An archive room multi-device heterogeneous system linkage control method

Abstract

The application discloses a file warehouse multi-device heterogeneous system linkage control method, and particularly relates to the technical field of file warehouse intelligent management and control, which classifies file warehouses of each campus, generates differentiated control strategies and corresponding strategy fingerprints; collects environmental parameter data and behavior parameter data, and constructs a coupling data set; in a network interruption state, the strategy fingerprint is matched and verified with the warehouse type, and the differentiated control strategy is reconstructed when the strategy is abnormal or missing; then, the warehouse operation state is determined in combination with the coupling data set, linkage control instructions are generated, and the temperature and humidity boundaries are controlled by constraint through a minimum constraint set; the method can maintain differentiated linkage control of different warehouses in the case of network interruption or strategy abnormality, and reduce the file preservation risk caused by unified default control.

Description

Technical Field

[0001] This invention relates to the field of intelligent management and control technology for archive storage facilities, specifically to a method for coordinated control of a multi-device heterogeneous system in an archive storage facility. Background Technology

[0002] With the continuous improvement of information technology and intelligent management of archives, various types of equipment, such as air conditioning, dehumidification, security, fire protection, and environmental monitoring, have been widely deployed in archive storage facilities. These devices are usually provided by different manufacturers and use multiple communication protocols, forming a heterogeneous system of multiple devices. In the current technology, most archive storage facilities still adopt a decentralized control method, with each system operating independently and lacking a unified data interaction and linkage control mechanism. This makes it difficult to achieve collaborative control of multiple devices in actual operation, resulting in a low overall level of automation and intelligence.

[0003] When implementing differentiated management for different types of archive repositories (such as faculty archives and student archives), existing technologies typically achieve this through preset fixed control parameters or simple rules. However, this approach lacks the ability to comprehensively analyze the attributes and actual operational status of the repositories, making it difficult to dynamically adjust control strategies based on environmental changes and user behavior, resulting in insufficient control precision. Furthermore, in multi-campus or distributed deployment scenarios, the system is highly dependent on the network. If a network outage occurs, cloud-based policies cannot be distributed, and the edge devices often revert to a uniform default control method, failing to maintain the original differentiated control strategies.

[0004] In the event of a network outage or abnormal situation, the lack of effective strategy identification and reconstruction methods, as well as constraint and control mechanisms for key environmental parameters, can easily lead to the mixed use or failure of control strategies for different types of warehouses, resulting in problems such as deviations in temperature and humidity control and reduced security levels. In severe cases, this may affect the long-term safe storage of archives. Summary of the Invention

[0005] The purpose of this invention is to provide a method for the coordinated control of a multi-device heterogeneous system in an archive storage facility, in order to address the shortcomings of the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for linkage control of a multi-device heterogeneous system in an archive storage facility, comprising: The archives of each campus are classified, and the types of archives and corresponding environmental control parameters are obtained to generate differentiated control strategies. The differentiated control strategies are then feature-encoded to form corresponding strategy fingerprints. Collect environmental parameter data and behavioral parameter data from each warehouse, and construct a coupled dataset of environmental parameter domain and behavioral parameter domain; Under normal network conditions, the differentiated control strategy and its strategy fingerprint are issued. When a network outage is detected, the current strategy fingerprint is read and matched with the warehouse type for verification. When the matching fails or the strategy is missing, the differentiated control strategy of the corresponding warehouse is reconstructed based on the pre-stored strategy fingerprint and the coupled dataset of the environmental parameter domain and the behavioral parameter domain. Based on the reconstructed differentiated control strategy and the coupled dataset, the current warehouse operating status is determined, transient disturbances and continuous offsets are distinguished, and corresponding linkage control commands are determined. During the execution of the linkage control command, the minimum constraint set corresponding to the warehouse type is invoked to perform boundary constraint control on the environmental parameters; Collect execution results and perform feedback verification. After the network is restored, synchronize and update the running data and policy adjustment results during the network outage period.

