A fluid pressure dynamic regulation method and system based on differential pressure feedback

Abstract

The application provides a kind of fluid pressure dynamic regulation method and system based on differential pressure feedback, it is related to fluid pressure control technical field, the method comprises: according to the connection order between region constructs directed association structure, and the differential pressure change direction of each adjacent region is as connection direction, along connection direction step by step tracking differential pressure change path, step by step backtracking each region in effective path, identify the response intensity of regional differential pressure, obtain differential pressure active value, mark the increase of the differential pressure of each region as positive direction value, mark the decrease of the differential pressure as negative direction value, identify the length of section in the same direction, extract the highest proportion of direction and proportion, identify the change degree of differential pressure dominant direction, obtain differential pressure offset factor, determine the adjustment priority and adjustment direction of each region, adjust the inflow intensity and outflow intensity of fluid, obtain partition adjustment data.The application can perceive local differential pressure change, and identify disturbance source.

Description

Technical Field

[0001] This invention relates to the field of fluid pressure control technology, and in particular to a method and system for dynamic regulation of fluid pressure based on differential pressure feedback. Background Technology

[0002] In cleanroom environments, maintaining a stable pressure difference between different zones is typically necessary to prevent the entry of external contaminants or the spread of internal pollution. Current technologies generally employ differential pressure sensors installed between the clean zone and adjacent zones to collect pressure difference signals in real time. These signals are compared with a set target pressure difference, and the operating status of air supply or exhaust actuators, such as dampers or fans, is adjusted based on the comparison result to regulate the airflow and thus control the pressure within the cleanroom. In multi-zone cleanroom systems, the pressure difference in critical zones, such as the core clean zone, is usually used as the control benchmark, and the overall air supply and exhaust system is regulated through a single feedback control loop.

[0003] However, a single differential pressure feedback control method may not accurately reflect pressure difference changes between different areas, easily leading to localized pressure imbalances. For example, in a semiconductor cleanroom, when personnel or materials enter the clean area through an airlock, the opening of the airlock door causes localized airflow disturbances, resulting in instantaneous changes in the pressure difference between adjacent areas. However, because existing systems only adjust based on the pressure difference between the core clean area and the outside, they may not be able to detect pressure fluctuations between the airlock and adjacent clean areas in a timely manner, leading to insufficient pressure difference in localized areas for short periods and increasing the risk of external contaminated air entering the clean area. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for dynamic regulation of fluid pressure based on differential pressure feedback, in order to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a method for dynamic regulation of fluid pressure based on differential pressure feedback, the method comprising:

[0007] Acquire differential pressure data between multiple regions;

[0008] Based on the differential pressure data, a directed correlation structure is constructed according to the connection order between regions, and the direction of differential pressure change in each adjacent region is used as the connection direction to obtain differential pressure transmission data.

[0009] Based on the differential pressure transmission data, the differential pressure change path is traced step by step along the connection direction, and the differential pressure change of each differential pressure change path is compared before and after to obtain the effective path set;

[0010] Based on the set of effective paths, starting from the termination region of each effective path, we trace back through each region in the effective path step by step, and distribute the pressure difference changes in each region in a decreasing manner to obtain the reverse influence weight data.

[0011] Based on the reverse impact weight data, the reverse impact weights of the same area in different effective paths are superimposed and accumulated to identify the response intensity of regional pressure difference and obtain the pressure difference activity value.

[0012] Based on the pressure difference transmission data, the increase of pressure difference in each region is marked as a positive direction value and the decrease of pressure difference is marked as a negative direction value. The direction values ​​of multiple consecutive cycles are spliced ​​together to obtain the regional direction sequence data.

[0013] Based on the regional directional sequence data, identify the length of segments with the same continuous direction, extract the direction and proportion with the highest proportion, identify the degree of change of the pressure difference dominant direction, and obtain the pressure difference offset factor.

[0014] By mapping the differential pressure activity value and differential pressure offset factor to a preset judgment interval, the adjustment priority and direction of each region are determined, and the inflow and outflow intensities of the fluid are adjusted to obtain the zoned adjustment data.

[0015] Furthermore, based on the differential pressure data, a directed correlation structure is constructed according to the connection order between regions, and the direction of differential pressure change in each adjacent region is used as the connection direction to obtain differential pressure transmission data, including:

[0016] Based on the differential pressure data, the spatial adjacency relationship between each region is identified, and adjacent regions are paired according to the connection order to obtain the region connection data;

[0017] Based on the regional connectivity data, the pressure difference data of each group of adjacent regions is extracted, and the pressure difference changes are arranged in chronological order to obtain the connectivity pressure difference sequence.

[0018] Based on the connected differential pressure sequence, the differential pressure data at the same time are compared to identify the changing trends of differential pressure from high to low and from low to high, and the changing trends are converted into directional markers to obtain the connected direction data;

[0019] Based on the connection direction data, the connection relationships of each region are combined with the direction labels, and then arranged in an orderly manner according to the region connection order to obtain directed association data;

[0020] Based on the directional correlation data, the pressure difference change amplitude of each connection relationship is written into the directional connection to determine the directional information and change amplitude of each connection relationship, thus obtaining the pressure difference transmission data.

[0021] Furthermore, based on the differential pressure transmission data, the differential pressure change path is traced step by step along the connection direction, and the differential pressure changes along each path are compared before and after to obtain the effective path set, including:

[0022] Based on the pressure difference transmission data, the connection directions of each region are traversed, and the regions with outward connections are taken as the starting nodes of the path to obtain the path starting point set;

[0023] Based on the path starting point set, extract subsequent connection nodes level by level along the connection direction of each starting node, and record the node arrangement relationship according to the connection order to obtain the initial path data;

[0024] Based on the initial path data, the pressure difference change magnitude and direction of each adjacent node in each initial path are read and combined according to the order in the initial path to obtain the path change sequence;

[0025] Based on the path change sequence, the pressure difference changes at adjacent locations in the initial path are compared sequentially to identify the consistency of the direction of pressure difference change and the continuity of the change amplitude, thus obtaining the path comparison results.

[0026] Based on the path comparison results, the initial paths that satisfy the conditions of continuous direction and orderly change are retained, while the initial paths with direction interruption or abnormal change are removed, thus obtaining the effective path set.

[0027] Furthermore, based on the effective path set, starting from the termination region of each effective path, the regions within the effective paths are traced back level by level, and the pressure difference changes in each region are progressively distributed to obtain the reverse influence weight data, including:

[0028] Based on the set of valid paths, extract the region nodes located at the end of each valid path and use these region nodes as the path termination nodes to obtain the set of termination regions.

[0029] Based on the set of termination regions, each valid path is rearranged in reverse order from the termination region to the starting region, and the order of nodes in the reverse arrangement is recorded to obtain the reverse path data.

[0030] Based on the reverse path data, read the pressure difference change amplitude corresponding to each node in each reverse path, and arrange them according to the node order of the reverse path to obtain the reverse change sequence.

[0031] Based on the reverse change sequence, the pressure difference change amplitude in the termination region is used as the initial allocation value, and then gradually reduced according to the position order of the nodes in the reverse path to obtain the path allocation sequence.

[0032] Based on the path allocation sequence, the nodes in each reverse path are associated with the allocation values, and the allocation values ​​of the same region in different paths are aggregated to obtain the reverse influence weight data.

[0033] Furthermore, based on the reverse impact weight data, the reverse impact weights of the same region in different effective paths are superimposed and accumulated to identify the response intensity of regional pressure difference and obtain the pressure difference activity value, including:

[0034] Based on the reverse impact weight and pressure difference change magnitude of the same region on a single effective path in the reverse impact weight data, the response intensity of the region to pressure difference disturbance in the effective path is identified, and the regional impact term is obtained.

[0035] The number of nodes in each effective path is extracted based on the reverse path data, and the path balance of effective paths of different lengths to the regional response is identified to obtain the path balance term.

[0036] Based on the reverse influence weight data, the degree of path linkage between the pressure difference transmission of each node in the effective path and the target area is identified, and the path coupling term is obtained.

[0037] By fusing regional impact terms, path equilibrium terms, and path coupling terms, the overall response intensity of the target area to differential pressure disturbances under the combined action of multiple effective paths is identified, and the differential pressure activity value is obtained.

[0038] Furthermore, based on the regional directional sequence data, the lengths of segments with consecutive identical directions are identified, the direction and proportion with the highest percentage are extracted, the degree of change in the dominant pressure differential direction is identified, and the pressure differential offset factor is obtained, including:

[0039] Based on the regional directional sequence data, the sum of the lengths of the positive and negative segments in the same region is extracted to identify the bias direction of the pressure difference change in the target region and obtain the directional bias term.

[0040] Based on the regional direction sequence data, the sum of segment lengths is compared with the sum of all segment lengths in the region to identify the degree of control of the dominant direction over the overall direction in the target region, and the dominant enhancement term is obtained.

[0041] Based on the regional directional sequence data, the degree of weakening of the stability of the dominant direction when the pressure difference change direction in the target region frequently switches is identified, and the switching inhibition term is obtained;

[0042] By fusing the directional offset term, the dominant enhancement term, and the switching suppression term, the degree of offset, the duration, and the stability of directional switching of the dominant direction of pressure difference change in the target area are identified, and the pressure difference offset factor is obtained.

[0043] Furthermore, by mapping the differential pressure activity value and differential pressure offset factor to a preset judgment interval, the adjustment priority and direction of each region are determined, and the inflow and outflow intensities of the fluid are adjusted to obtain zonal adjustment data, including:

[0044] By mapping the differential pressure activity value to a preset activity level range and the differential pressure offset factor to a preset direction determination range, area determination data is obtained;

[0045] Based on the regional assessment data, the activity level ranges of each region are compared to determine the adjustment order between different regions, and then sorted according to the adjustment order to obtain regional priority data.

[0046] Based on the regional determination data, the differential pressure adjustment direction of each region is determined by the direction determination interval of the differential pressure offset factor, and the adjustment direction is associated with the region to obtain the regional direction control data.

[0047] Based on the regional priority data and regional direction control data, the adjustment direction is converted into fluid inflow intensity and outflow intensity, and the adjustment amplitude is set in combination with the regional priority to obtain the zonal adjustment data.

