Clean laboratory ventilation energy consumption optimization method based on multi-parameter fusion

Abstract

The invention discloses a clean laboratory ventilation energy consumption optimization method based on multi-parameter fusion, particularly relates to the field of clean laboratory intelligent ventilation regulation and control, and is used for solving the problems of air quantity redundancy and energy consumption waste caused by the fact that an existing ventilation system fixes a static mode of initial debugging parameters. According to the method, an operation data set is constructed by collecting a room pressure difference time sequence, a ventilation cabinet opening degree time sequence and an energy consumption performance indication quantity, safe and stable samples are screened in the operation data set, an air supply and exhaust setting table is generated according to ventilation cabinet opening degree combination induction, and the ventilation cabinet opening degree is subjected to dynamic mining to achieve self-adaptive refining. Reading a current state to calculate target air supply and exhaust control quantity in real-time operation, recording data calculation parameters after execution under the same opening degree combination, and updating an air supply and exhaust setting table.

Description

Technical Field

[0001] This invention relates to the field of intelligent ventilation control in clean laboratories, and more specifically, to a method for optimizing ventilation energy consumption in clean laboratories based on multi-parameter fusion. Background Technology

[0002] Cleanrooms typically house multiple fume hoods, air supply terminals, and other localized exhaust systems. The fume hoods must protect operators from harmful gases and particulate matter while simultaneously drawing away large amounts of air treated by the cleanroom's air conditioning system. Existing technologies utilize parallel control of multiple fume hoods to adjust the total exhaust and make-up air volumes based on the door opening height. A make-up air system within the cleanroom introduces treated fresh air into the exhaust ductwork, creating an air curtain at the hood opening to reduce the overall air intake of the cleanroom. This type of solution requires initial commissioning based on face velocity, pressure differential, and process requirements. This involves setting valve openings, frequency converter operating ranges, and the corresponding relationships between make-up and exhaust air. The system then operates under these settings long-term, with a monitoring interface recording operational data such as pressure differential, temperature and humidity, fan frequency, and fume hood status for routine inspections and troubleshooting.

[0003] However, although the ventilation system has accumulated a large amount of data related to ventilation operation, the existing control methods mainly use this data for display and alarm triggering. The control logic for exhaust and makeup air of fume hoods still follows the empirical parameters and fixed formulas formed during the initial commissioning. It neither automatically judges whether the safety margin is too large based on the stability of pressure difference and face velocity during long-term operation, nor can it automatically tighten to the minimum necessary supply and exhaust air combination under the premise of meeting cleanliness and protection requirements by combining historical energy consumption curves and fume hood usage patterns. As a result, the system maintains an operating state with high air volume and high energy consumption for a long time. There is a lack of a clean laboratory ventilation energy consumption optimization method through multi-parameter fusion and data processing to comprehensively utilize these historical and real-time data.

[0004] To address the aforementioned problems, a technical solution is provided. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method for optimizing ventilation energy consumption in clean laboratories based on multi-parameter fusion. This method constructs an operational dataset by collecting time series data of room pressure difference, time series data of fume hood opening, and energy consumption performance indicators. Safe and stable samples are selected from the operational dataset. A supply and exhaust air setting table is generated by summarizing the fume hood opening combinations. The target supply and exhaust air control quantity is calculated by reading the current state during real-time operation. Furthermore, the method records the calculated parameters after execution and updates the supply and exhaust air setting table under the same opening combination. This approach aims to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: S1: Collect time series data of room pressure difference, time series data of fume hood opening, and energy consumption performance indicators to construct an operational dataset; S2: Identify records in the running dataset where the room pressure difference fluctuation is within a stable range and there are no related alarm markers, and mark the records that meet the conditions as safe and stable samples; S3: Based on safe and stable samples, summarize the correspondence between room pressure difference and energy consumption performance according to the combination of fume hood opening degree, and generate a supply and exhaust air setting table describing the supply and exhaust air control quantity of each opening degree combination. S4: Read the current fume hood opening and room pressure difference during real-time operation, select an item close to the current state according to the supply and exhaust air setting table, and calculate the target supply and exhaust air control quantity used to drive the supply and exhaust air equipment. S5: Record the room pressure difference and energy consumption performance after the target supply and exhaust air control is executed under the same fume hood opening combination. Combine the safe and stable samples and the historical supply and exhaust air adjustment records of the operation data to perform a learning-type comprehensive analysis. When the analysis results indicate that the air volume can be reduced and the room pressure difference is kept up to standard, update the supply and exhaust air control in the supply and exhaust air setting table according to the analysis results.

[0007] Furthermore, step S1 includes the following: The instantaneous values ​​of room pressure difference, fume hood opening percentage, energy consumption performance indicators, and supply and exhaust air control quantities are read at a fixed sampling period and associated with timestamps. The fume hood opening percentage is discretized and then sequentially concatenated to generate a fume hood opening combination code. The code is then appended to the structured storage medium in chronological order to construct the running dataset.

[0008] Furthermore, step S2 includes the following: For the running dataset, check whether the room pressure difference is within the design allowable range and whether the energy consumption performance indicator is lower than the preset abnormal upper limit to generate an alarm flag field. Scan and divide the continuous normal flag sequence according to the sampling time sequence as candidate stable intervals. Calculate the difference between the maximum and minimum pressure difference values ​​from the interval room pressure difference value sequence as the pressure difference fluctuation amplitude. If it is lower than the preset stable threshold, it is determined to be a stable interval. Add safe and stable sample flags to the stable interval records and summarize them into a safe and stable sample set.