[0007] Preferably, the archives storage rooms of each campus are classified, including: Extract the confidentiality level, usage frequency within a unit statistical period, and temperature and humidity control range of each warehouse to form a warehouse attribute set; determine the warehouse type based on the warehouse attribute set; retrieve the corresponding environmental control parameters according to the warehouse type as the basic data for generating differentiated control strategies.

[0008] Preferably, the generation of differentiated control strategies includes: Environmental control parameters are segmented according to the sampling time sequence to form stable intervals, slowly changing intervals, and abrupt change intervals; Based on the warehouse type, a control priority order is assigned to each segmented area, and the corresponding control boundaries are determined; The execution intensity is determined based on the degree of deviation of environmental parameters from the center value of the control boundary, forming a differentiated control strategy that includes control boundary, response sequence and execution intensity.

[0009] Preferably, the formation of the strategy fingerprint includes: Extract the control boundaries, response sequence, and execution intensity from the differentiated control strategies; The control boundary is formed into a boundary feature, the response order is formed into a sequence feature, and the execution intensity is formed into an intensity feature; The boundary features, sequence features, and intensity features are combined and encoded to form a policy fingerprint corresponding to the differentiated control strategy.

[0010] Preferably, the construction of the coupled dataset includes: Temperature and humidity are collected at a fixed sampling period, and the variation between adjacent sampling points is extracted to form an environmental parameter sequence. Record the frequency of personnel entry and exit, borrowing operations, and warehouse door opening and closing according to the statistical time window, and form a sequence of behavioral parameters based on the intensity of the behavior; The environmental parameter sequence and the behavioral parameter sequence are aligned according to a unified time period to form a correspondence between environmental changes and behavioral changes.

[0011] Preferably, the construction of the coupled dataset also includes: Based on the magnitude of environmental changes and the intensity of behavior within a continuous time period, the degree of correlation between environmental changes and behavioral changes is determined; based on the degree of correlation, environmental changes caused by behavior and environmental changes not caused by behavior are distinguished; and the time period, environmental change state, behavioral change state, and correlation results are combined to form a coupled dataset.

[0012] Preferably, the current policy fingerprint is read and matched with the warehouse type for verification, including: Under normal network conditions, the differentiated control strategy, strategy fingerprint, and warehouse type are bound and stored. When a network outage is detected, the currently executing strategy fingerprint and corresponding warehouse type are read. The boundary features, sequence features, and intensity features in the current strategy fingerprint are compared with the pre-stored strategy fingerprint of the corresponding warehouse type. If the comparison is inconsistent, it is marked as a strategy anomaly.

[0013] Preferably, the differentiated control strategy for the corresponding warehouse is restructured, including: Retrieve the pre-stored policy fingerprint set corresponding to the current warehouse type and filter out policy fingerprints that are inconsistent with the warehouse type; based on the current environment parameter domain and behavior parameter domain, select feature fingerprints that meet similar requirements from the pre-stored policy fingerprint set; parse the control boundary, response order and execution intensity corresponding to the feature fingerprints, and make corrections in combination with the coupled dataset to obtain the reconstructed differentiated control strategy.

[0014] Preferably, determining the current warehouse operating status and identifying the linkage control command includes: extracting the environmental change magnitude, behavior intensity, and correlation results of the current time period and several consecutive time periods preceding it from the coupled dataset to form a state change sequence; comparing the state change sequence with the control boundary in the reconstructed differentiated control strategy to distinguish between transient disturbances and continuous offsets; selecting the corresponding control response mode based on the determination result, and generating a linkage control command that includes the target warehouse, the main control object, the collaborating object, the execution intensity, the response order, and the duration.