[0048] Secondly, a fluid pressure dynamic regulation system based on differential pressure feedback, the system comprising:

[0049] The data module is used to acquire differential pressure data between multiple regions;

[0050] The transmission module is used to construct a directed correlation structure based on the differential pressure data and the connection order between regions, and to use the direction of differential pressure change in each adjacent region as the connection direction to obtain differential pressure transmission data.

[0051] The path module is used to trace the pressure difference change path step by step along the connection direction based on the pressure difference transmission data, and compare the pressure difference change of each pressure difference change path before and after to obtain the effective path set;

[0052] The weighting module is used to backtrack through each region in the effective path, starting from the termination region of each effective path, and to distribute the pressure difference changes in each region in a decreasing manner to obtain the reverse influence weight data.

[0053] The differential pressure activity module is used to accumulate the weights of the reverse influences of the same area in different effective paths based on the reverse influence weight data, identify the response intensity of the regional differential pressure, and obtain the differential pressure activity value.

[0054] The direction module is used to mark the increase of pressure difference in each region as a positive direction value and the decrease of pressure difference as a negative direction value based on the pressure difference transmission data, and to splice the direction values ​​of multiple consecutive cycles to obtain the regional direction sequence data;

[0055] The differential pressure offset module is used to identify the length of segments with the same continuous direction based on the regional direction sequence data, extract the direction and proportion with the highest proportion, identify the degree of change of the dominant differential pressure direction, and obtain the differential pressure offset factor.

[0056] The zone control module is used to determine the control priority and direction of each zone by mapping the differential pressure active value and differential pressure offset factor to a preset judgment interval, and to adjust the inflow and outflow intensity of the fluid to obtain zone control data.

[0057] The above-described solution of the present invention has at least the following beneficial effects:

[0058] This invention traces differential pressure transmission data step-by-step along the connection direction and compares each differential pressure change path before and after, expanding the data processing results from a single connection relationship to a path sequence data spanning multiple regions. It records the continuous transmission trajectory of differential pressure changes in space as a sequence of regional nodes, and combines the change direction and amplitude of each node in the path to form a composite data structure with both temporal and spatial order attributes. This allows the system to express the propagation chain of differential pressure disturbances at the data level, rather than just reflecting local changes. This path information can serve as the basis for judging the range and direction of disturbance propagation in the regulation logic, ensuring that regulation behavior corresponds to the differential pressure propagation path. It transforms the original differential pressure change data into a path sequence with continuous constraints, thus describing the dynamic transmission process of differential pressure.

[0059] This invention determines the consistency of change direction and the continuity of change amplitude among nodes in the path, and eliminates paths with directional interruptions or abnormal changes, ensuring that subsequent calculations conform to continuous propagation characteristics. This processing is based on the relationship between sequences to filter the data, giving the path data a consistent attribute. This data filtering method avoids including discontinuous fluctuations or local disturbances in the adjustment calculation scope, and performs structural filtering on the path data to form a data subset that meets the conditions of directional continuity and amplitude order, so that subsequent data calculations are based on input data with consistent propagation characteristics.

[0060] This invention uses a step-by-step backtracking approach, starting from the path's termination region, and progressively assigns pressure difference changes in each region along the path. This establishes a numerical correlation between each node in the path and the terminal region, assigning each regional node a value based on its path location. This results in the terminal change being a result of the joint contribution of all nodes along the path. The progressive assignment method makes the influence of nodes at different locations on the same path present a sequential distribution, generating a weight structure within the path. This data can be used to represent the degree of participation of each region in the overall pressure difference change. The path data is further transformed into node data with assigned weights, thus re-encoding the original change information into hierarchically distributed influence data.

[0061] This invention aggregates the weights of the reverse influences of the same region in different effective paths, enabling data to be summarized from the path dimension to the region dimension, achieving cross-path data integration. This unifies the dispersed influences of the same region in multiple paths into a single numerical indicator, reflecting the region's participation in the multi-path transmission structure. Furthermore, it converts path-level data into region-level data, which can be used as input variables for regional regulation calculations, uniformly describing the region's role in overall pressure difference changes. This completes the conversion from multi-path distributed data to single-region aggregated data, providing a data foundation for subsequent regional ranking and regulation calculations.

[0062] This invention marks the pressure difference changes in each region as positive or negative directions and splices together the direction values ​​of multiple consecutive cycles, transforming the original pressure difference changes from numerical changes into a sequence of directional changes. This sequence data records the changing trend of the region within a continuous time period and retains the temporal sequence information of the directional changes, thus achieving discretized encoding of the direction of pressure difference changes. This type of sequence data can be used to describe the changing characteristics of the controlled object in the time dimension, transforming the original continuous changing signal into a discrete directional sequence, giving the data a time structure that allows for segment analysis and trend extraction. Attached Figure Description

[0063] Figure 1 This is a flowchart of a fluid pressure dynamic regulation method based on differential pressure feedback provided by an embodiment of the present invention. Detailed Implementation

[0064] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0065] like Figure 1 As shown, an embodiment of the present invention proposes a method for dynamic regulation of fluid pressure based on differential pressure feedback, the method comprising:

[0066] Acquire differential pressure data between multiple regions;

[0067] Based on the differential pressure data, a directed correlation structure is constructed according to the connection order between regions, and the direction of differential pressure change in each adjacent region is used as the connection direction to obtain differential pressure transmission data.

[0068] Based on the differential pressure transmission data, the differential pressure change path is traced step by step along the connection direction, and the differential pressure change of each differential pressure change path is compared before and after to obtain the effective path set;

[0069] Based on the set of effective paths, starting from the termination region of each effective path, we trace back through each region in the effective path step by step, and distribute the pressure difference changes in each region in a decreasing manner to obtain the reverse influence weight data.

[0070] Based on the reverse impact weight data, the reverse impact weights of the same area in different effective paths are superimposed and accumulated to identify the response intensity of regional pressure difference and obtain the pressure difference activity value.

[0071] Based on the pressure difference transmission data, the increase of pressure difference in each region is marked as a positive direction value and the decrease of pressure difference is marked as a negative direction value. The direction values ​​of multiple consecutive cycles are spliced ​​together to obtain the regional direction sequence data.

[0072] Based on the regional directional sequence data, identify the length of segments with the same continuous direction, extract the direction and proportion with the highest proportion, identify the degree of change of the pressure difference dominant direction, and obtain the pressure difference offset factor.

[0073] By mapping the differential pressure activity value and differential pressure offset factor to a preset judgment interval, the adjustment priority and direction of each region are determined, and the inflow and outflow intensities of the fluid are adjusted to obtain the zoned adjustment data.

[0074] In this embodiment of the invention, differential pressure data between multiple regions are acquired, and the differential pressure measurement results are organized into standardized time series data to provide a data foundation for subsequent analysis. Based on the differential pressure data, a directed correlation structure is constructed according to the connection order between regions, and the direction of differential pressure change in each adjacent region is used as the connection direction to obtain differential pressure transmission data, describing the transmission process of differential pressure in space and providing a foundation for subsequent path tracing. Based on the differential pressure transmission data, the path of differential pressure change is traced step by step along the connection direction, and the differential pressure change of each path is compared before and after to obtain a set of effective paths. This achieves structured screening of differential pressure change data, eliminating discontinuous changes caused by random fluctuations or local disturbances, and providing data for subsequent analysis. Based on the set of effective paths, starting from the termination region of each effective path, each region in the effective path is traced back step by step, and the differential pressure change of each region is distributed in a decreasing manner to obtain reverse influence weight data. This establishes a numerical correlation between terminal changes and each region in the path, realizing the transformation of data from path structure to node influence distribution, and providing basic data for regional analysis.

[0075] Based on the reverse influence weight data, the reverse influence weights of the same region in different effective paths are superimposed and accumulated to identify the response intensity of regional pressure difference, obtaining the pressure difference activity value. This achieves the aggregation from multi-path data to regional single-value data, and describes the region's participation in the overall pressure difference change based on this index. Based on the pressure difference transmission data, the increase of pressure difference in each region is marked as a positive direction value, and the decrease of pressure difference is marked as a negative direction value. The direction values ​​of multiple consecutive cycles are spliced ​​to obtain regional direction sequence data, representing the pressure difference change process as a symbolic sequence, supporting subsequent segment division and trend extraction. Based on the regional direction sequence data, the length of segments with the same continuous direction is identified, the direction and proportion with the highest proportion are extracted, the degree of change of the pressure difference dominant direction is identified, and the pressure difference offset factor is obtained, reflecting both direction and persistence, providing data basis for determining the adjustment direction. By mapping the pressure difference activity value and the pressure difference offset factor to a preset judgment interval, the adjustment priority and adjustment direction of each region are determined, the inflow and outflow intensities of fluid are adjusted, and the zonal adjustment data is obtained to achieve multi-regional differentiated adjustment.

[0076] This includes acquiring pressure difference data between multiple regions, specifically:

[0077] Based on the spatial layout of the cleanroom or controlled environment, a zone division model is pre-established, clearly defining the adjacency relationships and connection boundaries between zones, such as clean areas, transition zones, airlocks, and connecting channels between external areas. Differential pressure sensors are deployed at the boundary positions of each pair of adjacent zones. The sensors are preferably installed on both sides of doors, near air vents, or along airflow exchange paths to ensure that the collected differential pressure signals reflect the true pressure differences between zones. The system synchronously samples each sensor according to a preset sampling period through a data acquisition module, converting the acquired analog signals into digital signals and adding a unified timestamp to ensure time consistency of data between different zones. The differential pressure data between adjacent zones are categorized and stored according to zone pairs. For example, the differential pressure data between zone A and zone B is labeled as AB channel data, forming a continuous time series. Preprocessing operations are performed on the raw differential pressure data, including identifying and removing abrupt outliers, interpolating and compensating for missing sampling points, and uniformly calibrating the outputs of different sensors to eliminate deviations caused by equipment errors. Finally, the acquired data is processed by sliding window filtering or low-pass filtering to reduce the impact of high-frequency noise on subsequent analysis. After the above processing, the pressure difference data is obtained, which can simultaneously reflect the spatial correspondence between regions and the change process of pressure difference in the time dimension, providing a basic input for subsequent processing.