[0009] Furthermore, step S3 includes the following: The safe and stable sample set is grouped into sample groups according to the fume hood opening combination code. The room pressure difference, energy consumption performance indication, and supply and exhaust air control quantity are read from each sample group. The sample records are sorted and selected according to the energy consumption performance indication, and the priority sample record subset with pressure difference far away from the design lower limit is selected. The representative value of supply and exhaust air control quantity is calculated from the priority sample record subset as the basic supply and exhaust air control quantity, and the reference pressure difference and reference energy consumption are recorded. The supply and exhaust air setting table is constructed by creating a supply and exhaust air setting table entry for each fume hood opening combination, which includes the fume hood opening combination code, the basic supply and exhaust air control quantity, the reference pressure difference, and the reference energy consumption.

[0010] Furthermore, step S4 includes the following: Read the current fume hood opening and current room pressure difference from the new monitoring data frame and generate the current fume hood opening combination code. Query the supply and exhaust air setting table to find the matching entry and extract the basic supply and exhaust air control quantity as the target supply and exhaust air control quantity.

[0011] Furthermore, step S4 also includes the following: If no matching entry is found, the sum of the absolute differences between the current fume hood opening combination code and the corresponding cabinet door level of each fume hood opening combination code in the supply and exhaust air setting table is calculated as the difference degree. The entry with the smallest difference degree is selected as the approximate entry, and the basic supply and exhaust air control quantity is extracted from it as the initial target supply and exhaust air control quantity. The deviation between the current room pressure difference and the reference pressure difference of the approximate entry is calculated to correct the initial target supply and exhaust air control quantity. The final target supply and exhaust air control quantity is output to the supply air equipment and exhaust air equipment as control commands.

[0012] Furthermore, step S5 includes the following: Record the time series of room pressure difference and energy consumption performance indicators after the target supply and exhaust air control is executed under the same fume hood opening combination. Extract records from the safe and stable sample set to calculate the safe and stable average pressure difference and the safe and stable minimum pressure difference. Calculate the new average pressure difference and the new minimum pressure difference for the room pressure difference time series. Weighted combination to obtain the combined average pressure difference and take the minimum as the combined minimum pressure difference. Calculate the difference and divide it by the average to obtain the pressure difference redundancy buffer ratio trajectory parameter.

[0013] Furthermore, step S5 also includes the following: Retrieve historical supply and exhaust air adjustment records from the running dataset to calculate the energy consumption response ratio under the condition of meeting the standard reduction. After sorting and excluding extreme values, calculate the representative value as the energy consumption ventilation response slope profile parameter.

[0014] Furthermore, step S5 also includes the following: Using historical supply and exhaust air adjustment results as a sample set, the system divides the experience blocks in a two-dimensional parameter plane to statistically analyze the success rate and recommended adjustment range, and combines these to obtain the supply and exhaust air adjustment coefficient. If the room pressure difference remains within the standard and the supply and exhaust air adjustment coefficient is greater than the preset minimum effective adjustment threshold, then the basic supply and exhaust air control quantity in the supply and exhaust air setting table is corrected using the supply and exhaust air adjustment coefficient as a proportional factor.

[0015] The technical effects and advantages of this invention, based on a multi-parameter fusion-based method for optimizing ventilation energy consumption in clean laboratories, are as follows: This invention constructs an operational dataset by collecting time series data of room pressure difference, time series data of fume hood opening, and energy consumption performance indicators. It then selects safe and stable samples to generate a supply and exhaust air setting table. During real-time operation, it calculates and executes the target supply and exhaust air control quantities. By fusing current records with historical samples, it calculates the pressure difference redundancy buffer ratio trajectory parameters and the energy consumption ventilation response slope profile parameters. Through analysis of historical adjustment results, it obtains the adjustment coefficient update setting table, thus achieving complete closed-loop optimization through dynamic fusion of multiple parameters.

[0016] First, this method significantly reduces the overall energy consumption of cleanroom ventilation systems. While meeting differential pressure and protection requirements, it automatically reduces airflow to the minimum necessary combination, avoiding the long-term high operating levels caused by traditional static commissioning. This adaptive adjustment based on real-time data far exceeds expectations and cannot be predicted by traditional empirical parameters. Second, it improves system stability and safety. Through quantitative analysis of differential pressure redundancy and energy consumption response, it ensures that updates only occur when redundancy is sufficient, preventing potential fluctuation risks and achieving a precise balance between protection and energy saving. The synergistic effect of this parameter fusion mechanism provides non-obvious long-term reliable operation. Third, it promotes the comprehensive utilization of data resources, transforming accumulated historical and real-time multi-type data into operable control strategies, gradually approaching the optimization boundary, leading to a continuous decrease in laboratory operating costs and improved environmental adaptability. This globally coordinated nonlinear optimization path surpasses the linear improvement of conventional methods. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the energy consumption optimization method for cleanroom ventilation based on multi-parameter fusion, as described in this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example 1: Figure 1This invention presents a method for optimizing ventilation energy consumption in clean laboratories based on multi-parameter fusion, including: S1: Collect time series of room pressure difference, time series of fume hood opening, and energy consumption indicators to construct an operational dataset.