[0015] Preferably, the boundary constraints and feedback updates after executing the linkage control command include: Call the minimum set of constraints corresponding to the warehouse type, adjust the current linkage control command when the environmental parameters are close to the boundary value, and output the constraint control command when the environmental parameters exceed the boundary value; Continuously collect environmental parameter changes and control response data to form an execution result sequence, and compare it with the control objectives in the differentiated control strategy to mark the deviation status; After the network is restored, upload the execution result sequence and policy adjustment results during the network outage, and update the differentiated control policy and corresponding policy fingerprint based on the deviation status.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention addresses the problem of warehouse control easily degenerating into a uniform default policy after network outages. It employs a binding relationship between warehouse type, differentiated control strategy, and policy fingerprint. This allows the edge device to still identify whether the current control strategy matches the physical attributes of the warehouse based on the local policy fingerprint, even when unable to receive cloud commands. Different warehouses have varying tolerances to temperature and humidity fluctuations, personnel disturbances, and equipment response intensity. The policy fingerprint solidifies control boundaries, response sequences, and execution intensity into verifiable characteristics, thereby preventing the teacher warehouse from being mistakenly operated as a student warehouse or using a standard energy-saving strategy. This ensures that the temperature and humidity regulation process remains adapted to the archival storage conditions even during network outages.

[0017] This invention also distinguishes between short-term thermal and humidity disturbances caused by personnel entry / exit and borrowing operations and continuous deviations caused by non-behavioral factors by using coupled datasets of environmental parameter domains and behavioral parameter domains. Furthermore, it introduces a minimum constraint set for boundary constraints when executing linkage control commands. Personnel activities typically cause short-term localized heat and moisture input, while leaks, equipment failures, or the intrusion of external humid air can cause continuous deviations. Based on this, this invention selects delayed, low-intensity, or enhanced control, and changes the equipment output direction when temperature and humidity approach safe boundaries. If the boundaries are exceeded, the original command is overridden to restore the system to the safe range, directly reducing the risks of erroneous linkage, over-adjustment, and network outages leading to loss of control. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of a multi-device heterogeneous system linkage control method for an archive storage facility according to the present invention.

[0020] Figure 2 This is a flowchart of the method for distinguishing between transient disturbances and continuous offsets according to the present invention. Detailed Implementation

[0021] 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. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0022] Example 1, please refer to Figure 1 As shown in this embodiment, a method for coordinated control of a multi-device heterogeneous system in an archive storage facility includes: The archives of each campus are classified, and the types of archives and corresponding environmental control parameters are obtained to generate differentiated control strategies. The differentiated control strategies are then feature-encoded to form corresponding strategy fingerprints.

[0023] In this implementation, the basic attributes of each warehouse are first extracted and quantified. Confidentiality levels are represented by hierarchical numerical values, assigned 1, 2, and 3 for low, medium, and high, respectively. Usage frequency is calculated by counting the number of borrowings per unit time, with a statistical period of 24 hours. The number of borrowings is recorded as n, and divided into low-frequency (n < 10), medium-frequency (10 ≤ n ≤ 50), and high-frequency (n > 50) intervals. Environmental control intervals include temperature and humidity ranges, with the temperature control interval set to 18°C ​​to 24°C and the humidity control interval set to 40% to 60%.

[0024] The above-mentioned confidentiality level values, usage frequency range identifiers, and upper and lower limits of environmental control ranges are combined to form a set of warehouse attributes. The warehouse type is determined according to preset mapping rules. When the confidentiality level is 3 and the usage frequency is low, it is determined to be the teacher warehouse type. When the confidentiality level is 2 or 1 and the usage frequency is medium or high, it is determined to be the student warehouse type.

[0025] After determining the warehouse type, the environmental control parameters were serialized. First, temperature and humidity data were sampled over time at 5-minute intervals to form a discrete data sequence. For any parameter sequence, the difference between adjacent sampling points was calculated to obtain the change sequence. The change was calculated by subtracting the previous sampling value from the current sample value. When the absolute value of the change was less than 0.5 for three consecutive sampling periods, the time period was classified as a stable interval; when the absolute value of the change was greater than or equal to 0.5 and less than or equal to 2, it was classified as a slowly changing interval; and when the absolute value of the change was greater than 2, it was classified as a sudden change interval.

