Clean room temperature and humidity data processing method and system based on partition comparison

Abstract

The application discloses a clean room temperature and humidity data processing method and system based on partition comparison, relates to the technical field of data processing, and comprises the following steps: obtaining temperature and humidity sampling data and partition basic data to generate partition window records; generating double-reference-direction records based on current reference boundary records and historical reference boundary records; forming single-reference-difference sections, double-reference-same-direction sections and reference conflict sections according to the double-reference-direction records, and identifying same-direction boundary-crossing clusters; and generating partition deviation source records and a partition comparison report in combination with boundary sharing marks and adjacent partition direction records. The application is used for solving the problem that reference sources are difficult to distinguish in clean room temperature and humidity partition comparison, direction offset and adjacent offset, and improving the traceability of partition deviation source judgment.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method and system for processing cleanroom temperature and humidity data based on zonal comparison. Background Technology

[0002] Cleanrooms are used in semiconductor manufacturing, biopharmaceutical production, precision instrument manufacturing, and electronic component packaging. Temperature and humidity are environmental management parameters in cleanrooms, and they also affect process stability, material condition, product consistency, and the reliability of test results. For example, if the temperature in a local area of ​​a semiconductor manufacturing zone is consistently too high, it may affect the thermal stability of the local process environment; if the humidity in a local area of ​​a biopharmaceutical cleanroom is consistently too high, it may affect the moisture absorption state of materials, microenvironmental fluctuations, or the risk of local condensation. Therefore, cleanrooms typically set up multiple sampling points in different zones and continuously acquire temperature and humidity sampling values.

[0003] Current methods for monitoring cleanroom temperature and humidity typically employ fixed threshold alarms at a single sampling point, or average the values ​​from multiple sampling points within the same zone and compare them to fixed upper and lower limits. While these methods can detect temperature and humidity anomalies significantly exceeding control limits, they still have shortcomings in zone-wide comparison scenarios. First, if a sampling point generates an isolated out-of-limit value, the single-point threshold method may directly classify it as a zone-wide anomaly. Second, if there are differences between the zone to be evaluated and the current reference zone, but the zone to be evaluated has not deviated from its historical reference state, relying solely on the current reference zone will result in a reference source mismatch judgment. Third, if upper and lower temperature deviations or upper and lower humidity deviations alternate, simply counting the number of anomalies or averaging may cause the upper and lower deviations to cancel each other out in the results, resulting in the loss of continuous positional relationships. Finally, if the zone to be evaluated and its adjacent zones simultaneously exhibit the same directional shift, analyzing the zone to be evaluated alone makes it difficult to determine whether the deviation is a continuous shift within the zone or a shift propagated from adjacent zones. Summary of the Invention

[0004] This invention provides a method and system for processing cleanroom temperature and humidity data based on partitioned comparison, which solves the technical problems of reference source mismatch, orientation cancellation, and adjacency offset differentiation in cleanroom temperature and humidity partitioned comparison.

[0005] A method for processing cleanroom temperature and humidity data based on partition comparison includes: acquiring temperature and humidity sampling data and partition basic data, and generating partition window records; generating current reference boundary records and historical reference boundary records based on the partition window records; comparing the partition window to be evaluated with the current reference boundary records and the historical reference boundary records respectively to generate dual reference direction records; generating dual reference segment records including single reference difference segments, dual reference co-directional segments, and reference conflict segments based on the dual reference direction records, and generating co-directional boundary crossing cluster records based on consecutive co-directional boundary crossing positions in the dual reference segment records; generating adjacency consistency records based on the co-directional boundary crossing cluster records, boundary sharing markers, and adjacent partition direction records; generating partition deviation source records based on the dual reference segment records and the adjacency consistency records; and generating a partition comparison report based on the partition deviation source records, sampling verification records, and the partition window records.

[0006] The beneficial effects of this invention are reflected in:

[0007] After acquiring temperature and humidity sampling data and zoning basic data, this invention first records the data in zoning windows, organizing the sampled values ​​according to zoning, sampling points, and a unified time point, rather than directly averaging multiple sampling points within the same zoning. Therefore, subsequent processing can preserve the source of the sampling points, the location of the unified time point, and missing locations, avoiding the masking of isolated sampling anomalies, continuous offset locations, and shared sampling point sources at boundaries by averaging.

[0008] This invention further generates current reference boundary records and historical reference boundary records based on the partition window records, and compares the partition window to be evaluated with the two types of reference boundary records to generate dual reference direction records. This ensures that each temperature and humidity sample value can form an upward, downward, or in-boundary state relative to the current and historical reference sources. Therefore, it can distinguish between single-reference difference cases that are "abnormal relative to the current reference partition but normal relative to their own history," and continuous offset cases where "both types of references cross the boundary in the same direction," reducing the reference source mismatch judgment caused by a single reference source.

[0009] This invention also generates single-reference difference segments, dual-reference co-directional segments, and reference conflict segments based on dual-reference direction records, and generates co-directional boundary overflow cluster records based on continuous co-directional boundary overflow position segments. Furthermore, it combines boundary sharing markers and adjacent partition direction records to generate adjacency-consistent records. Thus, single boundary overflow positions, alternating upward and downward offset fluctuations, continuous co-directional offsets, and synchronous offsets of adjacent partitions can be processed separately, further generating local area continuous offset records, adjacent propagation offset records, or reference source mismatch records. This allows partition comparison reports to trace back to specific reference sources, sampling point identifiers, unified time point locations, and adjacent partition direction records. Attached Figure Description

[0010] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0011] Figure 1 This is a schematic diagram illustrating the steps of the cleanroom temperature and humidity data processing method based on partition comparison according to the present invention. Figure 2 This is a schematic diagram of some steps in S1 of the cleanroom temperature and humidity data processing method based on partition comparison of the present invention. Figure 3 This is a schematic diagram of the data flow timing in S2 of the cleanroom temperature and humidity data processing method based on partition comparison in this invention; Figure 4 This is a schematic diagram of part of step S2 in the cleanroom temperature and humidity data processing method based on partition comparison of the present invention. Figure 5 This is a schematic diagram of the data flow timing in S3 of the cleanroom temperature and humidity data processing method based on partition comparison in this invention; Figure 6 This is a schematic diagram of some steps in step S3 of the cleanroom temperature and humidity data processing method based on partition comparison of the present invention. Figure 7 This is a schematic diagram of the data flow timing in S4 of the cleanroom temperature and humidity data processing method based on partition comparison in this invention; Figure 8 This is a schematic diagram of part of step S4 in the cleanroom temperature and humidity data processing method based on partition comparison of the present invention. Figure 9 This is a schematic diagram of the module structure and data flow of the cleanroom temperature and humidity data processing system based on partition comparison according to the present invention. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0013] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0014] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0015] This invention provides a method for processing cleanroom temperature and humidity data based on zonal comparison, such as... Figure 1 As shown, in one embodiment, the method includes: S1. Acquire temperature and humidity sampling data and basic partition data, and generate partition window records; S2. Generate the current reference boundary record and the historical reference boundary record based on the partition window record. Compare the partition window to be evaluated with the current reference boundary record and the historical reference boundary record respectively to generate a dual reference direction record. S3. Generate a dual-reference segment record including a single-reference difference segment, a dual-reference co-directional segment, and a reference conflict segment based on the dual-reference direction record, and generate a co-directional boundary crossing cluster record based on the continuous co-directional boundary crossing position segments in the dual-reference segment record; generate an adjacency consistency record based on the co-directional boundary crossing cluster record, the boundary sharing marker, and the adjacent partition direction record; and generate a partition deviation source record based on the dual-reference segment record and the adjacency consistency record. S4. Generate a partition comparison report based on the partition deviation source record, sampling verification record, and partition window record.