[0020] S2: Identify records in the running dataset where the room pressure difference fluctuation is within a stable range and there are no related alarm markers, and mark the records that meet the conditions as safe and stable samples.

[0021] S3: Based on safe and stable samples, summarize the correspondence between room pressure difference and energy consumption performance according to the combination of fume hood opening degree, and generate a supply and exhaust air setting table describing the supply and exhaust air control quantity of each opening degree combination.

[0022] S4: During real-time operation, read the current fume hood opening and room pressure difference, select an item close to the current state according to the supply and exhaust air setting table, and calculate the target supply and exhaust air control quantity used to drive the supply and exhaust air equipment.

[0023] S5: Record the room pressure difference and energy consumption performance after the target supply and exhaust air control is executed under the same fume hood opening combination. Combine the safe and stable samples and the historical supply and exhaust air adjustment records of the operation data to perform a learning-type comprehensive analysis. When the analysis results indicate that the air volume can be reduced and the room pressure difference is kept up to standard, update the supply and exhaust air control in the supply and exhaust air setting table according to the analysis results.

[0024] In cleanroom ventilation systems, existing control strategies are often limited to the static application of initial commissioning parameters, resulting in the system operating at high airflow and high energy consumption for extended periods, failing to fully utilize accumulated operational data for dynamic optimization. By systematically collecting multi-parameter time series data and constructing an operational dataset, a reliable foundation can be provided for subsequent screening of safe and stable samples and generation of supply and exhaust air setting tables, thereby enabling adaptive adjustment of ventilation energy consumption based on data fusion. Step S1 focuses on real-time sampling and data structuring, ensuring that all key quantities such as room pressure difference, fume hood opening combination codes, energy consumption performance indicators, and supply and exhaust air control quantities are stored in a unified format, supporting direct referencing and analysis in subsequent steps.

[0025] S1.1 Data sampling and instantaneous value reading.

[0026] To monitor the cleanroom's operational status, sampling is performed at fixed intervals. At each sampling moment, instantaneous values ​​of the room's differential pressure measurement points are read from the monitoring link. These instantaneous values, output in real-time by differential pressure sensors, represent the pressure difference between the room and the external environment. Simultaneously, the position codes or opening percentages of each fume hood door are read. These opening percentages are calculated using position sensors to determine the ratio of the door height to its fully open height. Furthermore, energy consumption indicators representing fan or overall energy consumption are read. These indicators are extracted from fan power meters or electricity meters, reflecting the immediate energy consumption level. These read values ​​are correlated with the current sampling timestamp to form preliminary time-correlation data. To support subsequent optimization, each preliminary time-correlation data entry includes corresponding supply and exhaust air control parameters. These parameters include normalized command values ​​for supply fan frequency, exhaust fan frequency, or valve opening, extracted directly from the control system log. The reading operations must be performed synchronously to maintain time consistency. If sensor delays exist, adjustments are made through a timestamp calibration mechanism to ensure that the preliminary time-correlation data accurately captures the instantaneous state of the ventilation system. The initial time-related data set is ready to serve as input for subsequent encoding processing.

[0027] S1.2 Fume Hood Opening Combination Code Generation.

[0028] Based on the preliminary time-related data generated in sub-step S1.1, the opening percentage of each fume hood is discretized, mapping continuous percentage values ​​to preset level intervals. These intervals are divided according to fume hood operating specifications, such as dividing the opening percentage into multiple equally spaced or non-equally spaced levels. Each level corresponds to a unique identifier, such as a number or letter, to reflect the stepwise impact of opening on exhaust air. Subsequently, the discretized opening levels of multiple fume hoods are connected in a fixed order, following the installation serial number or identification number of the fume hood, forming a fume hood opening combination code. This code can be in string form, such as a concatenation of level identifiers, or in integer vector form, such as a concatenation of level numbers, ensuring the uniqueness and comparability of the code. For example, if a laboratory has three fume hoods, each with a specific level identifier, the code is an ordered combination of these identifiers. This fume hood opening combination code is embedded in the preliminary time-related data, replacing the original scattered opening percentage values ​​to form a coded time record. Through this generation process, the coded time record has a standardized combination representation, facilitating centralized grouping and comparison of the dataset. The encoding time record is fully formed, supporting the construction of running datasets.

[0029] S1.3 Runs dataset construction and storage.

[0030] The encoded time records from sub-step S1.2 are received and appended to a unified structured storage medium in the order of sampling timestamps. This medium can be a database table or a serialized file. The append operation only expands existing records without overwriting them. Each record in the running dataset contains the sampling time, room pressure difference, fume hood opening combination code, energy consumption performance indicator, and the corresponding supply and exhaust air control quantity. The sampling time is in timestamp format, the room pressure difference is recorded in pressure units, the energy consumption performance indicator is recorded in power or energy consumption units, and the supply and exhaust air control quantity is recorded as a normalized command value. To improve access efficiency, indexes are created in the storage for the fume hood opening combination code and sampling time fields, supporting fast queries by combination or time range. This running dataset is organized in a time-series structure, integrating all ventilation-related key quantities to form the data foundation for selecting safe and stable samples in the subsequent step S2. The running dataset is fully established, providing complete data support for the optimization method.