[0026] Based on the above interval division results, the environmental control parameters are divided into multiple segmented intervals according to time sequence, and control priority is assigned according to the type of storage room. For the teacher storage room type, the priority of the stable interval is set to 1, the gradual change interval to 2, and the sudden change interval to 3; for the student storage room type, the priority of the sudden change interval is set to 1, the gradual change interval to 2, and the stable interval to 3, thus forming a parameter sequence with sequential correlation.

[0027] After obtaining the parameter sequence, it is combined and mapped with the warehouse type. During the mapping process, the control boundary is first determined. The control boundary is taken as the upper and lower limits of the environmental control range. For the teacher warehouse type, the control bandwidth is narrowed based on the original temperature and humidity ranges. The temperature range is adjusted to 20 degrees Celsius to 22 degrees Celsius, and the humidity range is adjusted to 45% to 55%. For the student warehouse type, the original range remains unchanged.

[0028] Then, the response order is determined based on the interval types and their priority in the parameter sequence. When a sudden change interval is detected, the control action is triggered first, followed by a gradually changing interval, and finally a stable interval. The execution intensity is determined by the degree of interval deviation, which is represented by the absolute value of the difference between the current value and the center value of the control boundary. The center value is the average of the upper and lower limits. When the deviation is less than 1, the execution intensity is set to low; when it is greater than or equal to 1 and less than or equal to 3, it is set to medium; and when it is greater than 3, it is set to high. Through the combination of the above control boundary, response order, and execution intensity, a differentiated control strategy structure is formed.

[0029] After forming the differentiated control strategy structure, feature encoding processing is performed on it. First, control boundary features are extracted, and the upper and lower limits of temperature and humidity are arranged in order to form a boundary vector. Then, sequence features are extracted, and the priority of each segment interval is arranged in chronological order to form a sequence sequence. Finally, intensity features are extracted, and the execution intensity of the corresponding time period is mapped to the values ​​1, 2, and 3 respectively as low, medium, and high to form an intensity sequence.

[0030] The boundary vector, sequence sequence, and intensity sequence are combined and encoded by concatenating the numerical values ​​of each sequence sequentially. The resulting numerical sequence is then weighted and summed, with weighting coefficients cyclically set to 1, 2, and 3. The final result is a unique numerical value that serves as the strategy fingerprint. This method allows different warehouse types and their control strategies to correspond to different strategy fingerprints, thus achieving a unique identifier for differentiated control strategies and facilitating subsequent matching.

[0031] Collect environmental parameter data and behavioral parameter data from each warehouse to construct a coupled dataset of environmental parameter domain and behavioral parameter domain.

[0032] In this embodiment, the environmental parameter data is first processed by continuous time segmentation. The environmental parameters include temperature and humidity, and the sampling period is set to a fixed time interval, preferably 5 minutes, that is, the temperature and humidity values ​​are recorded every 5 minutes to form time series data.

[0033] For any environmental parameter sequence, the difference between two adjacent sampling points is taken as the amplitude of change, which is equal to the absolute value of the current sampled value minus the previous sampled value. Based on multiple consecutive sampling points, the amplitude of change is statistically analyzed. When the amplitude of change is less than 0.3 for three consecutive sampling periods, the time interval is divided into a stable interval; when the amplitude of change is greater than or equal to 0.3 and less than or equal to 1.5, the time interval is divided into a slowly changing interval; when the amplitude of change is greater than 1.5, the time interval is divided into a drastically changing interval.

[0034] Furthermore, to avoid short-term abnormal interference, a sliding window is introduced during the interval division process. The window length is set to 3 sampling points. The average value of the variation amplitude within the window is taken. The average value is calculated by summing the variation amplitudes within the window and dividing by the window length, thus obtaining a smoothed variation amplitude sequence. Based on the above division results, the environmental parameters are arranged in chronological order to form an environmental parameter sequence consisting of stable intervals, slowly changing intervals, and drastically changing intervals. The start time, duration, and corresponding average variation amplitude of each interval are recorded.