[0078] Specifically, based on the pressure difference transmission data, an increase in pressure difference in each region is marked as a positive direction value, and a decrease in pressure difference is marked as a negative direction value. The direction values ​​from multiple consecutive periods are then concatenated to obtain the regional direction sequence data, which includes:

[0079] For each regional node, the system extracts the pressure difference changes of that region within each time period from all its associated connections. Specifically, for each sampling period, the system reads the pressure difference change amplitude corresponding to that region and calculates the difference between the current period's pressure difference value and the previous period's pressure difference value to obtain the pressure difference change amount. When the change amount is greater than zero, the region is determined to be in a state of increasing pressure difference within that period, and this state is marked as a positive value; when the change amount is less than zero, it is determined to be in a state of decreasing pressure difference, and is marked as a negative value; when the change amount is close to zero or within a preset dead zone, it can be marked as a neutral value or continue the direction of the previous period as needed to avoid noise interference.

[0080] After determining the direction in a single period, the direction markers for each period are arranged sequentially in chronological order within a fixed-length time window, such as N consecutive sampling periods, forming a direction value sequence for the region. Isolated direction points caused by sampling jitter or short-term reversals are smoothed, for example, by using majority rules or minimum segment length constraints to merge or correct direction segments shorter than a set threshold, ensuring the temporal continuity of the direction sequence. The direction sequences from multiple consecutive time windows are then spliced ​​together to form a regional direction sequence data covering a longer time range. To ensure the comparability of direction sequences between different regions, the sequence length can be standardized, for example, by truncating a fixed-length sequence or generating multiple equal-length subsequences using a sliding window method. Simultaneously, the direction sequence is bound to its corresponding timestamp for storage, enabling it to reflect the trend of direction changes and to be traced back to the original differential pressure data. The regional direction sequence data is obtained through the above processing.

[0081] In a preferred embodiment of the present invention, a directed correlation structure is constructed based on the differential pressure data according to the connection order between regions, and the direction of differential pressure change in each adjacent region is used as the connection direction to obtain differential pressure transmission data, including:

[0082] Based on the differential pressure data, the spatial adjacency relationship between each region is identified, and adjacent regions are paired according to the connection order to obtain the region connection data;

[0083] Based on the regional connectivity data, the pressure difference data of each group of adjacent regions is extracted, and the pressure difference changes are arranged in chronological order to obtain the connectivity pressure difference sequence.

[0084] Based on the connected differential pressure sequence, the differential pressure data at the same time are compared to identify the changing trends of differential pressure from high to low and from low to high, and the changing trends are converted into directional markers to obtain the connected direction data;

[0085] Based on the connection direction data, the connection relationships of each region are combined with the direction labels, and then arranged in an orderly manner according to the region connection order to obtain directed association data;

[0086] Based on the directional correlation data, the pressure difference change amplitude of each connection relationship is written into the directional connection to determine the directional information and change amplitude of each connection relationship, thus obtaining the pressure difference transmission data.

[0087] In this embodiment of the invention, based on differential pressure data, the spatial adjacency relationships between regions are identified, and adjacent regions are paired according to their connection order to obtain region connection data, establishing a topological connection relationship between regions and providing a data foundation for subsequent processing. Based on the region connection data, differential pressure data for each group of adjacent regions is extracted, and the pressure difference changes are arranged in chronological order to obtain a connected differential pressure sequence. This transforms the differential pressure data between regions into a time-sorted sequence, providing a data foundation for subsequent trend analysis. Based on the connected differential pressure sequence, differential pressure data at the same time are compared to identify trends in pressure difference from high to low and from low to high, and these trends are then converted into... As a directional marker, connection direction data is obtained, realizing the transformation from continuous numerical changes to discrete directional markers, reducing data complexity. Based on the connection direction data, the connection relationships of each region are combined with the directional markers and arranged in an orderly manner according to the region connection order to obtain directed correlation data. This transforms the undirected region connection relationships into directed relationships with clear directions, providing a basic data structure for path tracing. Based on the directed correlation data, the pressure difference change amplitude of each connection relationship is written into the directional connection to determine the direction information and change amplitude of each connection relationship, thus obtaining pressure difference transmission data. This expands the data from directional expression to a joint expression of direction and amplitude.

[0088] Specifically, based on differential pressure data, the spatial adjacency relationships between regions are identified, and adjacent regions are paired according to their connection order to obtain region connectivity data, which includes:

[0089] A regional topology model is established based on the structural layout information of the cleanroom or controlled environment. This model can be generated from pre-imported building floor plans, BIM models, or manually configured regional division tables. The system reads the topology model to determine if there are direct connections between regions, such as airflow exchange paths formed through doors, airlocks, vents, or passageways, and identifies regions with direct gas exchange pathways as adjacent regions. All adjacent regions are then numbered and arranged in an orderly manner according to preset rules, such as from inside to outside, from high cleanliness level to low cleanliness level, or according to region numbering order, forming a connection sequence. Each pair of adjacent regions is then paired to generate regional connection data; for example, region A is paired with region B to form an AB connection, and region B is paired with region C to form a BC connection. Simultaneously, each set of regional connections is associated with its corresponding differential pressure data, so that each connection not only includes a region identifier but also corresponds to a set of time-series differential pressure values. For multi-channel or multi-point regional connections, multiple sensor data are fused, such as by taking the average or weighted average, to form a single representative value.

[0090] Specifically, based on the connected pressure difference sequence, by comparing the pressure difference data at the same time, the changing trends of pressure difference from high to low and from low to high are identified, and these trends are converted into directional markers to obtain the connected direction data, which includes:

[0091] The system analyzes the pressure difference changes of each connected region point by point over a continuous time period. For each time sampling point, the pressure difference value at the current moment is read and the difference is calculated by subtracting the pressure difference value at the previous sampling moment to obtain the pressure difference change. If the change is greater than a preset positive threshold, the pressure difference at the connection is determined to show a trend of change from low to high; if the change is less than a negative threshold, it is determined to show a trend of change from high to low; if the change is within the threshold range, it can be regarded as a stable state or maintaining the direction of the previous moment. To avoid the influence of noise, a dead zone interval can be introduced so that small fluctuations do not participate in the direction determination. The change trend at each time point is converted into discrete direction markers, for example, +1 is used to represent the direction of pressure difference increase, -1 is used to represent the direction of pressure difference decrease, and 0 is used to represent no significant change. These direction markers are arranged in chronological order to form the direction sequence of the connection. Short-term isolated directions are smoothed, for example, by using a sliding window majority decision method to correct or merge direction segments with a length less than a set threshold to form connection direction data, describing the evolution of the pressure difference change direction of the connection over multiple periods.

[0092] Specifically, based on the directed correlation data, the pressure difference change amplitude of each connection relationship is written into the directional connection to determine the directional information and change amplitude of each connection relationship, thus obtaining the pressure difference transmission data, which specifically includes:

[0093] For each connection, the pressure difference amplitude is read from its corresponding time series, i.e., the difference between the current and previous pressure difference values. This amplitude value is then associated with the direction marker at the corresponding time point. The amplitude value is symbolized according to its direction attribute, for example, by maintaining the original positive or negative value or uniformly converting it into a combination of direction marker and absolute amplitude value, to ensure consistent representation of direction and amplitude information in the data structure. Each connection is expanded into a data structure containing multiple fields, including a start region identifier, an end region identifier, a sequence of direction markers, and a corresponding sequence of amplitude changes, and stored in chronological order. For multi-period data, a matrix or linked list structure can be used for organization, so that each time point corresponds to a complete set of direction and amplitude information. The amplitude data is normalized or scaled to eliminate the influence of dimensional differences between different connections, forming pressure difference transmission data that describes the direction of pressure difference transmission and its intensity of change between regions.

[0094] In a preferred embodiment of the present invention, based on differential pressure transmission data, the differential pressure change path is traced step by step along the connection direction, and the differential pressure changes of each differential pressure change path are compared before and after to obtain an effective path set, including:

[0095] Based on the pressure difference transmission data, the connection directions of each region are traversed, and the regions with outward connections are taken as the starting nodes of the path to obtain the path starting point set;

[0096] Based on the path starting point set, extract subsequent connection nodes level by level along the connection direction of each starting node, and record the node arrangement relationship according to the connection order to obtain the initial path data;

[0097] Based on the initial path data, the pressure difference change magnitude and direction of each adjacent node in each initial path are read and combined according to the order in the initial path to obtain the path change sequence;

[0098] Based on the path change sequence, the pressure difference changes at adjacent locations in the initial path are compared sequentially to identify the consistency of the direction of pressure difference change and the continuity of the change amplitude, thus obtaining the path comparison results.

[0099] Based on the path comparison results, the initial paths that satisfy the conditions of continuous direction and orderly change are retained, while the initial paths with direction interruption or abnormal change are removed, thus obtaining the effective path set.

[0100] In this embodiment of the invention, based on pressure difference transmission data, the connection directions of each region are traversed, and regions with outward connections are taken as the starting nodes of the path, resulting in a path starting point set. This enables the identification of key nodes in the directed association structure and avoids invalid nodes from participating in path generation. Based on the path starting point set, subsequent connection nodes are extracted level by level along the connection directions of each starting node, and the node arrangement relationship is recorded according to the connection order to obtain initial path data. This achieves an explicit expression of the pressure difference change propagation path and provides structured input for subsequent path analysis. Based on the initial path data, the pressure difference change amplitude and direction of each adjacent node in each initial path are read, and the initial path is then analyzed. The path data is sequentially combined to obtain a path change sequence, realizing an integrated expression of path data and change data, providing a data foundation for subsequent analysis. Based on the path change sequence, the pressure difference changes at adjacent positions in the initial path are compared sequentially to identify the consistency of the direction of pressure difference change and the continuity of the change amplitude, obtaining path comparison results that can distinguish between data with continuous transmission characteristics and randomly changing data. Based on the path comparison results, initial paths that meet the requirements of continuous direction and orderly change are retained, while initial paths with directional interruptions or abnormal changes are removed, resulting in an effective path set. This completes the screening and optimization of path data, providing data input for subsequent reverse weight calculation.