[0031] By implementing step S1, the cleanroom ventilation system achieves real-time acquisition and structured storage of multiple parameters, ensuring that the data provides a traceable time-series basis for energy consumption optimization while meeting cleanroom protection requirements.

[0032] The system reads instantaneous values ​​of room pressure difference, fume hood opening percentages, and energy consumption indicators from the monitoring link at a fixed sampling period, along with corresponding supply and exhaust air control values ​​at the same time. This data is then correlated with the sampling timestamps to form preliminary time-related data. The opening percentages of each fume hood are discretized, and the discretized opening levels are connected in a fixed order to generate fume hood opening combination codes. These codes are embedded in the preliminary time-related data to form coded time records. Finally, the data is appended to the structured storage medium in the order of the sampling timestamps to construct a running dataset. The running dataset records include the sampling time, room pressure difference, fume hood opening combination codes, energy consumption indicators, and supply and exhaust air control values.

[0033] While the running dataset provides multi-parameter time series data, it contains potentially outlier records that could lead to optimization biases. By rigorously screening and identifying stable operating conditions, high-quality, safe, and stable samples can be constructed, supporting the accurate generation of supply and exhaust ventilation setting tables, thereby achieving adaptive reduction of ventilation energy consumption. Step S2 focuses on filtering records in the running dataset, ensuring that only the portions with stable differential pressure and no alarms are extracted, forming the data foundation for the subsequent group analysis in step S3.

[0034] S2.1 Alarm tag field generation.

[0035] For each record in the operational dataset, the room pressure differential is checked to ensure it is within the design limits. This range is preset based on laboratory cleanliness standards and process requirements and can be dynamically adjusted based on historical operation data. Subsequently, the energy consumption indicator is checked to ensure it is below the preset abnormal upper limit. If the room pressure differential exceeds the range or the energy consumption indicator exceeds the upper limit, the alarm flag field is set to abnormal; if both meet the requirements, it is set to normal. This process iterates through all records in the operational dataset to form an extended operational dataset, where each record has an additional alarm flag field. This extended operational dataset provides anomaly identification, facilitating subsequent segmentation. Once the extended operational dataset is established, it serves as input for time segmentation processing.

[0036] S2.2 Time segmentation and candidate stable interval division.

[0037] Based on the extended running dataset from sub-step S2.1, records are scanned sequentially according to sampling time to check if all alarm markers in the continuous sequence are normal. Starting from the first record, if the current record's alarm marker is normal, it is included in the current sequence; otherwise, the current sequence is terminated and a new sequence is started. This division considers timestamp continuity; if the sampling interval varies, continuity is determined by the timestamp difference, ensuring that there are no abnormal records within the candidate stable intervals. Each candidate stable interval records the start timestamp, end timestamp, and an index list of included records, forming a candidate stable interval list. Through this processing, the extended running dataset is differentiated into an ordered interval structure, supporting targeted evaluation. The candidate stable interval list is fully generated, providing the basis for calculating fluctuation amplitude.

[0038] S2.3 Calculation of differential pressure fluctuation amplitude and determination of stable range.

[0039] For each candidate stable interval, the sequence of recorded room differential pressure values ​​within the interval is extracted from the extended runtime dataset. First, a smoothing filter is applied to remove sensor noise. Then, the maximum and minimum differential pressure values ​​in the sequence are calculated, and the difference is taken as the differential pressure fluctuation amplitude. If the differential pressure fluctuation amplitude is lower than a preset stability threshold (set according to laboratory differential pressure specifications and historical stable data), the interval is determined to be within a stable range; otherwise, it is discarded. This process iterates through all candidate stable intervals, forming a stable interval list, with each stable interval retaining its original index and time information. Through this calculation, candidate stable intervals are selected as stable intervals. The stable interval list is then established for sample labeling.

[0040] S2.4 Safety and Stable Sample Labels and Set Composition.

[0041] The system receives the list of stable intervals from sub-step S2.3 and adds a safe and stable sample marker to the records within each stable interval. This marker is a dedicated field indicating that the record meets the conditions of no alarms and low fluctuations. All marked records are aggregated into a safe and stable sample set, which is stored in an independent structure, such as a database view or file, and associated with the running dataset for querying, ensuring no duplicate records. This set excludes any records with abnormal pressure differences or energy consumption, reflecting the system's state under clean and reasonable energy consumption conditions. Through this marker, the safe and stable sample set is formed, supporting the grouping in step S3. The comprehensive safe and stable sample set provides a reliable data source for the optimization method.

[0042] By implementing step S2, the cleanroom ventilation system extracts safe and stable samples from the operational dataset, ensuring the rigor of data filtering, laying the foundation for subsequent steps of energy consumption optimization, and achieving stable and efficient ventilation control.

[0043] Through step S2, the set of safe and stable samples has been screened, providing stable operating data without anomalies. By grouping and summarizing these samples and extracting corresponding relationships, a supply and exhaust air setting table can be constructed to correlate fume hood opening combinations with recommended control quantities, thereby guiding real-time airflow adjustments and ensuring the system achieves minimum energy consumption while meeting protection standards. Step S3 focuses on data summarization and table generation, ensuring that historical stable operating conditions are transformed into operable control references.