[0035] Behavioral parameters include the number of personnel entering and exiting, the number of borrowing operations, and the frequency of opening and closing the warehouse door. Using time as the main thread, various behavioral events are recorded to form an original behavioral record sequence. For behavioral data within any given time period, the number of events per unit time is first counted. A statistical window of 10 minutes is set, within which the number of personnel entering and exiting (m1), the number of borrowing operations (m2), and the frequency of opening and closing (m3) are counted respectively. To unify the impact of different types of behavior, each behavior is weighted. The weight for personnel entering and exiting is set to 0.5, the weight for borrowing operations to 0.3, and the weight for opening and closing frequency to 0.2. The frequency of each of the three types of behavior is multiplied by its corresponding weight and then summed to obtain the behavior intensity value. The behavior intensity value is equal to m1 multiplied by 0.5, m2 multiplied by 0.3, and m3 multiplied by 0.2.

[0036] The behavior intensity values ​​are used for classification. A value less than 5 is classified as a low behavior interval, a value greater than or equal to 5 and less than or equal to 20 is classified as a medium behavior interval, and a value greater than 20 is classified as a high behavior interval. The behavior intervals corresponding to each time window are arranged in chronological order to form a sequence of behavior parameters, and the time position and corresponding intensity level of each interval are recorded.

[0037] The environmental parameter sequences and behavioral parameter sequences are aligned according to their temporal correspondence. The alignment process is based on a unified time axis, mapping the environmental and behavioral parameter sequences to the same time scale. Since the environmental parameter sampling period is 5 minutes, while the behavioral parameter statistical window is 10 minutes, a unified alignment unit of 10 minutes is used. Two adjacent environmental parameter intervals are merged into one time period, and the average value of the environmental change amplitude within that time period is calculated. The average value equals the sum of all change amplitudes within that time period divided by the number of data points. Subsequently, a one-to-one correspondence is established between the environmental interval type and the corresponding behavioral interval type within that time period, establishing a time-period-level correlation mapping relationship. For each time period, the environmental interval type, the average change amplitude, and the behavioral interval type are recorded, thus forming a set of correspondences between environmental changes and behavioral changes.

[0038] To enhance the effectiveness of the correlation mapping, the correlation between environmental changes and behavioral changes is quantified. For multiple consecutive time periods, the correlation coefficient between the environmental change amplitude sequence and the behavioral intensity sequence is calculated. The correlation coefficient is calculated as follows: First, the average value of the environmental change amplitude sequence and the average value of the behavioral intensity sequence are calculated separately; then, for each time period, the product of the difference between the environmental change amplitude and its average value and the difference between the behavioral intensity and its average value is calculated, and the products for all time periods are summed; next, the sum of squares of the differences between the environmental change amplitude and its average value and the sum of squares of the differences between the behavioral intensity and its average value are calculated separately; finally, the sum of the products is divided by the product of the square roots of the two sums of squares to obtain the correlation coefficient value.

[0039] The correlation coefficient is used to determine the degree of association between environmental changes and behavioral changes. When the correlation coefficient is greater than 0.6, the two are considered to be strongly associated, and when it is less than or equal to 0.6, the association is considered to be weak.

[0040] For each time period, if the environmental interval type is a slow change or a drastic change interval and the corresponding behavior interval is a high behavior interval, and the correlation coefficient is greater than 0.6, then the environmental change in that time period is determined to be mainly caused by the behavior change; if the environmental interval is a drastic change interval and the behavior interval is a low behavior interval, and the correlation coefficient is less than or equal to 0.6, then the environmental change is determined to be an abnormal shift.