[0016] In this embodiment, the basic data for zoning includes zoning boundary records, sampling point spatial location records, and zoning level records. Zoning boundary records represent the spatial extent of each zoning zone, sampling point spatial location records represent the location of each sampling point within the cleanroom, and zoning level records represent the cleanliness level or temperature and humidity control level of each zoning zone. Temperature and humidity sampling data includes sampling point identification, sampling time, temperature sampling value, and humidity sampling value. The unit for temperature sampling values ​​is °C, and the unit for humidity sampling values ​​is %RH, where RH represents relative humidity.

[0017] When multiple sampling points exist within the same partition window, the temperature or humidity sampling values ​​from multiple sampling points are not averaged. Instead, temperature and humidity sampling value subsequences are formed separately according to the sampling point identifiers. Each subsequence is arranged according to a unified time point. In this embodiment, the sampling point spatial location record indicates the location of the sampling point within the cleanroom space. The first, second, and nth positions in the partition window record, arranged according to a unified time point, represent the sampling sequence positions, referred to as the unified time point positions. The result of the partition window to be evaluated being located between the upper and lower boundaries after comparison with the reference boundary record is called the boundary-in-boundary state.

[0018] Furthermore, the temperature and humidity sampling data includes sampling point identifiers, sampling time, temperature sampling values, and humidity sampling values. Zonal baseline data includes zonal boundary records, sampling point spatial location records, and zonal level records. Zonal deviation source records include local persistent offset records, adjacent propagation offset records, and reference source mismatch records. The zonal comparison report is used to record whether the temperature and humidity deviation of the zonal being evaluated originates from local persistent offset, synchronous offset from adjacent zonals, or inconsistency between the current reference source and historical reference sources.

[0019] In S1, the temperature and humidity sampling data are first organized into raw sampling records indexed by sampling point identifiers and sampling times, and then converted into aligned sampling records according to a unified time grid. The unified time grid is not an arbitrarily set set of time points, but is determined based on the main sampling intervals of the same sampling point in historical valid sampling records. Taking the current processing cycle from 10:00:00 to 10:29:30 as an example, if the sampling time difference of sampling point P041 in adjacent historical sampling records is mainly 30 seconds, then 60 unified time points are formed within the current processing cycle.

[0020] When the spatial location records of sampling points P041, P042, P043, and P044 all fall within the corresponding range of the partition boundary record of partition A15 to be evaluated, the aligned sampling records of the above sampling points are written into the partition window record of partition A15, but the temperature sampling values ​​of multiple sampling points are not merged into an average value. If P041 forms a temperature sampling value of 23.0℃ and a humidity sampling value of 45.0%RH at the 31st unified time point, then this record is written into the temperature sampling value subsequence and humidity sampling value subsequence of A15 with P041 as the sampling point identifier. This step transforms the discrete temperature and humidity sampling data into data records arranged by partition, sampling point, and unified time point, so that subsequent boundary generation, direction comparison, and same-direction cluster identification have a consistent data basis.

[0021] In S2, partition window records are used to generate current and historical reference boundary records. The current reference boundary record originates from the current reference window within the current processing cycle that has the same cleanliness level or control level as the partition being evaluated. The historical reference boundary record originates from the historical reference window of the partition being evaluated itself. Taking partition A15 as an example, if partitions A08, A09, and A10 have the same partition level as A15, and A08 can form complete temperature and humidity boundary records, the effective sampled value subsequence of A09 shows repeated reversals in direction, and A10 has not formed a complete humidity boundary record, then A08 is selected to generate the current reference boundary record. If the historical window H15-029 of A15 can form a complete boundary record, then a historical reference boundary record is generated based on this historical window.

[0022] The partition window to be evaluated is compared with two types of boundary records. Temperature samples are compared only with temperature boundary records, and humidity samples are compared only with humidity boundary records. For example, if the temperature sample value of A15 at the 31st unified time point is 23.0℃, the current upper boundary of the reference temperature is 22.7℃, and the historical upper boundary of the reference temperature is 22.8℃, then this temperature sample value is upward biased relative to both types of reference sources. This step uses both reference sources from the same period and its own historical reference sources simultaneously to reduce mismatch judgments caused by a single reference source.

[0023] In S3, the dual-reference direction records are further divided into single-reference difference segments, dual-reference same-direction segments, and reference conflict segments. Continuous same-direction boundary-crossing locations within the same sampling point identifier, the same detection quantity type, and adjacent unified time point locations are merged to generate same-direction boundary-crossing cluster records. If the temperature sampling values ​​of P041 at the 31st to 36th unified time point locations all exceed the upper boundary of the current reference temperature and the upper boundary of the historical reference temperature, then these locations form a temperature dual-reference same-direction segment, which further forms a temperature upward bias same-direction cluster record. If a temperature downward bias occurs at the 12th unified time point location, but the 11th and 13th unified time point locations are both within the boundary, then a sampling verification record is generated instead of directly generating a partition bias source record.

[0024] Subsequently, based on the partition boundary records, the partitions adjacent to A15, such as A14 and A16, are determined. The same-direction out-of-bounds cluster records of A15 are then matched with the adjacent partition direction records for the same detection type, same offset direction, and unified time point range. If there are no same-direction records in the adjacent partitions, a continuous offset record for that region is generated; if there are same-direction records in the adjacent partitions, an adjacency propagation offset record is generated. This step further transforms "whether it is out of bounds" into traceable records for "whether it is continuous, whether it is in the same direction, and whether it is synchronized with adjacent partitions."

[0025] In S4, a partition comparison report is generated based on the partition deviation source record, sampling verification record, and partition window record. The partition comparison report also records the processing cycle, the partition identifier to be evaluated, the current reference partition identifier, the historical reference window identifier, the main temperature direction, the main humidity direction, the start and end positions of the same-direction out-of-bounds clusters, the identifiers of participating sampling points, the number of participating sampling points, the boundary sharing participation marker, the adjacency consistency record, and the partition handling prompt record.

[0026] For example, when A15 forms a temperature-biased cluster record at the 31st to 36th unified time points, and adjacent partitions A14 and A16 do not form temperature-biased records, the partition comparison report writes the continuous offset record for this region and generates a verification prompt for this region. If non-boundary shared sampling points in A16 also form temperature-biased records within the overlapping unified time point range, the partition comparison report writes the adjacent propagation offset record and generates an adjacent linkage prompt. If only one of the current reference direction record and the historical reference direction record is out of bounds, while the other is within the boundary, a reference source mismatch record is generated, and a reference review prompt is generated. This step converts the intermediate data processing results into verifiable report fields, enabling subsequent verifiers to trace the reference source, sampling point source, and adjacent partition directions.

[0027] In summary, this invention, through continuous processing of partition window records, current reference boundary records, historical reference boundary records, dual reference direction records, dual reference segment records, same-direction out-of-bounds cluster records, and adjacent consistency records, classifies the sources of temperature and humidity deviation in the partition to be evaluated into local continuous offset, adjacent propagation offset, or reference source mismatch, and reduces the impact of single sampling position, direction cancellation, and single reference source on the partition comparison results.

[0028] like Figure 2 As shown, in one specific embodiment, S1 includes: S11. Generate aligned sampling records based on temperature and humidity sampling data and a unified time grid; S12. Generate a sampling point partition mapping record based on the partition boundary record, sampling point spatial location record, and partition level record; S13. Generate partition window records based on the aligned sampling records and the sampling point partition mapping records.