[0044] S3.1 Grouping of safe and stable samples.

[0045] For the safe and stable sample set, records are categorized according to the fume hood opening combination code in each record. Records with the same fume hood opening combination code are assigned to the same sample group, and each sample group corresponds to a unique fume hood opening combination. If an empty sample group is encountered or the number of samples is less than the preset minimum value, it is marked as pending addition and not processed temporarily. This grouping traverses all records in the safe and stable sample set, using the fume hood opening combination code as the classification key to form a sample group list. Each sample group contains the record's sampling time, room pressure difference, energy consumption performance indicators, and supply and exhaust air control parameters. Through this classification, the safe and stable sample set is organized into a grouped structure, supporting independent analysis. The complete sample group list serves as input for data retrieval.

[0046] S3.2 Sample group data reading and priority sample search.

[0047] Based on the sample group list in sub-step S3.1, for each sample group, read all recorded room differential pressure, energy consumption performance indicators, and supply and exhaust ventilation control parameters. Sort the sample group records in ascending order of energy consumption performance indicators. Identify records in the list whose differential pressure values ​​are far from the design lower limit, which is derived from laboratory cleanliness standards and initial commissioning parameters. The degree of deviation is quantified by the ratio of the differential pressure value to the lower limit. Select the record with the lowest energy consumption performance indicator from these records until the preferred sample record subset reaches a preset proportion of the sample group size, based on historical data coverage. This search covers all sample groups, ensuring that the preferred sample record subset represents a low-consumption, stable state. Through this reading and search, the sample group obtains a preferred sample record subset. The preferred sample record subset is now ready, providing the basis for calculations.

[0048] S3.3 Calculation and reference value recording of basic air supply and exhaust control quantities.

[0049] The priority sample record subset from sub-step S3.2 is received. For each priority sample record subset, the supply and exhaust air control quantity sequence is extracted, and the median value of the sequence is calculated using the median algorithm as the basic supply and exhaust air control quantity. If the size of the priority sample record subset is less than a preset threshold, the calculation is expanded to adjacent samples. The room pressure difference of the record with the smallest supply and exhaust air control quantity is selected from the priority sample record subset as the reference pressure difference, and its energy consumption performance indicator is used as the reference energy consumption. This calculation and recording traverses all sample groups to ensure that the values ​​reflect the optimization potential. Through this process, each fume hood opening combination is associated with the basic supply and exhaust air control quantity, the reference pressure difference, and the reference energy consumption. These values ​​are established completely for entry creation.

[0050] S3.4 Establishment of supply and exhaust ventilation settings table entries and table construction.

[0051] For each fume hood opening combination, a supply and exhaust air setting table entry is generated. This entry includes the fume hood opening combination code, basic supply and exhaust air control parameters, reference differential pressure, and reference energy consumption. All fume hood opening combinations corresponding to the sample group list are traversed, and each entry is added to the supply and exhaust air setting table. This table uses a database table or key-value storage format, with the fume hood opening combination code as the primary key, supporting fast retrieval and update interfaces. This construction ensures the uniqueness of entries and integrates typical environmental responses. Through this construction, the supply and exhaust air setting table is fully constructed. Once the supply and exhaust air setting table is constructed, it supports the query application in step S4.

[0052] Through the implementation of the above sub-step S3, the clean laboratory ventilation system extracts the supply and exhaust air setting table from the safe and stable sample, realizes multi-parameter fusion control optimization, and provides a data-driven framework for energy consumption reduction.

[0053] The supply and exhaust air setting table contains recommended control values ​​for each combination of fume hood openings. By reading the current status in real time and matching it with the setting table, the target supply and exhaust air control values ​​can be obtained, enabling dynamic adjustment of equipment drives and ensuring that the system balances cleanliness protection and energy consumption optimization during actual operation. This step S4 focuses on real-time processing and output, ensuring that the control logic adapts to varying operating conditions.

[0054] S4.1 Current fume hood opening and room pressure difference reading and encoding generation.

[0055] For new monitoring data frames, the current opening degree of each fume hood is read, obtained from the proportion of the hood door height from the position sensor. If there is sensor delay, synchronization calibration is performed using timestamps. The current fume hood opening degree is then discretized, mapping the proportion to a preset level range. This range is divided at equal intervals according to fume hood exhaust specifications, with each level assigned a numerical identifier. The discretized levels of multiple fume hoods are connected sequentially by installation number to form a combined code for the current fume hood opening degree; this code is an integer vector. Simultaneously, the current room differential pressure is read, directly output from the differential pressure sensor. Through this process, the combined code for the current fume hood opening degree and the current room differential pressure form a real-time status description. This complete real-time status description serves as input for lookups.

[0056] Search for identical entries in the S4.2 supply and exhaust air setting table.

[0057] Based on the real-time status description in sub-step S4.1, a query is performed in the supply and exhaust air setting table using the current fume hood opening combination code as the key. If a completely matching entry exists, the basic supply and exhaust air control value for that entry is extracted as the target supply and exhaust air control value. This query uses a hash index to ensure constant-time retrieval. If no match is found, the query fails and the process proceeds to difference calculation. If a match is successful, the target supply and exhaust air control value is initially obtained; otherwise, preparations are made for entry selection.