[0041] Based on this determination, the environmental interval type, behavioral interval type, and associated determination results for each time period are combined to form coupled data items, which are then arranged in chronological order to form a coupled dataset. This coupled dataset not only contains information on environmental changes but also reflects the influence of behavioral factors on environmental changes, thus providing a basis for subsequent control strategy determination.

[0042] Under normal network conditions, the differentiated control strategy and its strategy fingerprint are issued. When a network outage is detected, the current strategy fingerprint is read and matched with the warehouse type for verification. When the match fails or the strategy is missing, the differentiated control strategy for the corresponding warehouse is reconstructed based on the pre-stored strategy fingerprint and the coupled dataset of the environmental parameter domain and the behavioral parameter domain.

[0043] In this implementation, under normal network conditions, differentiated control policies and their corresponding policy fingerprints are bound to the same policy record. This policy record includes the warehouse type, control boundary, response order, execution strength, policy fingerprint, issuance time, and validity flag. During binding, the consistency between the warehouse type and the policy fingerprint is first verified. Once the consistency is established, the policy record is issued to the local storage area of ​​the corresponding warehouse, and at least three of the most recent valid policy records are retained. If multiple policy records exist for the same warehouse, the policy record with the issuance time closest to the current time and a valid flag of "true" is selected as the currently executable policy.

[0044] Network connectivity is confirmed through periodic communication. The communication confirmation period is set to 30 to 120 seconds, preferably 60 seconds. If no valid communication confirmation is received three consecutive times, the network is considered interrupted; if valid communication is restored within two consecutive confirmation periods, the network is considered restored. After determining a network interruption, the currently executing policy record is read, and the policy fingerprint and warehouse type are extracted. If the current policy record does not exist, the policy fingerprint is empty, the warehouse type is empty, or the valid flag is false, the current policy is directly recorded as an abnormal state.

[0045] It should be noted that consistency comparison includes type comparison and fingerprint comparison. During type comparison, the warehouse type in the current policy record is compared with the warehouse type corresponding to the locally stored warehouse attribute set. If the two are inconsistent, it is determined that the types do not match.

[0046] During fingerprint comparison, control boundary features, sequence features, and intensity features are read from the current policy fingerprint and compared with the pre-stored policy fingerprint set corresponding to the warehouse type. The comparison process is as follows: First, the control boundary difference is calculated as the average of the absolute values ​​of the differences in the upper and lower temperature limits, the upper and lower humidity limits; then, the sequence difference is calculated as the proportion of the number of items with inconsistent positions in the response sequence to the total number of sequential items; finally, the intensity difference is calculated as the average of the absolute values ​​of the differences in execution intensity levels across different time periods. If the control boundary difference is less than or equal to 1, the sequence difference is less than or equal to one-third, and the intensity difference is less than or equal to 1, the policy fingerprint is determined to match the warehouse type; if any condition is not met, the current policy is determined to be in an abnormal state, and verification result information is generated.

[0047] The verification result information includes the cause of the anomaly, the current warehouse type, the current policy fingerprint, the most recent valid policy fingerprint, and the time when the anomaly occurred, which can be used for subsequent reconstruction of differentiated control policies.

[0048] When the comparison results are inconsistent or the strategy is missing, retrieve the pre-stored strategy fingerprint set corresponding to the current warehouse type.

[0049] Each record in the pre-stored policy fingerprint set contains a policy fingerprint, control boundary, response order, execution intensity, and applicable time period. When extracting feature fingerprints, records with inconsistent warehouse types are first filtered out. Then, the similarity is calculated based on the current environmental parameter domain and behavioral parameter domain. The similarity calculation process is as follows: The absolute value of the difference between the current environmental change magnitude and the environmental change magnitude in the pre-stored record, and the absolute value of the difference between the current behavioral intensity and the behavioral intensity in the pre-stored record are calculated separately. Both are then converted to percentage differences, with environmental differences accounting for 60% and behavioral differences accounting for 40%. The two are added together to obtain the comprehensive difference value. The similarity value is obtained by subtracting the comprehensive difference value from 100. Policy fingerprints with a similarity value greater than or equal to 80 are used as feature fingerprints; when multiple feature fingerprints exist, the one with the highest similarity value is selected.