[0101] Specifically, based on the path change sequence, the pressure difference changes at adjacent locations in the initial path are compared sequentially to identify the consistency of the direction of pressure difference change and the continuity of the change amplitude, thus obtaining the path comparison results, which include:

[0102] For each initial path, the system extracts the corresponding path change sequence. For example, for an initial path consisting of regions A, B, C, and D in sequence, the system extracts the pressure difference change information of the three adjacent connecting positions: A to B, B to C, and C to D. These are then arranged according to the path's progression order to form an ordered change sequence. The system performs a segment-by-segment sequential comparison of the pressure difference changes at adjacent positions in the sequence. Starting from the beginning of the sequence, the system compares the direction marker of the current connecting position with the direction marker of the next connecting position to identify whether the pressure difference changes at two adjacent positions maintain the same direction of propagation. When the direction markers are the same, or although the direction markers are expressed differently but belong to the same propagation trend according to preset rules, the system determines that the adjacent positions meet the directional consistency condition. When the directions are opposite, or the previous position shows an increase in pressure difference while the next position shows a decrease in pressure difference, and the preset transition exemption condition is not met, the system determines that there is a directional interruption between the adjacent positions. To avoid misjudgments caused by short-term fluctuations within a single sampling period, the system can introduce a direction stability determination window. This means that instead of relying solely on the direction value at a single moment, the system compares the dominant direction of the current connection position over several consecutive sampling periods. For example, if a connection position has a majority of positive direction values ​​in multiple consecutive sampling points, it can be considered as a positive segment and then compared with the dominant direction of the next connection position. This ensures that the determination of direction consistency is based on continuous segments rather than isolated point values.

[0103] The system continues to analyze the continuity of pressure difference changes between adjacent locations, reading the change amplitude values ​​of the preceding and following connected locations, and calculating the difference, absolute value of the difference, or percentage change between them. If the amplitude difference between adjacent locations is within a preset allowable range, or their amplitude change relationship conforms to a preset continuous change rule, such as gradual attenuation, gradual increase, or change within a defined amplitude fluctuation range, the system determines that the adjacent locations meet the amplitude continuity condition. Conversely, if the amplitude difference between adjacent locations exceeds a set threshold, exhibiting a sudden increase, sudden decrease, or irregular jump, the system determines that there is an anomaly between the adjacent locations. By calculating the amplitude change rate between two adjacent connected locations, it determines whether it exceeds a preset upper limit; or by calculating the gradient sequence of local segments in the entire path, it determines whether there are any anomalies at a certain location that significantly deviate from the surrounding change trend. The system summarizes the comparison results of each segment in path order to generate path comparison results. The path comparison results can be stored in the form of structured data, which records the direction consistency judgment result, amplitude continuity judgment result, abnormal position number, and comprehensive judgment status of the entire path at each adjacent position of each initial path.

[0104] Based on the path comparison results, initial paths that satisfy the conditions of continuous direction and orderly change are retained, while initial paths with direction interruptions or abnormal changes are removed, resulting in a set of valid paths, specifically including:

[0105] The system performs validity checks on each initial path, reading the comparison results for each path and judging them according to pre-set filtering rules. These filtering rules include at least two aspects: firstly, directional continuity requirements, meaning the directional comparison results between adjacent connecting positions in the path should meet a preset consistency condition; secondly, orderly change requirements, meaning the amplitude comparison results between adjacent connecting positions in the path should meet a preset continuous change condition. Only when an initial path simultaneously meets both of these preset rules is it recognized as a candidate valid path conforming to the pressure difference transmission law. If a path has a generally consistent direction but exhibits significant amplitude jumps at one or more connecting positions, it is not directly included in the valid path set. Similarly, if a path's amplitude changes smoothly but experiences directional reversal or interruption during propagation, it is also not considered a valid path. The system performs more detailed classification of initial paths based on path length, anomaly location distribution, and anomaly type. Paths that fully meet the conditions of continuous direction and orderly change are directly retained and marked as valid paths. For paths with only a few minor anomalies in localized areas, but which maintain a consistent dominant direction and continuous trend, the system can decide whether to correct the path before retention based on preset fault tolerance rules. For example, if a path contains only one isolated anomaly segment of length one, and the direction and amplitude trends before and after this segment are consistent, the system can treat this isolated anomaly as a local deviation caused by sampling noise or transient disturbances, and correct the path by merging adjacent segments, smoothly replacing anomaly points, or removing short segments. After correction, the system re-evaluates whether the path meets the valid path conditions. Conversely, paths with multiple directional interruptions, consecutive abrupt amplitude changes, excessively long anomaly segments, or anomaly points located at critical connection points are directly rejected by the system and not further processed.

[0106] After determining the validity of a single path, the system performs a set-level check on all retained paths. The system statistically analyzes the distribution of currently retained paths across dimensions such as starting region, ending region, path length, and dominant direction. If some paths, while individually meeting local continuity requirements, exhibit significantly different overall characteristics from other paths in the same batch—for example, showing completely opposite transmission directions within the same starting region, or having a path length far exceeding the system's set reasonable propagation range—a secondary review mechanism can be triggered. During the secondary review, the system re-accesses the original differential pressure transmission data to perform a backtracking check on the path, confirming whether it is a pseudo-path formed under abnormal operating conditions. If irrationality is confirmed, it is removed from the retention set. The system then aggregates all ultimately retained paths to form a valid path set. Each path in the valid path set retains information such as the original path node order, corresponding direction sequence, amplitude sequence, and screening judgment status, which serves as the basic input for subsequent reverse backtracking, influence weight allocation, and regional activity calculation.

[0107] In a preferred embodiment of the present invention, based on the effective path set, starting from the termination region of each effective path, each region in the effective path is traced back level by level, and the pressure difference change in each region is progressively allocated to obtain the reverse influence weight data, including:

[0108] Based on the set of valid paths, extract the region nodes located at the end of each valid path and use these region nodes as the path termination nodes to obtain the set of termination regions.

[0109] Based on the set of termination regions, each valid path is rearranged in reverse order from the termination region to the starting region, and the order of nodes in the reverse arrangement is recorded to obtain the reverse path data.

[0110] Based on the reverse path data, read the pressure difference change amplitude corresponding to each node in each reverse path, and arrange them according to the node order of the reverse path to obtain the reverse change sequence.

[0111] Based on the reverse change sequence, the pressure difference change amplitude in the termination region is used as the initial allocation value, and then gradually reduced according to the position order of the nodes in the reverse path to obtain the path allocation sequence.

[0112] Based on the path allocation sequence, the nodes in each reverse path are associated with the allocation values, and the allocation values ​​of the same region in different paths are aggregated to obtain the reverse influence weight data.

[0113] In this embodiment of the invention, based on the effective path set, the terminal nodes of each effective path are extracted and used as path termination nodes to obtain a termination region set. This clarifies which regions are at the end of the pressure differential transmission chain, providing a starting point for subsequent data. Based on the termination region set, each effective path is rearranged in reverse order from the termination region to the starting region, and the order of the reversed nodes is recorded to obtain reverse path data. The path data is then converted into a sequence starting from the termination region, enabling a process of tracing back to the cause from the result. Based on the reverse path data, the pressure differential change amplitude corresponding to each node in each reverse path is read and processed according to the node order of the reverse path. The data is arranged in rows to obtain a reverse change sequence, which enables the reintegration of path structure and numerical data, providing a data foundation for subsequent calculations. Based on the reverse change sequence, the pressure difference change amplitude in the termination region is used as the initial allocation value, and then gradually reduced according to the position of the nodes in the reverse path to obtain a path allocation sequence. This converts the change amplitude in the path into node-level allocation values, realizing the transformation of data from change records to influence allocation. Based on the path allocation sequence, the nodes in each reverse path are associated with the allocation values, and the allocation values ​​of the same region in different paths are aggregated to obtain reverse influence weight data. This realizes the data fusion from single-path node allocation to multi-path region aggregation, providing input for subsequent calculations.

[0114] Specifically, based on the reverse change sequence, the pressure difference change amplitude in the termination region is used as the initial allocation value, and then gradually decreased according to the position order of the nodes in the reverse path to obtain the path allocation sequence, which specifically includes:

[0115] The system uses a single reverse path as the processing unit, analyzing the node sequence and variation amplitude segment by segment. A reverse path is a sequence of nodes rearranged from the terminating region of the valid path back to the starting region. The corresponding reverse variation sequence is a sequence of amplitudes formed by organizing the pressure difference variation amplitudes between adjacent nodes in the reverse path according to the reverse path's arrangement. For example, for an original valid path ABCD, its corresponding reverse path is DCBA. In this reverse path, the system first determines the terminating region D as the allocation starting point and reads the pressure difference variation amplitude value associated with that terminating region from the reverse variation unit directly corresponding to D. This amplitude value is then used as the initial allocation value for this reverse path. The system progressively decreases the initial allocation value according to the node's position in the reverse path. The location of the termination region is defined as the first-level position, and its allocation value is directly equal to the initial allocation value. Then, for the second node in the reverse path, i.e., the upstream node directly adjacent to the termination region, the initial allocation value is decreased once according to a preset attenuation rule to obtain the first decreased allocation value for that node. The same decrease operation is performed on the third node in the path, based on the previous allocation value, to obtain the allocation value for the next node, and so on, until all nodes in the reverse path have received their corresponding allocation results. The attenuation rule can be preset according to specific implementation needs. For example, a fixed-proportion attenuation method can be used, making the current node's allocation value equal to the previous node's allocation value multiplied by a set attenuation coefficient; a fixed-difference attenuation method can be used, making the allocation value gradually decrease by a fixed value for each forward path position; or the attenuation magnitude can be dynamically determined based on the total path length, making the decrease in allocation value more gradual in longer paths and more compact in shorter paths. The system can modify the decreasing allocation process by incorporating local change information in the reverse change sequence. The system considers not only the relative position of nodes in the path but also the difference in pressure difference, the percentage change, or the smoothness of the change between the current node and the previous node to adjust the base decreasing value. For example, when the pressure difference between adjacent positions decreases more gradually compared to the preceding position, the system can retain a larger proportion of the allocation based on the preset decreasing value; conversely, when the pressure difference at a certain position shows a significant decrease, the system increases the decreasing degree of nodes after that position accordingly. The system records each node and its corresponding allocation value in reverse path order, forming the path allocation sequence corresponding to that reverse path.