[0029] In S11, the temperature and humidity sampling data includes sampling point identifiers, sampling times, temperature sampling values, and humidity sampling values. The unified time grid includes multiple unified time points arranged according to the processing cycle. The unified time grid interval is determined based on the main time interval between consecutive sampling times of the same sampling point in the temperature and humidity sampling data. During specific processing, the continuous sampling records saved before the current processing cycle are read, and the time difference between adjacent sampling times is calculated according to the same sampling point identifier. The time difference that occurs most frequently is used as the unified time grid interval. For example, sampling point P041 has consecutive sampling records at 09:58:00, 09:58:30, 09:59:00, and 09:59:30. The time difference between adjacent sampling times is 30 seconds. Therefore, the current processing cycle uses 30 seconds as the unified time grid interval.

[0030] The unified time point alignment range is determined based on the sampling time offset of the same sampling point in the historical valid sampling records. In practice, the offset between the sampling time of the historical valid sampling records and the most recent unified time point is first calculated. After excluding obviously duplicate sampling records, the offset range that covers the normal sampling offset without covering adjacent duplicate sampling records is taken as the alignment range. For example, the historical normal offset of sampling point P021 is concentrated between -0.5s and 0.6s, so ±1.0s is used as the unified time point alignment range for the current processing cycle; therefore, 10:00:30.4 falls within the alignment range of 10:00:30, while 10:00:34.8 does not fall within this alignment range and is marked as an alignment failure sampling record. This processing logic can eliminate the impact of small sampling time offsets on the partition window records, while avoiding filling unaligned data with fictitious values.

[0031] In step S12, a sampling point zoning mapping record is generated based on the zoning boundary record, sampling point spatial location record, and zoning level record. The zoning boundary record includes the zoning identifier, zoning boundary range, and zoning level. The sampling point spatial location record includes the sampling point identifier and the corresponding spatial coordinates. The zoning level record is used to represent the cleanliness level or temperature and humidity control level of each zoning zone. If the spatial location of a sampling point falls within the zoning range corresponding to the zoning boundary record, a mapping relationship between the corresponding sampling point and the corresponding zoning zone is generated.

[0032] For example, the spatial location record of sampling point P026 is X=18.5m, Y=42.0m, Z=2.2m, where X represents the coordinates of the sampling point in the horizontal reference direction of the cleanroom plane, Y represents the coordinates of the sampling point in the vertical reference direction of the cleanroom plane, and Z represents the height coordinates of the sampling point relative to the cleanroom floor reference plane, all in meters. If the partition boundary record of partition A06 covers this location, a partition mapping record between sampling point P026 and partition A06 is generated. If sampling point P026 is simultaneously located at the same level boundary position of partitions A06 and A07, a boundary sharing mark is generated, and the sampling record with the boundary sharing mark is written to the boundary sharing location record of the corresponding adjacent partition. This process ensures the traceability of the correspondence between sampling points and partitions and provides a basis for subsequently excluding self-adjacency consistency caused by boundary sharing sampling points.

[0033] In S13, partitioned window records are generated based on the aligned sampling records and sampling point partition mapping records. Each partitioned window record includes a partition identifier, window sequence number, window start and end time, temperature sampling value subsequence, humidity sampling value subsequence, valid sampling location record, missing location record, and boundary shared location record. When multiple sampling points exist within the same partitioned window record, the temperature or humidity sampling values ​​of the multiple sampling points are not averaged. Instead, temperature and humidity sampling value subsequences are generated separately according to the sampling point identifiers. Each temperature and humidity sampling value subsequence is arranged in a uniform time sequence.

[0034] When generating dual-reference direction records, same-direction out-of-bounds cluster records, and sampling verification records, adjacent unified time points corresponding to the same sampling point identifier are used as adjacent sampling positions. Positions between different sampling points are not directly considered as temporally adjacent. A valid sampling position record indicates that a valid temperature and humidity sample value exists simultaneously at the corresponding sampling point identifier and unified time point, and this sampling record is not marked as an alignment failure, a duplicate unselected record, or a missing position record. A missing position record indicates that the corresponding sampling point identifier did not obtain a valid temperature or humidity sample value at the corresponding unified time point. If multiple sampling records exist within the alignment range of the same sampling point at the same unified time point, the sampling record with the sampling time closest to that unified time point is selected as the valid sampling position record, and the remaining sampling records are written into the duplicate sampling record. For example, if sampling point P021 has records at both 10:00:30.4 and 10:00:30.7, the record at 10:00:30.4, which is closer to 10:00:30, is selected and written into the valid sampling position record.

[0035] In another specific implementation, the processing cycle is determined based on the reporting cycle of cleanroom temperature and humidity data and the continuous sampling capability of the sampling points. Specifically, the processing cycle is determined by reading the periodic report time periods already generated by the environmental monitoring system and prioritizing those consistent with the report time periods. If no separate report time periods are configured, a continuous sampling time period that includes at least two complete sampling sub-sequences and facilitates the formation of boundary records is used as the processing cycle. For example, if the sampling points generate temperature and humidity sampling values ​​every 30 seconds, and the cleanroom monitoring system generates an environmental record every 30 minutes, then the current processing cycle is set to 10:00:00 to 10:29:30, which includes 60 uniform time points. When there are boundary-shared location records within the same zone, these records can participate in the zone window record display. However, when generating adjacent consistent records, they need to be combined with non-boundary-shared location records for joint judgment to avoid misjudging the source of zone deviation due to single-point fluctuations in boundary positions.

[0036] like Figure 3 and Figure 4 As shown, in one specific embodiment, S2 includes: S21. Generate the current reference boundary record and historical reference boundary record based on the partition window record; S22. Compare the partition window to be evaluated with the current reference boundary record and the historical reference boundary record respectively to generate a dual reference direction record; S23. Generate a dual-reference segment record based on the dual-reference direction record, including single-reference difference segment, dual-reference same-direction segment, and reference conflict segment.

[0037] In S21, the current reference boundary record is derived from the current partition window record, and the historical reference boundary record is derived from the historical partition window record. The current reference window is obtained from partition windows that have the same cleanliness level or control level as the partition to be evaluated, excluding partition windows that have not formed temperature or humidity boundary records, as well as partition windows where the direction of change of adjacent sampling positions in the effective temperature sampling value subsequence or effective humidity sampling value subsequence is repeatedly opposite. The partition level record is derived from the cleanroom partition management data and is used to represent the cleanliness level or temperature and humidity control level corresponding to each partition. When generating the current reference window, partition windows of different cleanliness levels or different control levels are not mixed as the current reference source.

[0038] For example, if the partition level record of the partition to be evaluated, A15, is L2, and the partition level records of candidate partitions A08, A09, and A10 are also L2, then these candidate partitions can be included in the current reference window for filtering. If the partition level record of partition A03 is L1, then the current window of A03 will not be used as the current reference window of A15. The direction of change within the candidate reference window is repeatedly reversed, and the determination is based on the valid sample value subsequence corresponding to the same sample point identifier, without using the comparison result between the partition to be evaluated window and the candidate reference window.

[0039] In specific processing, the effective temperature sampling value subsequence or effective humidity sampling value subsequence under the same sampling point identifier is read in a unified time point order, and the increasing or decreasing directions between adjacent effective sampling values ​​are compared. If consecutive combinations of adjacent changing directions of rising, falling, rising, or falling appear in the same subsequence, the candidate reference window is marked as having repeatedly opposite directions and is not given priority as the current reference window. For example, if a certain temperature sampling value subsequence of candidate partition A09 is 22.3℃, 22.6℃, 22.2℃, and 22.7℃, and the adjacent changing directions are rising, falling, and rising in sequence, then this candidate reference window is not given priority as the current reference window.