[0058] S4.3 Selection of close items and calculation of differences.

[0059] The query result from sub-step S4.2 is received. If no duplicate entries are found, all entries in the supply and exhaust ventilation setting table are iterated through. The absolute difference between the current fume hood opening combination code and the corresponding door level of each entry's fume hood opening combination code is calculated, and the difference is summed in order of door number as the difference degree. The entry with the smallest difference degree is selected as the closest entry, and the basic supply and exhaust ventilation control quantity is extracted from it as the initial target supply and exhaust ventilation control quantity. This calculation ensures global optimization. The initial target supply and exhaust ventilation control quantity is determined, providing a basis for correction.

[0060] S4.4 Target supply and exhaust air control quantity correction and output.

[0061] For the initial target supply and exhaust air control value in sub-step S4.3, calculate the deviation between the current room differential pressure and the reference differential pressure of the adjacent item; this deviation is the difference between the two. Correct the initial target supply and exhaust air control value using a scaling algorithm. This algorithm multiplies the deviation by a preset sensitivity factor, which is set based on historical differential pressure responses, and adds it to the initial value. After correction, verify whether the target supply and exhaust air control value is within the equipment's allowable range; if it exceeds this range, revert to the initial value. Output the final target supply and exhaust air control value to the supply and exhaust air equipment as control commands to drive fan frequency or valve opening. Control adjustment is executed, and subsequent recording is supported.

[0062] Step S4 is responsible for calculating and outputting the target supply and exhaust air control values ​​using a pre-generated supply and exhaust air setting table during actual system operation. This drives the adjustment of the supply and exhaust air equipment, achieving dynamic response and minimizing energy consumption of the ventilation system. First, the current fume hood opening and current room pressure difference are read from the new monitoring data frame, and a combination code for the current fume hood opening is generated to form a real-time status description. Then, a perfectly matching entry is searched in the supply and exhaust air setting table. If a match is found, the basic supply and exhaust air control value is directly extracted as the target. If no match is found, the difference is calculated, a close entry is selected, and its basic supply and exhaust air control value is extracted as the initial value. Finally, corrections are made based on the deviation between the current room pressure difference and the reference pressure difference, and the final target supply and exhaust air control value is output as a control command. This process ensures the adaptability of the control logic, enabling timely adjustment of airflow when the fume hood opening changes or the pressure difference fluctuates, maintaining the room pressure difference within the cleanliness standard range, while avoiding energy waste caused by excessive supply and exhaust air.

[0063] In cleanroom ventilation systems, new operational data is generated after the target supply and exhaust air volume control is executed. By fusing these data with safe and stable samples to calculate parameters and obtaining adjustment coefficients through comprehensive analysis, the supply and exhaust air setpoints can be adaptively updated. This ensures that the system gradually reduces the air volume to the minimum necessary level while strictly maintaining the differential pressure to avoid the long-term continuation of high energy consumption. Step S5 focuses on the self-learning optimization process, supporting dynamic tightening of ventilation energy consumption.

[0064] S5.1 Differential Pressure Redundancy Buffer Proportional Trajectory Parameter Calculation.

[0065] For the current fume hood opening combination, all records belonging to this combination are extracted from the safe and stable sample set to form a safe and stable differential pressure sample set. For the room differential pressure values ​​in the safe and stable differential pressure sample set, the sample average is calculated as the safe and stable differential pressure average, and the sample minimum is calculated as the safe and stable differential pressure minimum. Simultaneously, for the room differential pressure time series recorded after implementing the target supply and exhaust air control, the average is calculated as the new differential pressure average, and the minimum is calculated as the new differential pressure minimum. The safe and stable differential pressure average and the new differential pressure average are then weighted and combined. This weight is based on a preset confidence level of historical stable data, and the formula is as follows: ,in This represents the average value of the combined pressure difference. To ensure a safe and stable average pressure difference, For the new average pressure difference, The weighting coefficient is calculated by dividing the number of records in the safe and stable pressure differential sample set by the sum of the number of records in the safe and stable pressure differential sample set and the number of current room pressure differential time series points. This ensures a balance between the confidence of historical data and the freshness of new data. The smaller of the safe and stable minimum pressure differential and the new minimum pressure differential is taken as the combined minimum pressure differential. The difference between the combined average pressure differential and the combined minimum pressure differential is calculated and divided by the combined average pressure differential, using the following formula: ,in The parameters for the differential pressure redundancy buffer ratio trajectory are... This represents the minimum combined differential pressure. This parameter is dimensionless and reflects the buffer ratio of the differential pressure relative to the safe minimum value under the current control level. The differential pressure redundancy buffer ratio trajectory parameter has been calculated and is used as input for analysis.

[0066] S5.2 Calculation of energy consumption ventilation response slope profile parameters.