[0116] Specifically, based on the path allocation sequence, nodes in each reverse path are associated with their allocated values, and the allocated values ​​for the same region in different paths are aggregated to obtain reverse influence weight data, which includes:

[0117] The system performs association processing on nodes and allocation values ​​within a single path, reading the node identifiers in the path allocation sequence and binding them one by one with the allocation values ​​at the corresponding positions in the same sequence, forming node-level allocation records. The system performs a summary traversal of all reverse paths, classifying and aggregating node allocation records according to region identifiers. It imports all node-level allocation records from all paths into a unified data cache or region index table, and then uses the region identifier as the search key to search and aggregate records of the same region from different paths. For example, a region might be in the second position in the first reverse path, the fourth position in the second reverse path, and appear as a termination region in the third reverse path. The system will extract all the multiple allocation values ​​corresponding to this region under the above different paths and group them into the data set corresponding to that region. Thus, each region corresponds to a set of allocation values ​​from different reverse paths, reflecting the distribution of the influence received by that region in multiple pressure differential transmission chains. The system performs aggregate calculations on these allocation values. It uses an accumulation method to sum the allocation values ​​of the same region across different reverse paths to obtain the total influence weight of that region. Alternatively, a weighted aggregation method can be used, introducing path length coefficients, terminal area change magnitude coefficients, or path effectiveness level coefficients when aggregating the allocation values, to differentiate the contribution of different source paths to the region's weight. For example, paths with shorter lengths and larger changes in terminal area pressure differentials can be assigned higher weights during aggregation; while paths with longer lengths and weaker changes can have their aggregation weight appropriately reduced.

[0118] After data aggregation, the system performs post-processing on the weight values ​​of each region. For example, it performs normalization based on the aggregated results of all regions, mapping the weights of each region to a unified range for direct comparison between different regions. Alternatively, it can categorize the weight values ​​according to preset level intervals, forming standardized input data that can be used for subsequent calculation of differential pressure activity values. Simultaneously, the system retains information on the composition of the weights; that is, in addition to the total regional weight, it synchronously stores supplementary information such as which paths contribute to the total weight, the percentage contribution of each path, and the region's position within each path. The system maps the path allocation results in each reverse path to specific regional nodes, and further aggregates and calculates the allocation values ​​of the same region across different paths to obtain the reverse influence weight data.

[0119] In a preferred embodiment of the present invention, based on the reverse influence weight data, the reverse influence weights of the same region in different effective paths are superimposed and accumulated to identify the response intensity of the regional pressure difference and obtain the pressure difference activity value, including:

[0120] Based on the reverse impact weight and pressure difference change magnitude of the same region on a single effective path in the reverse impact weight data, the response intensity of the region to pressure difference disturbance in the effective path is identified, and the regional impact term is obtained.

[0121] The number of nodes in each effective path is extracted based on the reverse path data, and the path balance of effective paths of different lengths to the regional response is identified to obtain the path balance term.

[0122] Based on the reverse influence weight data, the degree of path linkage caused by the pressure difference transmission of each node in the effective path to the target area is identified, and the path coupling term is obtained.

[0123] By fusing regional impact terms, path equilibrium terms, and path coupling terms, the overall response intensity of the target area to differential pressure disturbances under the combined action of multiple effective paths is identified, and the differential pressure activity value is obtained.

[0124] In this embodiment of the invention, based on the reverse influence weight and pressure difference change magnitude of the same region on a single effective path in the reverse influence weight data, the response intensity of the region to pressure difference disturbance in the effective path is identified, and a regional influence term is obtained. This realizes the transformation from path allocation weight to regional influence contribution and establishes a quantitative relationship between the region and pressure difference disturbance. Based on the reverse path data, the number of nodes in each effective path is extracted, and the path balance degree of effective paths of different lengths to the regional response is identified, resulting in a path balance term, making the data between different paths comparable. Based on the reverse influence weight data, the path linkage degree of pressure difference transmission of each node in the effective path to the target region is identified, resulting in a path coupling term, so that the regional response depends not only on a single path but also realizes the quantitative expression of the correlation between paths. The regional influence term, path balance term, and path coupling term are fused to identify the overall response intensity of the target region to pressure difference disturbance under the combined action of multiple effective paths, resulting in a pressure difference activity value. This completes the transformation from multi-dimensional path data to a single regional indicator, providing input for subsequent regulation.

[0125] Specifically, based on the reverse influence weights and pressure difference changes of the same region along a single effective path in the reverse influence weight data, the response intensity of the region to pressure difference disturbances in the effective path is identified, yielding a regional influence term; based on the reverse path data, the number of nodes in each effective path is extracted, and the path balance degree of effective paths of different lengths to the regional response is identified, yielding a path equilibrium term; based on the reverse influence weight data, the path linkage degree of pressure difference transmission at each node in the effective path to the target region is identified, yielding a path coupling term; the regional influence term, path equilibrium term, and path coupling term are fused to identify the overall response intensity of the target region to pressure difference disturbances under the combined action of multiple effective paths, yielding a pressure difference activity value, specifically including:

[0126] The system uses the target region as the computational object, traversing each valid path containing the target region and establishing a correspondence between the target region and each valid path. For the first... Given a target region, the system selects all valid paths passing through the set of valid paths, and denots the number of such valid paths as . ,in Indicates the target area The total number of valid paths with pressure differential transmission correlation. Then, the system is organized according to path number. These valid paths are read sequentially, and for each path... The first effective path is to extract the target region from the reverse influence weight data. The reverse influence weight in this path is denoted as The reverse influence weight is used to characterize the target region in the first... The share of influence that the target area occupies when receiving the pressure difference change result of the termination area within the effective path is derived from the reverse backtracking allocation result in the preceding steps. Therefore, this quantity itself already includes the path position relationship and decreasing allocation relationship of the target area relative to the termination area within the path. Simultaneously, the system reads the... The differential pressure change corresponding to each effective path is denoted as . In specific implementation, the aforementioned The target area can be selected in the first place. The local pressure difference change at the corresponding connection point in an effective path can also be taken as the representative pressure difference change at the termination region of the effective path, and then mapped in combination with the relative position of the target area in the path. By... and By performing correlation, the target region is obtained in the [number]th [location]. The basic value of the single-path response intensity to pressure difference disturbance on the effective path is the regional influence term.

[0127] The system continues to extract the data based on the reverse path. The number of nodes in a valid path is denoted as . The aforementioned Indicates the first The total number of regional nodes contained in a valid path essentially reflects the path length or the depth of the propagation hierarchy. The system will... As a path balancing factor, it is introduced into the normalization process of single-path response intensity, that is, the aforementioned and The product result divided by ,form This process ensures that pressure changes of the same magnitude do not directly affect the target area in the same way across paths of different lengths. Instead, they are distributed proportionally according to the number of nodes along the path. When a given effective path is long, it indicates that the pressure disturbance has been propagated through more regional nodes before reaching the target area. Therefore, the contribution of a single path to the target area is calculated according to... A reduction is performed; when an effective path is short, it indicates a shorter transmission chain between the pressure disturbance and the target area, resulting in a smaller path length allocation to the corresponding basic computational terms. In this way, the system obtains path equilibrium terms corresponding to the path length, ensuring that the contributions of paths of different lengths to the target area response are within the same comparable computational framework.

[0128] The system for the first Traverse all nodes in the valid path, read the reverse influence weight corresponding to each node on the path, and denot it as . ,in Indicates the first The node position number in the valid path. The value range is 1 to ,Right now Indicates the first The sum of the reverse influence weights of all nodes on the valid path. The system divides this sum by the total number of nodes on the path. The average reverse influence level of nodes within this path is obtained, i.e. This average value is used to describe the first time... In an effective path, the participation and linkage of all path nodes, excluding the target area itself, in the pressure difference transmission process are considered. A larger average value indicates that multiple nodes in the effective path have significant reverse influence weights, meaning the path is not driven solely by a single local node but rather exhibits a linked transmission state involving multiple regional nodes along the path. A smaller average value indicates that, except for a few nodes, the participation of other nodes in the pressure difference transmission is weak, and the path linkage effect is relatively limited. The system adds 1 to this average value to form a path coupling correction factor. Adding 1 ensures that even if the average linkage level of nodes within the path is low, the basic contribution value of a single path retains its original magnitude and is not weakened to near zero due to an excessively small coupling term. Simultaneously, as the overall linkage level within the path increases, this correction factor increases accordingly, allowing effective paths with more pronounced path linkage characteristics to occupy a larger proportion in the final pressure difference activity value calculation. The system quantifies the weight distribution characteristics of all nodes within the path into a path coupling term, directly reflecting the amplification effect on the target area's response intensity when multiple nodes in the path jointly transmit pressure difference disturbances.

[0129] The system performs the first step according to the preset formula. Pressure differential activity value of each target area The system performs fusion calculations on all data passing through the target area. The contribution value of each valid path is calculated, and the contribution values ​​of all paths are summed. The formula is as follows: ,in Indicates the first Active pressure differential values ​​for each target area; Indicates after the first The total number of valid paths to each target area; Indicates the first The target area in the first The weight of the reverse impact in the effective path; Indicates the first The target area in the first The magnitude of pressure difference change corresponding to each effective path; Indicates the first The number of nodes in a valid path; Indicates the first The first of the valid paths The reverse influence weight of each node.

[0130] In a preferred embodiment of the present invention, based on regional direction sequence data, the lengths of segments with consecutive identical directions are identified, the direction and proportion with the highest percentage are extracted, the degree of change in the dominant pressure differential direction is identified, and a pressure differential offset factor is obtained, including:

[0131] Based on the regional directional sequence data, the sum of the lengths of the positive and negative segments in the same region is extracted to identify the bias direction of the pressure difference change in the target region and obtain the directional bias term.

[0132] Based on the regional direction sequence data, the sum of segment lengths is compared with the sum of all segment lengths in the region to identify the degree of control of the dominant direction over the overall direction in the target region, and the dominant enhancement term is obtained.

[0133] Based on the regional directional sequence data, the degree of weakening of the stability of the dominant direction when the pressure difference change direction in the target region frequently switches is identified, and the switching inhibition term is obtained;

[0134] By fusing the directional offset term, the dominant enhancement term, and the switching suppression term, the degree of offset, the duration, and the stability of directional switching of the dominant direction of pressure difference change in the target area are identified, and the pressure difference offset factor is obtained.