[0040] In S22, the partition window to be evaluated is compared with both the current reference boundary record and the historical reference boundary record to generate a dual-reference direction record. When generating the current or historical reference boundary record, temperature and humidity sample values ​​are processed separately. For temperature sample values, all valid temperature sample values ​​within the corresponding reference window are read, excluding failed alignment records, missing location records, and duplicate unselected records. When the number of valid temperature sample values ​​is less than two, no temperature boundary record is generated. When the number of valid temperature sample values ​​is at least two, the minimum valid temperature sample value is used as the lower temperature boundary, and the maximum valid temperature sample value is used as the upper temperature boundary.

[0041] For humidity sample values, the lower and upper humidity boundaries are generated in the same manner. Temperature sample values ​​are used only to generate temperature boundary records, and humidity sample values ​​are used only to generate humidity boundary records. For example, if the minimum effective temperature sample value for the current reference window A08 is 22.1℃ and the maximum is 22.7℃, then the current reference temperature boundary record is 22.1℃ to 22.7℃. When the temperature sample value of the partition window to be evaluated exceeds the upper boundary of the current reference temperature, it is recorded as the upper offset position of the current reference temperature; when it does not reach the lower boundary of the current reference temperature, it is recorded as the lower offset position of the current reference temperature.

[0042] When a temperature sample value equals the upper boundary of the current reference temperature, the lower boundary of the current reference temperature, the upper boundary of the historical reference temperature, or the lower boundary of the historical reference temperature, the temperature sample value is recorded as being within the boundary, and not as being at an upper or lower temperature deviation. Humidity sample values ​​are processed in the same way. For example, if the upper boundary of the current reference temperature is 22.7℃, and the temperature sample value is 22.7℃, this sample value does not exceed the upper boundary of the current reference temperature, so it is recorded as being within the boundary; if the temperature sample value is 22.8℃, it is recorded as being at an upper deviation of the current reference temperature.

[0043] In S23, a dual-reference segment record is generated based on the dual-reference direction record. The dual-reference segment record is a position segment record formed based on the consistency relationship, unilateral boundary crossing relationship, and opposite direction relationship between the current reference direction record and the historical reference direction record. A single-reference difference segment is a position segment in which one of the current reference direction record and the historical reference direction record shows an upward or downward offset position, while the other is within the boundary. A dual-reference same-direction segment is a position segment in which both the current reference direction record and the historical reference direction record show the same detection quantity type and the same offset direction. A reference conflict segment is a position segment in which both the current reference direction record and the historical reference direction record show the same detection quantity type but opposite offset directions.

[0044] For example, if A15 is positioned at the 31st to 36th unified time points relative to both the current and historical reference boundary records, a temperature dual-reference unidirectional segment is generated. If A15 is positioned at the 41st to 44th unified time points relative to the current reference humidity boundary record, but within the boundary relative to the historical reference humidity boundary record, a humidity single-reference difference segment is generated. If A15 is positioned at the 50th to 52nd unified time points relative to the current reference temperature boundary record, but at the 50th to 52nd unified time points relative to the current reference temperature boundary record, but at the 50th to 52nd unified time points relative to the current reference temperature boundary record, but at the 50th to 52nd unified time points relative to the historical reference temperature boundary record, a temperature reference conflict segment is generated.

[0045] When the current or historical reference boundary record is unavailable, a reference availability flag is generated and written to the partition comparison report. The reference availability flag is not considered a partition offset source record and is not listed alongside local persistent offset records, adjacent propagation offset records, or reference source mismatch records. A partition offset source record is only generated after both the current and historical reference boundary records are available for comparison and a dual-reference direction record is formed.

[0046] In another specific implementation, when multiple candidate partition windows satisfying the same cleanliness level or control level exist within the same processing cycle, candidate partition windows that have not formed temperature or humidity boundary records are first excluded. Then, candidate partition windows whose adjacent sampling positions within the same window repeatedly change in opposite directions, failing to form a stable reference boundary, are also excluded. Among the remaining candidate partition windows, the one with the most valid sampling positions is selected as the current reference window. If the number of valid sampling positions is the same, the candidate partition window with fewer boundary-shared location records is selected. If they are still the same, the candidate partition window selected first according to the fixed order of the partition identifiers is chosen. For example, if candidate partition A08 has 232 valid sampling positions, A09 has 219 valid sampling positions with repeatedly opposite directions, and A10 has not formed a complete humidity boundary record, then A08 is selected as the current reference window.

[0047] The historical reference window searches through historical partition window records from most recent to oldest, without setting a fixed threshold for the number of historical windows. During processing, it first reads the previous historical partition window record adjacent to the current processing cycle, determining if it has valid temperature and humidity boundary records and if there are no repeated reversals in the internal direction of change. If these conditions are met, the historical partition window record is used as the historical reference window. If not, it continues reading even earlier historical partition window records until a usable historical reference window is found or the starting position of a saved historical record is located. For example, if H15-030 of A15 lacks a humidity boundary record, while H15-029 has complete temperature and humidity boundary records, then H15-029 is selected as the historical reference window.

[0048] like Figure 5 and Figure 6 As shown, in one specific embodiment, S3 includes: S31. Generate a cluster record of cross-boundary positions in the same direction based on the consecutive cross-boundary positions in the dual-reference segment record; S32. Generate adjacency-consistent records based on the same-direction out-of-bounds cluster records, boundary sharing markers, and adjacent partition direction records; S33. Generate partition deviation source records based on dual-reference segment records and adjacent consistency records.

[0049] In S31, the same-direction out-of-bounds cluster record is used to express the continuous directional structure of out-of-bounds positions within the partition window. Continuous same-direction out-of-bounds position segments are generated according to the same sampling point identifier, the same detection quantity type, and adjacent unified time points. If adjacent unified time points of the same sampling point all form out-of-bounds positions in the same direction, and there are no missing position records, failed alignment sampling records, or out-of-bounds positions in the opposite direction in between, then these positions are merged into a continuous same-direction out-of-bounds position segment; if a state within the boundary occurs in the middle, then the continuous same-direction out-of-bounds position segment is interrupted at the state within the boundary.

[0050] For example, if sampling point P041 forms an upwardly biased temperature position at all unified time points from the 31st to the 36th, then a single upwardly biased in-direction cluster record is generated. If the 34th unified time point is within the boundary, then unified time points from the 31st to the 33rd and unified time points from the 35th to the 36th form different in-direction out-of-bounds cluster records. The in-direction out-of-bounds cluster record includes the sampling point identifier, detection quantity type, offset direction, starting unified time point position, ending unified time point position, and the number of covered sampling positions. The number of covered sampling positions is calculated based on the number of consecutive unified time point positions corresponding to the same sampling point identifier, the same detection quantity type, and the same offset direction, not directly based on the start and end time difference. For example, if P041 forms an upwardly biased temperature position at all unified time points from the 31st to the 36th, then the number of covered sampling positions is 6.

[0051] When alternating upper and lower offset positions appear within the same detection quantity type corresponding to the same sampling point identifier, an alternating cluster record is generated. This alternating cluster record does not directly generate a continuous offset record or an adjacent propagation offset record for the current region; instead, it is used in the partition comparison report to indicate that the partition window exhibits repeated directional changes. When adjacent time points of a single out-of-bounds position are all within the boundary, a sampling verification record is generated, and the single out-of-bounds position is not directly used as a source of partition bias.

[0052] In step S32, before generating adjacency consistency records, adjacent partitions that share a boundary or are directly adjacent to the partition to be evaluated are identified based on the partition boundary records. Adjacent partition window records within the same processing cycle are read, and adjacent partition direction records are generated according to the same rules used to generate dual-reference direction records for the partition window to be evaluated. Adjacent partition direction records include adjacent partition identifiers, a unified time point, temperature upward offset direction records, temperature downward offset direction records, humidity upward offset direction records, and humidity downward offset direction records. Adjacent partitions are determined based on the spatial boundary relationships in the partition boundary records, not through manually preset distance thresholds.