[0050] After obtaining the characteristic fingerprint, the corresponding control boundary and execution order are analyzed to form a candidate control parameter sequence. The candidate control parameter sequence is arranged in chronological order, and each item includes the upper limit of temperature, the lower limit of temperature, the upper limit of humidity, the lower limit of humidity, the response order, and the execution intensity. If the current warehouse type is a teacher warehouse, the control boundary adopts a narrower interval; if the current warehouse type is a student warehouse, the control boundary adopts the corresponding environmental control interval.

[0051] During the analysis process, if there are missing items in the control boundary, the same item from the most recent valid strategy record in the same warehouse type will be used to fill the gap; if there are missing items in the response sequence, they will be filled in the order of humidity control, temperature control, and ventilation control.

[0052] If the coupled dataset shows a strong correlation between current environmental changes and personnel entry / exit, borrowing operations, or opening / closing frequency, and the environmental parameters do not exceed the control boundary within two consecutive time periods, the execution intensity will remain unchanged or be reduced by one level. If the environmental parameters exceed the control boundary for three consecutive time periods, and the behavioral parameters are in the low behavioral range, the execution intensity will be increased by one level, and the response order will be shifted forward by one position. If the environmental parameters are close to the control boundary, with the close range set to be no more than 10% of the control interval width from the boundary value, the execution intensity will not be reduced further.

[0053] The revised control parameter sequence is then recombined according to the order of control boundary, response sequence, and execution intensity to form a differentiated control strategy for the corresponding warehouse, and this differentiated control strategy serves as the basis for subsequent control execution during the network outage.

[0054] Based on the reconstructed differentiated control strategy and the coupled dataset, the current warehouse operating status is determined, transient disturbances and continuous offsets are distinguished, and corresponding linkage control commands are determined.

[0055] In this implementation, the current time period is aligned in 10-minute units. When determining the warehouse's operational status, data items for the current time period and the two preceding consecutive time periods are first read from the coupled dataset. Each data item includes the environmental interval type, the average environmental change range, the behavior interval type, the behavior intensity value, and the associated determination result. The environmental parameter change range includes temperature and humidity changes. The temperature change range is obtained by the absolute value of the difference between the average temperature of the current time period and the average temperature of the previous time period. The humidity change range is obtained by the absolute value of the difference between the average humidity of the current time period and the average humidity of the previous time period. The behavior parameter change characteristics are determined based on the behavior intensity values ​​corresponding to the number of personnel entering and exiting, the number of borrowing operations, and the frequency of opening and closing. A behavior intensity value less than 5 indicates a low behavior interval, greater than or equal to 5 and less than or equal to 20 indicates a medium behavior interval, and greater than 20 indicates a high behavior interval.

[0056] The environmental change magnitudes, behavioral interval types, and correlation determination results for the current time period and its two preceding consecutive time periods are arranged chronologically to form a state change sequence. Each sequence item in the state change sequence includes at least a time identifier, temperature change magnitude, humidity change magnitude, environmental interval type, behavioral interval type, and correlation determination result. This state change sequence is used to reflect whether environmental parameters change synchronously with behavioral parameters and whether the changes in environmental parameters are continuous. If the temperature change magnitude is less than 1 degree Celsius and the humidity change magnitude is less than 3% within a certain time period, it is recorded as a slight change; if the temperature change magnitude is greater than or equal to 1 degree Celsius and less than or equal to 2 degrees Celsius, or the humidity change magnitude is greater than or equal to 3% and less than or equal to 8%, it is recorded as a moderate change; if the temperature change magnitude is greater than 2 degrees Celsius or the humidity change magnitude is greater than 8%, it is recorded as a severe change.