[0135] In this embodiment of the invention, based on regional direction sequence data, the sum of the lengths of segments in the positive direction and the sum of the lengths of segments in the negative direction within the same region are extracted to identify the bias direction of pressure difference changes in the target region, thus obtaining a direction bias term. This achieves structured analysis of the direction sequence data and forms a quantitative expression of the bias of pressure difference changes. Based on the regional direction sequence data, the sum of the segment lengths is compared with the sum of the lengths of all segments in the region to identify the degree of control of the dominant direction over the overall direction in the target region, thus obtaining a dominant enhancement term. This allows the persistence of the dominant direction to be expressed in numerical form, forming a quantitative description of the stability and persistence of the direction. Based on the regional direction sequence data, the degree of weakening of the stability of the dominant direction when the pressure difference change direction in the target region frequently switches is identified, thus obtaining a switching suppression term. This allows the disturbance instability corresponding to frequent direction switching to be quantified and participate in subsequent calculations. The direction bias term, dominant enhancement term, and switching suppression term are fused to identify the degree of offset, persistence, and stability of the direction switching of the dominant direction of pressure difference changes in the target region, thus obtaining a pressure difference offset factor. This achieves the conversion from multi-dimensional direction feature data to a single comprehensive parameter, providing data input for subsequent adjustment direction determination.

[0136] Specifically, based on regional direction sequence data, the sum of the lengths of segments in the positive and negative directions within the same region is extracted to identify the bias direction of pressure difference changes in the target region, resulting in a direction bias term. Based on the regional direction sequence data, the sum of segment lengths is compared with the sum of all segment lengths in the region to identify the degree of control of the dominant direction over the overall direction in the target region, resulting in a dominant enhancement term. Based on the regional direction sequence data, the degree to which frequent switching of pressure difference change direction weakens the stability of the dominant direction in the target region is identified, resulting in a switching suppression term. The direction bias term, dominant enhancement term, and switching suppression term are fused to identify the degree of shift, persistence, and stability of the direction switching of the dominant direction of pressure difference changes in the target region, resulting in a pressure difference offset factor, specifically including:

[0137] The system processes a single region, reading the directional value sequence formed within that region over multiple consecutive sampling periods, cycle by cycle. Each element in the directional value sequence corresponds to the directional state of the pressure difference change in that region within a sampling period. A positive directional value indicates that the region exhibits an increasing pressure difference trend within the corresponding period, while a negative directional value indicates that the region exhibits a decreasing pressure difference trend within the corresponding period. The system scans the directional value sequence of the region in chronological order and merges consecutive directional values ​​with the same direction into a complete segment. For each directional segment, the system records the directional attribute of the segment and the number of consecutive periods, defining the number of consecutive periods as the segment length. Assuming the target region... The regional direction sequence is obtained after segmentation. Then the th consecutive segment, The length of each segment is denoted as ,in The value range is 1 to Simultaneously, the lengths of all positive direction segments are summed to obtain the total length of positive direction segments of the target region within the statistical period, denoted as . The lengths of all negative direction segments are summed to obtain the total length of negative direction segments in the target area within the statistical period, denoted as . Since the total effective period length covered by the regional direction sequence is equal to the sum of the lengths of each segment, the system records the sum of the lengths of all segments of the target region within the current statistical window as... ,in can be and The result can be obtained by adding them together, or by summing the lengths of each segment. .

[0138] The system calculates the directional bias term for the target area, indicating whether the pressure difference change in that area tends to increase or decrease over the entire statistical window. The system sums the lengths of the positive direction segments. Sum of the lengths of the negative direction segments Perform the difference operation and divide it by the sum of the lengths of all segments. ,form That is, the direction bias expression. This expression preserves the positive and negative attributes of the direction: when Greater than When the result is positive, it indicates that the target area generally exhibits a dominant trend of increasing pressure differential within the current analysis window; when... Less than When the result is negative, it indicates that the target area as a whole exhibits a dominant trend of decreasing pressure differential; when and When they are essentially equal, the result approaches zero, indicating that the region's persistence in the positive and negative directions is similar within the analysis window, and the overall bias is not significant. This expression is obtained through... Normalization is performed, so even if the statistical window lengths are different in different regions, the directional bias term is within a unified and comparable dimensional system.

[0139] The system constructs dominant enhancement terms based on regional direction sequence data to characterize the degree of control of the dominant direction over the overall direction change process in the target region. The system accumulates the length of the positive direction. Cumulative length in the negative direction The values ​​are compared, and the larger one is selected as the cumulative length of the dominant direction of the target region within the current statistical window. This maximum value reflects the extent to which the direction with the longer duration occupies the overall time window. The system divides this maximum cumulative length by the total length of all segments. This yields the proportion of the cumulative length of the dominant direction within the overall time window, i.e. The system adds 1 to this ratio to form the dominant enhancement term, namely... Even if the proportion of the dominant direction is not high, this term remains at least above 1, so that it will not have an eliminating effect on the overall offset result; while when the dominant direction lasts longer and accounts for a larger proportion throughout the time window, this enhancement term increases accordingly, indicating that the target area has a clearer dominant direction control feature during the statistical period.

[0140] The system calculates the length of all segments. Perform a segment-by-segment traversal and calculate the sum of squares of the lengths of each segment, i.e. Because the sum is accumulated in a squared form, the contribution of longer segments to the calculation results is further amplified, while the contribution of shorter segments is relatively smaller. Therefore, this term can highlight the temporal distribution characteristics of the concentrated occurrence of longer segments. The system divides this sum of squares by... To eliminate the influence of total time length on comparisons between different regions, forming This refers to the normalization term. Simultaneously, the system continues to identify the longest single segment among all segments and records it as... The maximum segment length represents the longest-lasting unidirectional change process within the entire statistical window, and is an important parameter characterizing the sustained concentration of the dominant direction in the target region. The system divides this maximum segment length by the total length. ,form And add it to the normalized term of the sum of squares of the aforementioned segments to obtain This combined term reflects, on the one hand, the overall concentration of long segments in the distribution of all segment lengths, and on the other hand, highlights the dominant role of the longest continuous segment in the overall direction of change. If the target region maintains the same direction for a relatively long period within the statistical window, both the sum of squares of the segments and the maximum segment length will increase accordingly, thus increasing the combined term; conversely, if the direction sequence is cut by a large number of short segments, the combined term will relatively decrease.

[0141] The system iterates through adjacent segments in pairs according to their segment numbers, calculates the product of the lengths of adjacent segments, and sums the products of all adjacent segments. The meaning of this expression is as follows: when adjacent segments all have a certain length, their product will be large, indicating that the target region maintains a certain direction for a relatively long period and then switches to the opposite direction for another relatively long period. This phenomenon shows that not only does the direction switch occur, but the new direction after the switch is also maintained for a relatively long time, thus significantly weakening the stability of the dominant direction. Conversely, when at least one of the adjacent segments is relatively short, their product is small, indicating that the direction switch may only be caused by a short-term disturbance, and its impact on the overall stability of the dominant direction is weak. The system then divides the sum of the products of the lengths of the adjacent segments by... This forms the normalized switching intensity term. To ensure that this switching intensity has a suppressive effect in the overall formula, the system adds 1 to it and takes the reciprocal to obtain the switching suppression term, i.e. When there are frequent alternations between long segments in the regional direction sequence, the switching intensity term in the denominator increases, which reduces the overall switching suppression term and weakens the final differential pressure offset factor. However, when there are few direction switches or only short-term scattered disturbances, the suppression term is close to 1 and has little impact on the overall result.

[0142] The system targets the area. Pressure offset factor Perform fusion calculations. The formula for calculating the differential pressure offset factor is: ,in Indicates the first Pressure differential offset factor for each target area; This represents the total length of all positive-direction segments within the statistical window for the target region; This represents the total length of all negative-direction segments within the statistical window for the target region; This represents the total length of all segments within the statistical window representing the target region; This represents the total number of segments obtained after dividing the target area into segments; Indicates the first The length of each directional segment; Indicates the first The length of each directional segment; This represents the maximum segment length among all directional segments. According to this formula, the first term... Used to characterize the direction and degree of deviation of pressure difference changes in the target area; the second item Used to characterize the degree of control that the dominant direction has over the overall directional change process; the third item Used to characterize the concentration of long segments in the directional segment distribution of the target region and the sustained effect of the longest continuous segment; the fourth item It is used to characterize the degree to which frequent changes in direction weaken the stability of the dominant direction. It retains both the positive and negative attributes of the directional bias, while also comprehensively considering information such as the duration of the dominant direction, the concentration of segment distribution, and the stability of direction switching. When a region maintains a certain direction for a relatively long period of time with few intermediate switching, its... The absolute value will be relatively large; when a certain region has a certain directional bias, but the segments are scattered and switching is frequent, the inhibitory effect of the fourth term and the weakening effect of the third term will work together to make The absolute value decreases; when the lengths in the positive and negative directions are close, the first term approaches zero, then the overall... A value approaching zero indicates that there is no clear dominant shift trend in this region within the current time window.

[0143] In a preferred embodiment of the present invention, by mapping the differential pressure activity value and differential pressure offset factor to a preset judgment interval, the adjustment priority and adjustment direction of each region are determined, and the inflow and outflow intensities of the fluid are adjusted to obtain zoned adjustment data, including:

[0144] By mapping the differential pressure activity value to a preset activity level range and the differential pressure offset factor to a preset direction determination range, area determination data is obtained;

[0145] Based on the regional assessment data, the activity level ranges of each region are compared to determine the adjustment order between different regions, and then sorted according to the adjustment order to obtain regional priority data.

[0146] Based on the regional determination data, the differential pressure adjustment direction of each region is determined by the direction determination interval of the differential pressure offset factor, and the adjustment direction is associated with the region to obtain the regional direction control data.

[0147] Based on the regional priority data and regional direction control data, the adjustment direction is converted into fluid inflow intensity and outflow intensity, and the adjustment amplitude is set in combination with the regional priority to obtain the zonal adjustment data.