[0053] In practice, the partition boundary range of the partition to be evaluated and the partition boundary range of other partitions are read. If two partitions share a boundary line, a boundary surface, or are registered as directly adjacent in the partition boundary record, then the corresponding partitions are determined to be adjacent partitions. If there are other partitions separating the two partitions, then they are not determined to be adjacent partitions. For example, if partition A15 and partition A16 share a boundary surface in the partition boundary record, then A16 is an adjacent partition of A15. If partition A15 and partition A18 are separated by partition A16, then A18 is not considered a directly adjacent partition of A15 and participates in the adjacency consistency judgment.

[0054] Adjacency-consistent records are matched according to detection type, offset direction, and uniform time point range. If a cross-boundary cluster record in the same direction of the partition to be evaluated overlaps with a directional record in an adjacent partition in the same detection type, the same offset direction, and at least one uniform time point, then the adjacency-consistent record is written to the adjacent partition's same-direction record; otherwise, the adjacency-consistent record is written to a record that does not include a directional record in an adjacent partition. The range of overlapping uniform time points is obtained by the intersection of the same uniform time point positions in the cross-boundary cluster records in the partition to be evaluated and the directional records in the adjacent partition, without setting an additional overlap ratio threshold.

[0055] In S33, a partition deviation source record is generated based on the dual-reference segment record and the adjacent consistency record. When there are dual-reference co-directional segments in the dual-reference segment record and the adjacent consistency record does not contain adjacent partition co-directional records, a local area continuous offset record is generated; when there are dual-reference co-directional segments in the dual-reference segment record and the adjacent consistency record contains adjacent partition co-directional records, an adjacent propagation offset record is generated; when there are single-reference difference segments or reference conflict segments in the dual-reference segment record and dual-reference co-directional segments are not formed within the same detection type and the same location segment, a reference source mismatch record is generated.

[0056] When different detection types or different unified time point ranges simultaneously exist within the same partition window record, corresponding to dual-reference co-directional segments, single-reference difference segments, and reference conflict segments, partition deviation source records are generated separately according to detection type and location segment, without forcibly merging different detection types or different location segments into a single source conclusion. For location segments forming dual-reference co-directional segments, adjacent consistent records are combined to generate local area continuous offset records or adjacent propagation offset records; for location segments forming only single-reference difference segments or reference conflict segments and not forming dual-reference co-directional segments within the same detection type and the same location segment, reference source mismatch records are generated. For example, if the temperature of A15 forms dual-reference co-directional segments at the 31st to 36th unified time point positions, and the humidity forms a single-reference difference segment at the 41st to 44th unified time point positions, then the temperature location segment and the humidity location segment will generate corresponding source records respectively.

[0057] In another specific implementation, when multiple sampling points within the same partition window form co-directional out-of-bounds cluster records with the same detection type and the same offset direction, a co-directional cluster summary record within the partition is generated according to the detection type, offset direction, and the overlap relationship of unified time points. If co-directional out-of-bounds cluster records from different sampling points overlap at at least one unified time point, they are written into the co-directional cluster summary record within the same partition, and the identifier of each sampling point, their respective start and end unified time point positions, and the range of overlapping unified time points are recorded; if there are no overlapping unified time point positions, they are written into different co-directional cluster summary records within different partitions. For example, if P041 forms a temperature-biased co-directional cluster record at unified time points 31 to 36, and P042 also forms a temperature-biased co-directional cluster record at unified time points 32 to 35, then both are written into the same co-directional cluster summary record within the same partition, with the overlapping unified time point range being unified time points 32 to 35.

[0058] When generating adjacency consistency records, if the orientation record of the partition to be evaluated and the orientation record of the adjacent partition come from the same sampling point identifier and that sampling point identifier has a boundary sharing mark, then the orientation record is only retained as a boundary sharing location record and is not used as the basis for the adjacent partition same orientation record alone. Only when at least one non-boundary sharing sampling point identifier in the adjacent partition forms an orientation record within the same detection quantity type, the same offset direction, and the same time point overlap, will the adjacency consistency record be written to include the adjacent partition same orientation record. For example, if P044 belongs to the boundary sharing location records of both A15 and A16, an adjacency propagation offset record cannot be generated solely based on P044 being temperature-biased in both partitions; only if the non-boundary sharing sampling point P049 in A16 also forms a temperature-biased orientation record at the corresponding time point will an adjacency consistency record containing the adjacent partition same orientation record be generated. If the adjacent partition cannot form a current reference boundary record or a historical reference boundary record, then the adjacent partition will not be used to generate an adjacency consistency record containing the adjacent partition same orientation record, and an adjacent partition reference unavailable mark will be written in the partition comparison report. If at least one of the multiple adjacent partitions forms a usable adjacent partition direction record, then only the usable adjacent partition direction record is used for adjacency consistency determination.

[0059] like Figure 7 and Figure 8 As shown, in one specific embodiment, S4 includes: S41. Generate partition result records based on partition deviation source records, sampling verification records, and partition window records; S42. Generate partition handling prompt records based on the partitioning results; S43. Generate a partition comparison report based on the partition result record and partition handling prompt record.

[0060] In S41, the partitioning result record includes the partition identifier to be evaluated, the current reference partition identifier, the historical reference window identifier, the main temperature direction, the main humidity direction, the start and end positions of the same-direction boundary crossing clusters, the identifiers of participating sampling points, the number of participating sampling points, the boundary sharing participation marker, and the source of partitioning deviation. The main temperature direction and the main humidity direction are generated based on the same-direction boundary crossing cluster records of the corresponding detection quantity type. If there is only one direction of same-direction boundary crossing cluster records for the corresponding detection quantity type, then that direction is the main direction; if there are same-direction boundary crossing cluster records in different directions, then the direction covering more sampling positions is taken as the main direction, and the other direction is written into the auxiliary direction record; if the number of covered sampling positions is the same, then the main direction record is an alternating direction; if there are no same-direction boundary crossing cluster records, then the main direction record is the state within the boundary.

[0061] The main directions of temperature and humidity are generated based on the number of sampling locations covered by the same-direction out-of-bounds cluster records of the corresponding detection type, and are not generated using the average of temperature and humidity sampling values. For example, if the temperature-biased same-direction cluster records in A15 cover 6 locations at the same time point, and there are no temperature-biased same-direction cluster records, then the main direction of temperature is temperature-biased. If only sampling point P041 forms a temperature-biased same-direction cluster record within partition A15, then the partition result record will include the sampling point identifier as P041, the number of participating sampling points as 1, and the partition comparison report will state that the result comes from the sampling point subsequence corresponding to P041, and will not express that all sampling points within the partition have experienced temperature-biased behavior.

[0062] In S42, partition handling prompt records are generated based on the partitioning result records. Local area continuous offset records correspond to local area verification prompts, adjacent propagation offset records correspond to adjacent linking prompts, reference source mismatch records correspond to reference review prompts, and sampling review records correspond to sampling review prompts. Partition handling prompt records are generated only based on partitioning deviation source records and sampling review records, and are used to indicate the direction of subsequent verification; they are not considered as confirmed physical causes.

[0063] In specific processing, continuous offset records in this area generate a local area verification prompt, adjacent propagation offset records generate an adjacent linkage prompt, reference source mismatch records generate a reference review prompt, and sampling review records generate a sampling review prompt. The prompt content can list the objects requiring verification, but should not list data not entered into this method as having identified causes. For example, when A15 generates a local area continuous offset record, the report can prompt verification of local heat sources, sampling arrangements, or obstruction in A15, but cannot state that the cost method has confirmed the offset was caused by a local heat source. When A15 generates an adjacent propagation offset record, an adjacent linkage prompt is generated, prompting simultaneous verification of the common boundary or common environmental impact of A15 and A16; when A15 generates a reference source mismatch record, a reference review prompt is generated, prompting verification of the current reference partition A08 or reselecting the current reference boundary record; when a single sampling location forms a sampling review record, a sampling review prompt is generated.