[0067] Based on the current fume hood opening combination, all relevant historical supply and exhaust air adjustment records in the operational dataset are retrieved. Each record includes the supply and exhaust air control values ​​before and after the adjustment, energy consumption performance indicators, and differential pressure compliance markers. For each historical supply and exhaust air adjustment record, if the differential pressure remains within the compliance range and the supply and exhaust air control values ​​are reduced, the reduction in energy consumption performance indicators is divided by the reduction in supply and exhaust air control values, which is used as the energy consumption response ratio. The formula is as follows: ,in This is the energy response ratio. The amount of reduction in the energy consumption indicator. This parameter represents the reduction in supply and exhaust ventilation control. All energy consumption response ratios meeting the criteria are collected, sorted by value, and extreme values ​​(both excessively high and low) are excluded. A representative value is calculated from the remaining intermediate range using the median algorithm, serving as the energy consumption ventilation response slope profile parameter. The energy consumption change and control quantity change after the current execution are incorporated into the set, and this parameter is updated. This parameter describes the typical energy consumption improvement resulting from each unit reduction in supply and exhaust ventilation control. The energy consumption ventilation response slope profile parameter is generated completely, providing a basis for comprehensive analysis.

[0068] The supply and exhaust air adjustment coefficients are obtained from the comprehensive analysis in S5.3.

[0069] Receive the parameters from sub-steps S5.1 and S5.2, use the historical supply and exhaust air adjustment results in the running dataset as the sample set, take each historical supply and exhaust air adjustment as a sample point, calculate its differential pressure redundancy buffer ratio trajectory parameter and energy consumption ventilation response slope profile parameter, mark the adjustment as successful or unsuccessful according to the differential pressure meeting the standard and the energy consumption reduction after adjustment, and record the actual reduction ratio of supply and exhaust air control quantity.

[0070] On the two-dimensional parameter plane, the parameter space is divided into a predetermined number of discrete empirical blocks. This number is optimized based on historical sample distribution. Each empirical block corresponds to a range of parameters for the differential pressure redundancy buffer ratio trajectory and a range of parameters for the energy consumption ventilation response slope profile. For each empirical block, the number of successful adjustments falling into the sample points and the total number are counted, and the success rate is calculated. A representative value of the actual reduction rate is obtained from the successful samples using an averaging algorithm, which serves as the recommended adjustment range. For the current parameter value, the corresponding empirical block is located, and the success rate and recommended adjustment range are read. The product of these two values ​​is used as the supply and exhaust air adjustment coefficient. After this sub-step is completed, the supply and exhaust air adjustment coefficient is established, supporting correction operations.

[0071] S5.4 Correct the supply and exhaust air setting table based on the supply and exhaust air adjustment coefficient.

[0072] Using the supply and exhaust air adjustment coefficients in sub-step S5.3, determine whether the current room pressure difference meets the cleanroom pressure difference requirements. These requirements are based on design standards. If the requirements are met and the supply and exhaust air adjustment coefficients are greater than the preset minimum effective adjustment threshold, then the basic supply and exhaust air control quantities for the current fume hood opening combination in the supply and exhaust air setting table are corrected using the supply and exhaust air adjustment coefficients as a proportional factor. The formula is as follows: ,in Control the supply and exhaust air volume for the new foundation. This is the original basic supply and exhaust air control volume. Adjust the supply and exhaust ventilation coefficients. Write the new baseline supply and exhaust ventilation control values ​​into the corresponding entries of the supply and exhaust ventilation setting table, replacing the original values, and verify the stability of the differential pressure after the update. Through continuous execution, the supply and exhaust ventilation setting table is gradually adjusted to the lowest control level. After this sub-step is completed, the supply and exhaust ventilation setting table is updated, achieving energy consumption optimization.

[0073] Step S5, after executing the target supply and exhaust air control, records the current room pressure difference time series and energy consumption performance indicator time series, and integrates them with a safe and stable sample set to calculate the pressure difference redundancy buffer ratio trajectory parameters and energy consumption ventilation response slope profile parameters. Then, based on the historical supply and exhaust air adjustment results concentrated in the operational data, a comprehensive analysis is performed to obtain the supply and exhaust air adjustment coefficient. When the current room pressure difference meets the standard and the adjustment coefficient is greater than the preset minimum effective adjustment threshold, this coefficient is used to correct the basic supply and exhaust air control quantities in the supply and exhaust air setting table, achieving gradual optimization of the setting table. This step first quantifies the pressure difference redundancy to assess the safety buffer, then refines the energy consumption response slope to measure energy-saving potential, then derives the adjustment coefficient through empirical block division and statistical analysis, and finally executes condition corrections to ensure that the update process both conservatively meets protection requirements and actively reduces airflow. Through continuous iteration, step S5 drives the entire system to evolve from initial static control to data-driven dynamic minimum energy consumption, avoiding long-term high operating conditions and supporting the long-term efficient operation of the clean laboratory.

[0074] Specifically, the above description is only a preferred embodiment of this application and is not intended to limit this application.

[0075] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0076] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A clean laboratory ventilation energy consumption optimization method based on multi-parameter fusion, characterized in that, The method comprises the steps of: S1: Collecting room differential pressure time series, fume hood opening time series and energy consumption performance indicators, and constructing an operation data set; S2: Identifying records in the operation data set in which the room differential pressure fluctuation is within a stable range and has no related alarm markers, and marking the records meeting the conditions as safe and stable samples; S3: Based on the safe and stable samples, the corresponding relationship between the room differential pressure and the energy consumption performance is summarized according to the fume hood opening combination to generate a supply and exhaust setting table describing the supply and exhaust control quantities of each opening combination; S4: In real-time operation, the current fume hood opening and room differential pressure are read, and the target supply and exhaust control quantity is calculated according to the supply and exhaust setting table to select an item close to the current state to drive the supply and exhaust equipment; S5: Under the same fume hood opening combination, the room differential pressure and energy consumption performance after the execution of the target supply and exhaust control quantity are recorded, and a learning type comprehensive analysis is conducted in combination with the safe and stable samples and historical supply and exhaust adjustment records in the operation data set, and when the analysis result indicates that the air volume can be reduced and the room differential pressure remains up to standard, the supply and exhaust control quantity in the supply and exhaust setting table is updated according to the analysis result.