[0148] In this embodiment of the invention, by mapping the differential pressure activity value to a preset activity level range and the differential pressure offset factor to a preset direction determination range, regional determination data is obtained. This achieves the data conversion from continuous characteristic quantities to discrete determination labels, providing a regional state expression for subsequent adjustment priority determination and direction control. Based on the regional determination data, the activity level ranges of each region are compared to determine the adjustment order between different regions. These regions are then sorted according to the adjustment order to obtain regional priority data. This achieves the conversion of regional state data to adjustment order data, establishing a sequential basis for the allocation of limited supply and exhaust ventilation adjustment resources among different regions. According to the method, the differential pressure adjustment direction of each region is determined by the direction determination interval of the differential pressure offset factor, and the adjustment direction is associated with the region to obtain the regional direction control data. This realizes the conversion of regional offset trend data into control direction labels, providing a directional basis for subsequent fluid inflow or outflow adjustment. Based on the regional priority data and regional direction control data, the adjustment direction is converted into fluid inflow intensity and outflow intensity, and the adjustment amplitude is set in combination with the regional priority to obtain the zonal adjustment data. This realizes the final mapping from the regional determination result to the execution control parameters, so that the adjustment direction and adjustment sequence are specifically converted into fluid inflow intensity and outflow intensity adjustment data that can be issued.

[0149] Specifically, by mapping differential pressure activity values ​​to preset activity level ranges and differential pressure offset factors to preset direction determination ranges, region determination data is obtained, including:

[0150] The system establishes a unified judgment mapping model for all areas. Based on cleanroom zoning control requirements, allowable differential pressure fluctuation range, the adjustment capabilities of supply and exhaust air actuators, and historical operating conditions, the system pre-sets multiple activity level intervals corresponding to differential pressure activity values ​​and multiple directional judgment intervals corresponding to differential pressure deviation factors. For differential pressure activity values, the system can divide them into multiple levels, such as low activity, medium activity, high activity, and extremely high activity, each level corresponding to different levels of disturbance participation or adjustment focus. For differential pressure deviation factors, the system classifies them according to their positive / negative attributes and absolute value into categories such as significant positive deviation intervals, weak positive deviation intervals, directional balance intervals, weak negative deviation intervals, and significant negative deviation intervals, used to characterize the dominant direction and significance of differential pressure changes over time. The system reads the differential pressure activity value of each area one by one and compares this value sequentially with the preset activity level interval boundaries to determine the interval range it falls within, assigning the corresponding activity level label to the area. Simultaneously, the differential pressure offset factor of the same region is read and matched with the preset direction determination interval to identify whether the region is currently in a positive offset, negative offset, or equilibrium state, as well as the strength of the offset.

[0151] The system sets boundary buffer zones or hysteresis zones for differential pressure activity values ​​and differential pressure offset factors, respectively. When a parameter value in a certain region is near the boundary of two adjacent judgment intervals, the system can combine the judgment result of the previous control cycle to make a maintenance judgment. That is, the level or direction category is only switched after the parameter has entered the new interval for multiple consecutive sampling windows; if it only briefly crosses the boundary within a single cycle, the judgment result of the previous cycle is maintained. The activity level label and direction judgment label of the same region are jointly encapsulated and stored with the region identifier and the current cycle number to form region judgment data.

[0152] Specifically, based on regional assessment data, the activity level ranges of each region are compared to determine the adjustment order among different regions. These regions are then sorted according to the adjustment order to obtain regional priority data, which includes:

[0153] The system groups all regions according to their activity levels. For example, regions in the extremely high activity range are assigned to the first regulation layer, those in the high activity range to the second, those in the medium activity range to the third, and those in the low activity range to the fourth. This grouping reflects the differences in the degree of participation of different regions in differential pressure disturbances within the current control cycle. The system determines the regulation order between groups according to preset rules, so that regions with higher activity levels correspond to earlier regulation orders, i.e., high-activity regions are processed first, followed by medium-activity regions, and finally low-activity regions. If multiple regions belong to the same activity level range, the system further reads the original values ​​of the differential pressure activity values ​​corresponding to these regions and sorts them in descending order of value, so that regions with higher activity values ​​within the same level are ranked first. If there are still cases where differential pressure activity values ​​are the same or nearly the same, the system can continue to introduce secondary comparison parameters for refined sorting, such as comparing the absolute value of the differential pressure offset factor of the region, the number of effective paths participated in by the region, the importance level of the functional area to which the region belongs, or comparing the priority position of the region in the previous cycle to form a sorting result.

[0154] The system binds the ranking results of each region to the region identifier, generating a corresponding priority number for each region. Priority numbers can be represented by integers; for example, 1 represents the highest priority region for adjustment in this cycle, 2 represents the second highest priority region, and so on. Simultaneously, the system can also attach priority weight coefficients to the priority numbers. For instance, the top-ranked regions are marked as high-priority groups, corresponding to higher adjustment resource allocation coefficients; the middle-ranked regions are marked as medium-priority groups, corresponding to medium adjustment weights; and the last-ranked regions are marked as low-priority groups, corresponding to lower adjustment allocation weights.

[0155] Specifically, based on the regional determination data, the differential pressure adjustment direction for each region is determined using the direction determination interval of the differential pressure offset factor, and the adjustment direction is correlated with the region to obtain regional direction control data, which specifically includes:

[0156] The system reads the direction determination label corresponding to the differential pressure offset factor for each region and determines the differential pressure adjustment direction of that region in the current cycle according to the preset direction mapping rules. If the differential pressure offset factor of a region falls into the positive significant offset range, it indicates that the region mainly exhibits a differential pressure offset in the direction of increasing over several consecutive cycles. Based on this offset trend, the system determines its adjustment direction as the direction to suppress the further increase of differential pressure. If the differential pressure offset factor of a region falls into the negative significant offset range, it indicates that the region mainly exhibits a differential pressure offset in the direction of decreasing. Based on this, the system determines its adjustment direction as the direction to compensate for the decrease in differential pressure. If the offset factor falls into the positive weak offset range or the negative weak offset range, the system further marks it as a weak adjustment direction while determining the adjustment direction. If the offset factor is in the direction balance range, the system marks the region as a maintenance direction or a monitoring direction, that is, no significant direction correction is performed in the current cycle, only the original flow configuration is maintained or a low-intensity fine-tuning state is entered.

[0157] The system determines the adjustment direction based on the direction determination result of the current cycle, and can also constrain the direction switching by combining the direction control status of historical cycles. For example, when a region switches from positive adjustment to negative adjustment in the current cycle, but the absolute value of the current offset factor only slightly crosses the boundary of the equilibrium interval, the system may not immediately change the adjustment direction of that region, but instead require the new direction determination to remain consistent for several consecutive cycles before formally updating the control direction. As another example, when the direction determination of a region remains consistent for multiple consecutive cycles, the system can assign a higher stability flag to that direction state and adopt a direction control strategy when setting subsequent adjustment amounts. The system associates the direction information with the region identifier to form region direction control data.

[0158] Specifically, based on regional priority data and regional direction control data, the adjustment direction is converted into fluid inflow and outflow intensities, and the adjustment amplitude is set in conjunction with the regional priority to obtain zonal adjustment data, which includes:

[0159] The system reads the adjustment direction label corresponding to each region. If a region is identified as a pressurization direction, the system converts it to one or more combinations of increasing the fluid inflow intensity, decreasing the fluid outflow intensity, or simultaneously adjusting both according to preset control rules. If a region is identified as a depressurization direction, the system converts it to decreasing the fluid inflow intensity, increasing the fluid outflow intensity, or a corresponding bidirectional linkage. If a region is identified as a maintenance direction, the system sets its inflow and outflow intensity adjustment to zero, a small compensation value, or maintains the output value of the previous cycle. The system pre-establishes basic adjustment templates for different intensity levels of different directions. For example, for a strong pressurization direction, a larger supply air increment or a larger exhaust air reduction can be preset; for a weak pressurization direction, a smaller supply air increment and / or a smaller exhaust air reduction corresponds to it; for a strong depressurization direction, a larger supply air reduction and / or a larger exhaust air increment corresponds to it; and for a weak depressurization direction, a smaller adjustment amount corresponds to it. The system reads the priority number or priority weight coefficient of each region and incorporates this priority information into the calculation of the basic adjustment amount. The higher the priority of a region, the larger its basic adjustment amount multiplied by the priority adjustment coefficient, resulting in a larger final adjustment range. The lower the priority of a region, the smaller its corresponding adjustment coefficient, so even if the direction is the same, the final adjustment amount is lower than that of the high-priority region.

[0160] The fluid inflow intensity adjustment value can be expressed as an increase or decrease in the opening of the air supply valve, a correction of the air supply fan frequency, or a correction of the air supply volume flow rate per unit time; the fluid outflow intensity adjustment value can be expressed as a correction of the opening of the exhaust valve, a correction of the exhaust fan frequency, or a correction of the exhaust volume flow rate per unit time. To ensure that the output adjustment data meets the actual execution capacity of the equipment, the system performs constraint checks on the adjustment values ​​calculated for each region, such as determining whether it exceeds the maximum step size of a single adjustment, whether it exceeds the minimum and maximum control boundaries allowed by the regional equipment, and whether it causes the total air supply volume or total exhaust volume of the system to exceed the overall capacity range. If the initial adjustment calculation result of a certain region exceeds the allowable range, the system will truncate, scale, or perform step-by-step processing according to preset rules. At the same time, when multiple regions simultaneously request an increase in inflow intensity or simultaneously request an increase in outflow intensity, resulting in a limitation on the overall system capacity, the system can allocate the available adjustment capacity in sequence according to the aforementioned priority number, that is, prioritize meeting the adjustment needs of high-priority regions, and then allocate it to subsequent regions within the remaining capacity range. The system encapsulates the corresponding area identifier, priority number, direction control result, inflow intensity adjustment value, and outflow intensity adjustment value for each area into a unified partitioned adjustment data.

[0161] Embodiments of the present invention also provide a fluid pressure dynamic regulation system based on differential pressure feedback, the system comprising:

[0162] The data module is used to acquire differential pressure data between multiple regions;

[0163] The transmission module is used to construct a directed correlation structure based on the differential pressure data and the connection order between regions, and to use the direction of differential pressure change in each adjacent region as the connection direction to obtain differential pressure transmission data.

[0164] The path module is used to trace the pressure difference change path step by step along the connection direction based on the pressure difference transmission data, and compare the pressure difference change of each pressure difference change path before and after to obtain the effective path set;

[0165] The weighting module is used to backtrack through each region in the effective path, starting from the termination region of each effective path, and to distribute the pressure difference changes in each region in a decreasing manner to obtain the reverse influence weight data.

[0166] The differential pressure activity module is used to accumulate the weights of the reverse influences of the same area in different effective paths based on the reverse influence weight data, identify the response intensity of the regional differential pressure, and obtain the differential pressure activity value.