[0064] In S43, a partition comparison report is generated based on the partition result record and partition handling prompt record. The partition comparison report includes the processing cycle, the partition identifier to be evaluated, the current reference partition identifier, the historical reference window identifier, the single reference difference segment, the dual reference same direction segment, the reference conflict segment, the same direction out-of-bounds cluster record, the same direction cluster summary record within the partition, the adjacency consistency record, the partition deviation source record, the sampling verification record, and the partition handling prompt record.

[0065] For example, a partition comparison report might record the following: the processing period is 2026-07-09-1000; the partition to be evaluated is identified as A15; the current reference partition is identified as A08; the historical reference window is identified as H15-029; the dual-reference unidirectional segment is from the 31st unified time point to the 36th unified time point; the unidirectional out-of-bounds cluster record is a temperature-biased unidirectional cluster record; the sampling point identifier is P041; the number of sampling points is 1; the adjacent consistent record is one that does not include adjacent partition unidirectional records; the partition deviation source record is a continuous offset record for this region; and the partition handling prompt record is a verification prompt for this region. This report records the reference source, offset direction, number of continuous sampling positions, and adjacent partition direction records, providing more verifiable fields compared to simply outputting out-of-limit results. If the current reference boundary record, historical reference boundary record, or adjacent partition direction record is unavailable, a corresponding reference availability flag or adjacent partition reference unavailable flag is written in the report, and it is not considered a partition deviation source record.

[0066] In another specific implementation, the zonal comparison report can be continuously output according to the processing cycle, and a historical report index is retained. If the same zonal to be evaluated generates continuous offset records for multiple consecutive processing cycles, the number of consecutive processing cycles can be indicated in subsequent reports; if the same zonal to be evaluated repeatedly generates reference source mismatch records in different processing cycles, the reference source can be indicated as needing verification. The above supplement is only for report organization and does not change the generation logic of zonal deviation source records. If different source records in the temperature direction and humidity direction exist simultaneously within the same zonal window, the zonal comparison report will list them separately according to the detection quantity type, without overwriting the temperature deviation source and humidity deviation source. For example, if the temperature is a continuous offset record for this zone at the 31st to 36th unified time points, and the humidity is a reference source mismatch record at the 41st to 44th unified time points, then the report will record the temperature source and humidity source separately.

[0067] This invention also provides a cleanroom temperature and humidity data processing system based on zonal comparison, such as... Figure 9 As shown, it includes: a data alignment module, a partition window module, a dual-reference processing module, a co-directional cluster identification module, a deviation source judgment module, and a report generation module.

[0068] The data alignment module is used to generate aligned sampling records based on temperature and humidity sampling data and a unified time grid; the temperature and humidity sampling data includes sampling point identifiers, sampling time, temperature sampling values, and humidity sampling values.

[0069] The partition window module is used to generate a sampling point partition mapping record based on the partition boundary record, sampling point spatial location record, and partition level record, and to generate a partition window record based on the aligned sampling record and the sampling point partition mapping record.

[0070] The dual-reference processing module is used to generate current reference boundary records and historical reference boundary records based on current partition window records and historical partition window records, generate dual-reference direction records based on the partition window records to be evaluated, current reference boundary records and historical reference boundary records, and generate dual-reference segment records based on dual-reference direction records.

[0071] The same-direction cluster identification module is used to generate same-direction out-of-bounds cluster records and sampling verification records based on the dual-reference segmented records.

[0072] The deviation source determination module is used to generate adjacency consistent records based on the same-direction out-of-bounds cluster records, boundary sharing markers, and adjacent partition direction records, and to generate partition deviation source records based on the dual-reference segment records and adjacency consistent records.

[0073] The report generation module is used to generate partition result records based on partition deviation source records, sampling verification records, and partition window records; generate partition handling prompt records based on partition result records; and generate partition comparison reports based on partition result records and partition handling prompt records.

[0074] In one specific implementation, the system can be deployed on a cleanroom environment monitoring server, industrial computer, or edge computing terminal. Modules can exchange data via database tables, cached records, or message queues. The system can complete zoned temperature and humidity data processing without requiring the use of digital twin models, reinforcement learning models, image recognition models, or air conditioning control models. This architecture is compatible with existing cleanroom temperature and humidity acquisition platforms; it only needs to acquire temperature and humidity sampling data, zone boundary records, sampling point spatial location records, zone level records, and historical zone window records to generate a zone comparison report.

[0075] To facilitate those skilled in the art to understand and implement the technical solutions described in this specification, the following describes the processing method and system for cleanroom temperature and humidity data processing based on zonal comparison, using a scenario containing specific parameters and data.

[0076] A cleanroom comprises 4 functional zones and 20 local zones, labeled A01 to A20. Eighty temperature and humidity sampling points are set up within the cleanroom, labeled P001 to P080. Temperature and humidity values ​​are generated every 30 seconds. The current processing cycle is from 10:00:00 to 10:29:30, resulting in 60 unified time points. If 10:00:00 is taken as the first unified time point, then the 31st unified time point is 10:15:00, and the 36th unified time point is 10:17:30.

[0077] The 31st to 36th unified time points cover 6 sampling locations, each corresponding to a unified time interval of 30 seconds. Therefore, the covered sampling period is 6 × 30 seconds = 180 seconds = 3 minutes. The start and end time difference between the 31st and 36th unified time points is 5 × 30 seconds = 150 seconds = 2.5 minutes. In this embodiment, the term "covered sampling period" is used to describe the same-direction out-of-bounds cluster records, instead of "start and end time difference" to replace the covered sampling period.

[0078] In S1, sampling points P041, P042, P043, and P044 are mapped to the evaluation zone A15 through sampling point partition mapping records. Sampling point P041 generates temperature sampling values ​​of 23.0℃, 23.1℃, 23.2℃, 23.2℃, 23.3℃, ​​and 23.4℃ at the 31st to 36th unified time points, and generates humidity sampling values ​​of 45.0%RH, 45.1%RH, 45.1%RH, 45.2%RH, 45.1%RH, and 45.0%RH.

[0079] The aforementioned temperature and humidity sampling values ​​are written into the P041 temperature sampling value subsequence and P041 humidity sampling value subsequence of A15, respectively, and are not averaged with the sampling values ​​of P042, P043, and P044. If P044 is located at the boundary between A15 and A16, the sampling record of P044 also carries a boundary sharing marker, and is not used as a separate basis for adjacent partition records in the same direction in subsequent adjacency consistency judgment.

[0080] In S2, the current reference zone A08 records a current reference temperature boundary of 22.1℃ to 22.7℃ and a current reference humidity boundary of 44.1%RH to 45.5%RH. The historical reference window H15-029 of the zone to be evaluated A15 records a historical reference temperature boundary of 22.0℃ to 22.8℃ and a historical reference humidity boundary of 44.0%RH to 45.6%RH. Both the temperature and humidity ranges mentioned above are within the common temperature and humidity operating range of cleanrooms. The temperature deviation is also within the range that can be continuously recorded by sensors in local zone comparisons, and does not involve situations below the freezing point, above the acceptable environmental range for humans and equipment, or humidity exceeding saturation.