2. The method for optimizing energy consumption of clean laboratory ventilation based on multi-parameter fusion according to claim 1, characterized in that, Step S1 includes the following contents: The room differential pressure instantaneous value, fume hood opening percentage, energy consumption performance indicator and supply and exhaust control quantity are read at a fixed sampling period and associated with a time stamp, the fume hood opening percentage is discretized and then sequentially connected to generate a fume hood opening combination code, and the code is appended in time sequence to a structured storage medium to construct an operation data set. 3.The method of claim 2, wherein, Step S2 includes the following contents: An alarm marker field is generated by checking whether the room differential pressure is within the design allowed range and whether the energy consumption performance indicator is lower than the preset abnormal upper limit for the operation data set records, the continuous normal marker sequence is scanned in sampling time sequence to divide the sequence into candidate stable intervals, the difference between the maximum differential pressure value and the minimum differential pressure value is calculated from the interval room differential pressure value sequence as the differential pressure fluctuation amplitude, and if it is lower than the preset stable threshold, it is determined as a stable interval, the safe and stable sample marker is added to the stable interval record and is collected as a safe and stable sample set.

4. The method for optimizing energy consumption of clean laboratory ventilation based on multi-parameter fusion according to claim 3, characterized in that, Step S3 includes the following contents: The safe and stable sample set is grouped according to the fume hood opening combination code to form sample groups, the room differential pressure, energy consumption performance indicator and supply and exhaust control quantity are read from each sample group, the priority sample record subset with differential pressure away from the design lower limit is selected by sorting the energy consumption performance indicator, the supply and exhaust control quantity representative value is calculated as the basic supply and exhaust control quantity from the priority sample record subset and the reference differential pressure and reference energy consumption are recorded, and the supply and exhaust setting table entry containing the fume hood opening combination code, basic supply and exhaust control quantity, reference differential pressure and reference energy consumption is created for each fume hood opening combination to construct the supply and exhaust setting table.

5. The method for optimizing energy consumption of clean laboratory ventilation based on multi-parameter fusion according to claim 4, characterized in that, Step S4 includes the following contents: The current fume hood opening and current room differential pressure in the new monitoring data frame are read and the current fume hood opening combination code is generated, and the matching entry is queried in the supply and exhaust setting table to extract the basic supply and exhaust control quantity as the target supply and exhaust control quantity.

6. The method for optimizing energy consumption of clean laboratory ventilation based on multi-parameter fusion according to claim 5, characterized in that, Step S4 further includes the following contents: If there is no matching entry, the sum of the absolute difference between the current fume hood opening combination code and the corresponding fume hood door level of each entry in the exhaust setting table is calculated as the difference degree, the entry with the smallest difference degree is selected as the approximate entry, and the basic exhaust control amount is extracted from the approximate entry as the initial target exhaust control amount. The deviation between the current room pressure difference and the reference pressure difference of the approximate entry is corrected to obtain the initial target exhaust control amount, and the final target exhaust control amount is output to the supply air equipment and the exhaust equipment as a control instruction.

7. The method for optimizing energy consumption of clean laboratory ventilation based on multi-parameter fusion according to claim 6, characterized in that, Step S5 Comprises the following contents: The room pressure difference time series and energy consumption performance indicator time series after the execution of the target exhaust control amount are recorded under the same fume hood opening combination, the safe and stable sample set is extracted to calculate the safe and stable pressure difference average value and the safe and stable pressure difference minimum value, the new pressure difference average value and the new pressure difference minimum value are calculated from the room pressure difference time series, the combined pressure difference average value is obtained by weighted combination, and the minimum value is taken as the combined pressure difference minimum value. The difference value is divided by the average value to obtain the pressure difference redundancy buffer proportion trajectory parameter.

8. The method for optimizing energy consumption of clean laboratory ventilation based on multi-parameter fusion according to claim 7, characterized in that, Step S5 also Comprises the following contents: The energy consumption response ratio under the compliance reduction condition is calculated by searching the historical exhaust adjustment record in the running data set, and the representative value is calculated as the energy consumption ventilation response slope profile parameter after sorting and excluding extreme values.

9. The method for optimizing energy consumption of clean laboratory ventilation based on multi-parameter fusion according to claim 8, characterized in that, Step S5 also comprises the following contents: The success proportion and recommended adjustment amplitude are calculated by dividing the two-dimensional parameter plane with the historical exhaust adjustment result as the sample set, and the exhaust adjustment coefficient is obtained by combination. If the room pressure difference remains compliant and the exhaust adjustment coefficient is greater than the preset minimum effective adjustment threshold, the exhaust adjustment coefficient is used as a proportional factor to correct the basic exhaust control amount in the exhaust setting table.

Images