[0167] The direction module is used to mark the increase of pressure difference in each region as a positive direction value and the decrease of pressure difference as a negative direction value based on the pressure difference transmission data, and to splice the direction values ​​of multiple consecutive cycles to obtain the regional direction sequence data;

[0168] The differential pressure offset module is used to identify the length of segments with the same continuous direction based on the regional direction sequence data, extract the direction and proportion with the highest proportion, identify the degree of change of the dominant differential pressure direction, and obtain the differential pressure offset factor.

[0169] The zone adjustment module is used to determine the adjustment priority and direction of each zone by mapping the differential pressure active value and differential pressure offset factor to a preset judgment interval, and to adjust the inflow and outflow intensity of the fluid to obtain zone adjustment data.

[0170] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0171] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0172] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0173] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for dynamic regulation of fluid pressure based on differential pressure feedback, characterized in that, The method includes: Acquire differential pressure data between multiple regions; Based on the differential pressure data, a directed correlation structure is constructed according to the connection order between regions, and the direction of differential pressure change in each adjacent region is used as the connection direction to obtain differential pressure transmission data. Based on the differential pressure transmission data, the differential pressure change path is traced step by step along the connection direction, and the differential pressure change of each differential pressure change path is compared before and after to obtain the effective path set; Based on the set of effective paths, starting from the termination region of each effective path, we trace back through each region in the effective path step by step, and distribute the pressure difference changes in each region in a decreasing manner to obtain the reverse influence weight data. Based on the reverse impact weight data, the reverse impact weights of the same area in different effective paths are superimposed and accumulated to identify the response intensity of regional pressure difference and obtain the pressure difference activity value. Based on the pressure difference transmission data, the increase of pressure difference in each region is marked as a positive direction value and the decrease of pressure difference is marked as a negative direction value. The direction values ​​of multiple consecutive cycles are spliced ​​together to obtain the regional direction sequence data. Based on the regional directional sequence data, identify the length of segments with the same continuous direction, extract the direction and proportion with the highest proportion, identify the degree of change of the pressure difference dominant direction, and obtain the pressure difference offset factor. By mapping the differential pressure activity value and differential pressure offset factor to a preset judgment interval, the adjustment priority and direction of each region are determined, and the inflow and outflow intensities of the fluid are adjusted to obtain the zoned adjustment data.

2. The fluid pressure dynamic regulation method based on differential pressure feedback according to claim 1, characterized in that, Based on the differential pressure data, a directed correlation structure is constructed according to the connection order between regions, and the direction of differential pressure change in each adjacent region is used as the connection direction to obtain differential pressure transmission data, including: Based on the differential pressure data, the spatial adjacency relationship between each region is identified, and adjacent regions are paired according to the connection order to obtain the region connection data; Based on the regional connectivity data, the pressure difference data of each group of adjacent regions is extracted, and the pressure difference changes are arranged in chronological order to obtain the connectivity pressure difference sequence. Based on the connected differential pressure sequence, the differential pressure data at the same time are compared to identify the changing trends of differential pressure from high to low and from low to high, and the changing trends are converted into directional markers to obtain the connected direction data; Based on the connection direction data, the connection relationships of each region are combined with the direction labels, and then arranged in an orderly manner according to the region connection order to obtain directed association data; Based on the directional correlation data, the pressure difference change amplitude of each connection relationship is written into the directional connection to determine the directional information and change amplitude of each connection relationship, thus obtaining the pressure difference transmission data.

3. The fluid pressure dynamic regulation method based on differential pressure feedback according to claim 2, characterized in that, Based on the differential pressure transmission data, the differential pressure change path is traced step by step along the connection direction, and the differential pressure changes of each path are compared before and after to obtain the effective path set, including: Based on the pressure difference transmission data, the connection directions of each region are traversed, and the regions with outward connections are taken as the starting nodes of the path to obtain the path starting point set; Based on the path starting point set, extract subsequent connection nodes level by level along the connection direction of each starting node, and record the node arrangement relationship according to the connection order to obtain the initial path data; Based on the initial path data, the pressure difference change magnitude and direction of each adjacent node in each initial path are read and combined according to the order in the initial path to obtain the path change sequence; Based on the path change sequence, the pressure difference changes at adjacent locations in the initial path are compared sequentially to identify the consistency of the direction of pressure difference change and the continuity of the change amplitude, thus obtaining the path comparison results. Based on the path comparison results, the initial paths that satisfy the conditions of continuous direction and orderly change are retained, while the initial paths with direction interruption or abnormal change are removed, thus obtaining the effective path set.

4. The fluid pressure dynamic regulation method based on differential pressure feedback according to claim 3, characterized in that, Based on the set of effective paths, starting from the termination region of each effective path, the regions within the effective paths are traced back level by level, and the pressure difference changes in each region are distributed in a decreasing manner to obtain the reverse influence weight data, including: Based on the set of valid paths, extract the region nodes located at the end of each valid path and use these region nodes as the path termination nodes to obtain the set of termination regions. Based on the set of termination regions, each valid path is rearranged in reverse order from the termination region to the starting region, and the order of nodes in the reverse arrangement is recorded to obtain the reverse path data. Based on the reverse path data, read the pressure difference change amplitude corresponding to each node in each reverse path, and arrange them according to the node order of the reverse path to obtain the reverse change sequence. Based on the reverse change sequence, the pressure difference change amplitude in the termination region is used as the initial allocation value, and then gradually reduced according to the position order of the nodes in the reverse path to obtain the path allocation sequence. Based on the path allocation sequence, the nodes in each reverse path are associated with the allocation values, and the allocation values ​​of the same region in different paths are aggregated to obtain the reverse influence weight data.

5. The fluid pressure dynamic regulation method based on differential pressure feedback according to claim 4, characterized in that, Based on the reverse impact weight data, the reverse impact weights of the same region in different effective paths are superimposed and accumulated to identify the response intensity of regional pressure difference and obtain the pressure difference activity value, including: Based on the reverse impact weight and pressure difference change magnitude of the same region on a single effective path in the reverse impact weight data, the response intensity of the region to pressure difference disturbance in the effective path is identified, and the regional impact term is obtained. The number of nodes in each effective path is extracted based on the reverse path data, and the path balance of effective paths of different lengths to the regional response is identified to obtain the path balance term. Based on the reverse influence weight data, the degree of path linkage caused by the pressure difference transmission of each node in the effective path to the target area is identified, and the path coupling term is obtained. By fusing regional impact terms, path equilibrium terms, and path coupling terms, the overall response intensity of the target area to differential pressure disturbances under the combined action of multiple effective paths is identified, and the differential pressure activity value is obtained.

6. The method for dynamic fluid pressure regulation based on differential pressure feedback according to claim 5, characterized in that, Based on the regional directional sequence data, the lengths of segments with consecutive identical directions are identified, the direction and proportion with the highest percentage are extracted, the degree of change in the dominant pressure differential direction is identified, and the pressure differential offset factor is obtained, including: Based on the regional directional sequence data, the sum of the lengths of the positive and negative segments in the same region is extracted to identify the bias direction of the pressure difference change in the target region and obtain the directional bias term. Based on the regional direction sequence data, the sum of segment lengths is compared with the sum of all segment lengths in the region to identify the degree of control of the dominant direction over the overall direction in the target region, and the dominant enhancement term is obtained. Based on the regional directional sequence data, the degree of weakening of the stability of the dominant direction when the pressure difference change direction in the target region frequently switches is identified, and the switching inhibition term is obtained; By fusing the directional offset term, the dominant enhancement term, and the switching suppression term, the degree of offset, the duration, and the stability of directional switching of the dominant direction of pressure difference change in the target area are identified, and the pressure difference offset factor is obtained.

7. The fluid pressure dynamic regulation method based on differential pressure feedback according to claim 6, characterized in that, By mapping differential pressure activity values ​​and differential pressure offset factors to preset judgment intervals, the regulation priority and direction of each region are determined, and the inflow and outflow intensities of fluid are adjusted to obtain zonal regulation data, including: By mapping the differential pressure activity value to a preset activity level range and the differential pressure offset factor to a preset direction determination range, area determination data is obtained; Based on the regional assessment data, the activity level ranges of each region are compared to determine the adjustment order between different regions, and then sorted according to the adjustment order to obtain regional priority data. Based on the regional determination data, the differential pressure adjustment direction of each region is determined by the direction determination interval of the differential pressure offset factor, and the adjustment direction is associated with the region to obtain the regional direction control data. Based on the regional priority data and regional direction control data, the adjustment direction is converted into fluid inflow intensity and outflow intensity, and the adjustment amplitude is set in combination with the regional priority to obtain the zonal adjustment data.

8. A fluid pressure dynamic regulation system based on differential pressure feedback, characterized in that, The system is used to perform the method as described in any one of claims 1 to 7, the system comprising: The data module is used to acquire differential pressure data between multiple regions; The transmission module is used to construct a directed correlation structure based on the differential pressure data and the connection order between regions, and to use the direction of differential pressure change in each adjacent region as the connection direction to obtain differential pressure transmission data. The path module is used to trace the pressure difference change path step by step along the connection direction based on the pressure difference transmission data, and compare the pressure difference change of each pressure difference change path before and after to obtain the effective path set; The weighting module is used to backtrack through each region in the effective path, starting from the termination region of each effective path, and to distribute the pressure difference changes in each region in a decreasing manner to obtain the reverse influence weight data. The differential pressure activity module is used to accumulate the weights of the reverse influences of the same area in different effective paths based on the reverse influence weight data, identify the response intensity of the regional differential pressure, and obtain the differential pressure activity value. The direction module is used to mark the increase of pressure difference in each region as a positive direction value and the decrease of pressure difference as a negative direction value based on the pressure difference transmission data, and to splice the direction values ​​of multiple consecutive cycles to obtain the regional direction sequence data; The differential pressure offset module is used to identify the length of segments with the same continuous direction based on the regional direction sequence data, extract the direction and proportion with the highest proportion, identify the degree of change of the dominant differential pressure direction, and obtain the differential pressure offset factor. The zone adjustment module is used to determine the adjustment priority and direction of each zone by mapping the differential pressure active value and differential pressure offset factor to a preset judgment interval, and to adjust the inflow and outflow intensity of the fluid to obtain zone adjustment data.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images