[0081] At the 31st unified time point, the temperature sample value of 23.0℃ exceeds the upper boundary of the current reference temperature of 22.7℃ by an excess of 23.0℃ - 22.7℃ = 0.3℃; it also exceeds the upper boundary of the historical reference temperature of 22.8℃ by an excess of 23.0℃ - 22.8℃ = 0.2℃. At the 36th unified time point, the temperature sample value of 23.4℃ exceeds the upper boundary of the current reference temperature by 23.4℃ - 22.7℃ = 0.7℃, and exceeds the upper boundary of the historical reference temperature by 23.4℃ - 22.8℃ = 0.6℃. Therefore, the 31st to 36th unified time points all form a dual-reference, same-direction temperature segment.

[0082] In S2, the humidity sampling values ​​of 45.0%RH to 45.2%RH are all within the current reference humidity boundary record of 44.1%RH to 45.5%RH, and also within the historical reference humidity boundary record of 44.0%RH to 45.6%RH. Therefore, the humidity direction is within the boundary. This comparison process shows that temperature and humidity are compared separately, and there is no dimensional mixing. If, in another processing cycle, A15 is temperature-biased relative to the current upper boundary of the reference temperature at the 20th to 25th unified time point, but is within the boundary relative to the historical reference temperature boundary record, then this position segment generates a single reference difference segment, instead of directly generating a continuous offset record for this area. If a position segment is temperature-biased relative to the current reference temperature boundary record, but temperature-biased relative to the historical reference temperature boundary record, then this position segment generates a reference conflict segment and is used for subsequent reference source mismatch records.

[0083] In S3, the temperature directions at the 31st to 36th unified time points are all upward-biased, and therefore are merged into a cluster record of upward-biased, same-direction temperature changes. This cluster record covers six consecutive unified time points, corresponding to a sampling period of 180 seconds (3 minutes), indicating that the upward temperature deviation is not an isolated jump at a single unified time point. If the temperature sample value at the 12th unified time point is 21.8℃, but the 11th and 13th unified time points are both within the boundary, then the 12th unified time point only generates a sampling verification record, and does not directly generate a zoning deviation source record. This 21.8℃ temperature sample value still belongs to the local low values ​​that may occur in the cleanroom environment and does not involve physical anomalies.

[0084] If P042 also forms a temperature-biased cluster record at the 32nd to 35th unified time points, then the temperature-biased cluster records of P041 and P042 are written into the same partition's unified cluster summary record, with the overlapping unified time point range being the 32nd to 35th unified time point positions.

[0085] Continuing in S3, if adjacent partitions A14 and A16 do not form temperature upward bias records at the 31st to 36th unified time points, then the adjacency consistency record does not include adjacent partition same-direction records. Combined with the dual-reference same-direction segment, a continuous offset record for this region is generated. If the non-boundary shared sampling point P049 of adjacent partition A16 also forms temperature upward bias records at the 32nd to 37th unified time points, then the temperature upward bias same-direction cluster records of A15 and A16 overlap at the 32nd to 36th unified time points, generating an adjacency consistency record containing adjacent partition same-direction records. Combined with the dual-reference same-direction segment, an adjacency propagation offset record is generated. If only the boundary shared sampling point P044 forms temperature upward bias records simultaneously in A15 and A16, and A16 has no non-boundary shared sampling points forming records in the same direction, then an adjacency propagation offset record is not generated separately based on P044.

[0086] In S4, the zoning comparison report organizes the above data into traceable results. For example, when a continuous offset record for this zone is generated, the zoning comparison report records: processing period 2026-07-09-1000, zone to be evaluated A15, current reference zone A08, historical reference window H15-029, temperature main direction is upward temperature deviation, humidity main direction is within the boundary, the start and end positions of the same-direction out-of-bounds clusters are the positions of the 31st to 36th unified time points, the sampling point identifier is P041, the number of sampling points is 1, the source of zoning deviation is the continuous offset record for this zone, and the zoning handling prompt record is the verification prompt for this zone.

[0087] This report does not indicate that all sampling points within the zone have experienced temperature upward deviation; instead, it records that the result originates from the subsequence of sampling points corresponding to P041. If only a sampling verification record is generated in another processing cycle, the report generates a sampling verification prompt, but does not expand it to a source of zone deviation. If both a continuous temperature deviation record and a humidity reference source mismatch record exist simultaneously, the report records them separately according to the type of detection quantity to avoid overlapping of temperature and humidity results.

[0088] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0089] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0090] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for processing cleanroom temperature and humidity data based on zonal comparison, characterized in that, The methods include: Acquire temperature and humidity sampling data and basic partition data, and generate partition window records; Based on the partition window record, generate the current reference boundary record and the historical reference boundary record, and compare the partition window to be evaluated with the current reference boundary record and the historical reference boundary record respectively to generate a dual reference direction record; A dual-reference segmented record is generated based on the dual-reference direction record, including a single-reference difference segment, a dual-reference co-directional segment, and a reference conflict segment. A co-directional out-of-bounds cluster record is generated based on the consecutive co-directional out-of-bounds position segments in the dual-reference segmented record. An adjacency-consistent record is generated based on the same-direction out-of-bounds cluster record, boundary sharing marker, and adjacent partition direction record; a partition deviation source record is generated based on the dual-reference segment record and the adjacency-consistent record. A partition comparison report is generated based on the partition deviation source record, sampling verification record, and partition window record.

2. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 1, characterized in that, The process of acquiring temperature and humidity sampling data and partition basic data, and generating partition window records, includes: Aligned sampling records are generated based on the temperature and humidity sampling data and the unified time grid; wherein, the temperature and humidity sampling data includes sampling point identifier, sampling time, temperature sampling value and humidity sampling value, and the temperature sampling value and humidity sampling value that fall within the same unified time point alignment range are assigned to the corresponding unified time point, and the sampling records whose sampling time does not fall within the corresponding unified time point alignment range are marked as alignment failure sampling records. A sampling point partition mapping record is generated based on the partition boundary record, the sampling point spatial location record, and the partition level record; wherein, when the spatial location of a sampling point falls within the partition range corresponding to the partition boundary record, a mapping relationship between the corresponding sampling point and the corresponding partition is established; The partition window record is generated based on the aligned sampling record and the sampling point partition mapping record; wherein, each partition window record includes a partition identifier, window sequence number, window start and end time, temperature sampling value subsequence, humidity sampling value subsequence, valid sampling location record, missing location record, and boundary shared location record.

3. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 2, characterized in that, The process of generating a sampling point partition mapping record based on partition boundary records, sampling point spatial location records, and partition level records includes: Based on the spatial location record of the sampling point and the partition boundary record, it is determined whether the position corresponding to the spatial location record of the sampling point falls within the partition range; When the location corresponding to the spatial location record of the sampling point falls within the partition range, a sampling point partition mapping record is generated, including sampling point identifier, partition identifier, partition level, partition grade, and boundary sharing marker. When the location corresponding to the spatial location record of the sampling point is at the boundary of two partitions at the same level, a boundary sharing mark is generated, and the sampling record with the boundary sharing mark is written into the boundary sharing location record of the corresponding adjacent partition.

4. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 1, characterized in that, The step of generating the current reference boundary record and the historical reference boundary record based on the partition window record includes: A current reference window is generated based on the current partition window record; wherein, the current reference window is obtained from the partition window with the same cleanliness level or the same control level as the partition to be evaluated, and partition windows that have not formed temperature boundary records or humidity boundary records, as well as partition windows in the effective temperature sampling value subsequence or effective humidity sampling value subsequence where the direction of change of adjacent sampling positions is repeatedly opposite. A historical reference window is generated based on the historical partition window records; wherein, the historical reference window is obtained from the historical partition window records of the partition to be evaluated itself, and historical windows that have not formed temperature boundary records or humidity boundary records, as well as historical windows in the effective temperature sampling value subsequence or effective humidity sampling value subsequence where the direction of change of adjacent sampling positions is repeatedly opposite; The current reference boundary record is generated based on the current reference window, and the historical reference boundary record is generated based on the historical reference window; wherein, the temperature sample value is only used to generate the temperature boundary record, and the humidity sample value is only used to generate the humidity boundary record.

5. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 4, characterized in that, The step of comparing the partition window to be evaluated with the current reference boundary record and the historical reference boundary record respectively to generate a dual-reference direction record includes: A current reference direction record is generated based on the partition window record to be evaluated and the current reference boundary record; wherein, when the temperature sample value exceeds the upper boundary of the current reference temperature, it is recorded as the upper offset position of the current reference temperature, and when the temperature sample value does not reach the lower boundary of the current reference temperature, it is recorded as the lower offset position of the current reference temperature; when the humidity sample value exceeds the upper boundary of the current reference humidity, it is recorded as the upper offset position of the current reference humidity, and when the humidity sample value does not reach the lower boundary of the current reference humidity, it is recorded as the lower offset position of the current reference humidity. Historical reference direction records are generated based on the partition window records to be evaluated and the historical reference boundary records; wherein, when the temperature sampling value exceeds the upper boundary of the historical reference temperature, it is recorded as the upper offset position of the historical reference temperature, and when the temperature sampling value does not reach the lower boundary of the historical reference temperature, it is recorded as the lower offset position of the historical reference temperature; when the humidity sampling value exceeds the upper boundary of the historical reference humidity, it is recorded as the upper offset position of the historical reference humidity, and when the humidity sampling value does not reach the lower boundary of the historical reference humidity, it is recorded as the lower offset position of the historical reference humidity. The dual reference direction record is generated based on the current reference direction record and the historical reference direction record, and the upward bias direction of temperature, the downward bias direction of temperature, the upward bias direction of humidity, the downward bias direction of humidity, and the state within the boundary are retained respectively.

6. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 5, characterized in that, The process of generating a dual-reference segmented record, including a single-reference difference segment, a dual-reference co-directional segment, and a reference conflict segment, based on the dual-reference direction record includes: A single reference difference segment is generated based on the current reference direction record and the historical reference direction record; wherein, the single reference difference segment is a position segment in which one of the current reference direction record and the historical reference direction record is in an upward or downward offset position, and the other is in a state within the boundary; A dual-reference co-directional segment is generated based on the current reference direction record and the historical reference direction record; wherein, the dual-reference co-directional segment is a position segment that appears in both the current reference direction record and the historical reference direction record with the same detection quantity type and the same offset direction; A reference conflict segment is generated based on the current reference direction record and the historical reference direction record; wherein, the reference conflict segment is a position segment in the current reference direction record and the historical reference direction record where the same detection quantity type appears but with opposite offset directions.

7. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 6, characterized in that, The step of generating a co-directional out-of-bounds cluster record based on consecutive co-directional out-of-bounds position segments in the dual-reference segmented record includes: Based on the continuous same-direction temperature upward offset positions in the dual-reference segmented records, generate temperature upward offset same-direction cluster records; based on the continuous same-direction temperature downward offset positions in the dual-reference segmented records, generate temperature downward offset same-direction cluster records. Humidity upward bias cluster records are generated based on the continuous upward bias positions of humidity in the same direction in the dual-reference segmented records, and humidity downward bias cluster records are generated based on the continuous downward bias positions of humidity in the same direction in the dual-reference segmented records. When alternating upper and lower offset positions appear in the same type of detection quantity corresponding to the same sampling point identifier, an alternating cluster record is generated; When the adjacent time points of a single out-of-bounds location are all within the boundary, a sampling verification record is generated, and the single out-of-bounds location is not directly used as the source of partitioning deviation.

8. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 7, characterized in that, The step of generating an adjacency-consistent record based on the same-direction out-of-bounds cluster record, boundary sharing marker, and adjacent partition direction record, and generating a partition deviation source record based on the dual-reference segment record and the adjacency-consistent record, includes: Based on the partition boundary records, determine the adjacent partitions that are adjacent to the partition to be evaluated, and generate the adjacent partition direction records based on the adjacent partition window records; Adjacent consistency records are generated based on the same-direction boundary crossing cluster records, the boundary sharing markers, and the adjacent partition direction records; wherein, the temperature-biased same-direction clusters of the partition to be evaluated are compared only with the temperature-biased direction records of the adjacent partitions, the temperature-biased same-direction clusters of the partition to be evaluated are compared only with the temperature-biased direction records of the adjacent partitions, the humidity-biased same-direction clusters of the partition to be evaluated are compared only with the humidity-biased direction records of the adjacent partitions, and the humidity-biased same-direction clusters of the partition to be evaluated are compared only with the humidity-biased direction records of the adjacent partitions. When there are dual-reference co-directional segments in the dual-reference segmented record and the adjacent consistent record does not contain co-directional records of adjacent partitions, a continuous offset record for this area is generated. When there are dual-reference co-directional segments in the dual-reference segment record and the adjacency consistent record contains co-directional records of adjacent partitions, an adjacency propagation offset record is generated. When a single reference difference segment or reference conflict segment exists in the dual-reference segment record, and a dual-reference co-directional segment is not formed within the same detection quantity type and the same location segment, a reference source mismatch record is generated.

9. The cleanroom temperature and humidity data processing method based on zonal comparison according to claim 8, characterized in that, The step of generating a partition comparison report based on the partition deviation source record, sampling verification record, and partition window record includes: A partitioning result record is generated based on the partitioning deviation source record, the sampling verification record, and the partitioning window record; wherein, the partitioning result record includes the partition identifier to be evaluated, the current reference partition identifier, the historical reference window identifier, the main temperature direction, the main humidity direction, the start and end positions of the same-direction out-of-bounds clusters, the identifier of the participating sampling points, the number of participating sampling points, the boundary sharing participation marker, and the partitioning deviation source; Based on the partitioning results, a partitioning handling prompt record is generated; among them, a continuous offset record in this area corresponds to a local area verification prompt, an adjacent propagation offset record corresponds to an adjacent joint verification prompt, a reference source mismatch record corresponds to a reference review prompt, and a sampling review record corresponds to a sampling review prompt. The partition comparison report is generated based on the partition result record and the partition handling prompt record.

10. A cleanroom temperature and humidity data processing system based on zone comparison, characterized in that, It includes a data alignment module, a partition window module, a dual-reference processing module, a co-directional cluster identification module, a deviation source judgment module, and a report generation module; The data alignment module is used to generate aligned sampling records based on temperature and humidity sampling data and a unified time grid; wherein, the temperature and humidity sampling data includes sampling point identifier, sampling time, temperature sampling value and humidity sampling value; The partition window module is used to generate a sampling point partition mapping record based on the partition boundary record, the sampling point spatial location record, and the partition level record, and to generate a partition window record based on the aligned sampling record and the sampling point partition mapping record; The dual-reference processing module is used to generate current reference boundary records and historical reference boundary records based on current partition window records and historical partition window records, generate dual-reference direction records based on the partition window records to be evaluated, the current reference boundary records and the historical reference boundary records, and generate dual-reference segment records based on the dual-reference direction records; The same-direction cluster identification module is used to generate same-direction out-of-bounds cluster records and sampling verification records based on the dual-reference segmentation records; The deviation source determination module is used to generate an adjacency consistent record based on the same-direction out-of-bounds cluster record, boundary sharing marker and adjacent partition direction record, and to generate a partition deviation source record based on the dual reference segment record and the adjacency consistent record; The report generation module is used to generate a partition result record based on the partition deviation source record, the sampling verification record, and the partition window record; generate a partition handling prompt record based on the partition result record; and generate a partition comparison report based on the partition result record and the partition handling prompt record.

Images