[0057] Please see Figure 2 As shown, after obtaining the state change sequence, the average temperature and average humidity of each time period are compared with the control boundary in the reconstructed differentiated control strategy.

[0058] When the average temperature is higher than the upper temperature limit or lower than the lower temperature limit, it is considered a temperature out-of-bounds condition; when the average humidity is higher than the upper humidity limit or lower than the lower humidity limit, it is considered a humidity out-of-bounds condition. If an environmental parameter is within the control boundary and only shows slight or moderate changes, it is considered a fluctuation within the boundary. If an environmental parameter exceeds the boundary for one time period but recovers to within the control boundary in a subsequent time period, and the corresponding behavior interval type for that time period is a medium or high behavior interval, and the correlation determination result shows that the environmental change and the behavior change are correlated, it is considered a transient disturbance. If an environmental parameter exceeds the boundary for two consecutive time periods, but the out-of-bounds amplitude does not exceed 20% of the control interval width, and recovers to within the control boundary in a subsequent time period, it is also considered a transient disturbance.

[0059] When environmental parameters are out of bounds for three or more consecutive time periods, or the out-of-bounds range exceeds 20% of the width of the corresponding control interval, and the behavior interval type for the corresponding time period is a low behavior interval, and the correlation judgment result does not show a corresponding relationship between environmental changes and behavior changes, it is judged as a continuous offset.

[0060] For humidity exceeding the limit, if the humidity continuously increases and remains above the upper limit, it is primarily considered a continuous deviation. For temperature exceeding the limit, if the temperature continuously deviates from the control boundary without being accompanied by personnel entering or leaving, borrowing operations, or an increase in the frequency of opening and closing, it is also considered a continuous deviation. If both temperature and humidity exceed the limit simultaneously, the parameter with the longer duration and larger magnitude of the deviation is designated as the primary control parameter, and the other parameter is designated as the coordinating parameter.

[0061] After determining whether a transient disturbance or persistent offset has occurred, the corresponding control response method is selected from the differentiated control strategy. If the disturbance is determined to be transient, a low-intensity or delayed response method is selected, with the delay duration set to one time period. During the delay, environmental parameters continue to be collected. If the disturbance recovers to within the control boundary, the execution intensity is not increased; only hold or fine-tune linkage control commands are output. If the offset is determined to be persistent, a response method one level higher than the current execution intensity is selected, and linkage control commands are formed according to the response order in the differentiated control strategy. For persistent humidity offset, the linkage control commands prioritize dehumidification actions, followed by air conditioning coordination actions; for persistent temperature offset, the linkage control commands prioritize air conditioning adjustment actions, followed by fresh air restriction or shutdown actions. The linkage control commands must include at least the target warehouse, the main controlled object, the cooperating object, the execution intensity, the response order, and the duration, and are output to subsequent control execution stages.

[0062] During the execution of the linkage control command, the minimum constraint set corresponding to the warehouse type is invoked to perform boundary constraint control on the environmental parameters.

[0063] In this embodiment, the minimum constraint set is pre-established according to the warehouse type and maintains a one-to-one correspondence with the warehouse type.

[0064] The minimum constraint set includes a lower temperature limit, an upper temperature limit, a lower humidity limit, an upper humidity limit, a proximity range, and a recovery target range. The minimum constraint set for the teacher database is preferably set to a temperature of 18°C ​​to 24°C and a humidity of 40% to 60%; the minimum constraint set for the student database is preferably set to a temperature of 16°C to 26°C and a humidity of 35% to 65%.

[0065] When record management standards or local configuration requirements are more stringent, a narrower boundary range should be used as the final boundary value. The approach range is determined based on the width of the boundary interval, which is the value obtained by subtracting the lower limit from the upper limit boundary. The approach range is 10% to 20% of the boundary interval width, preferably 15%. Taking the temperature of the teacher's library as an example, the temperature interval width is 6 degrees Celsius, and the approach range is preferably 0.9 degrees Celsius; when the current temperature is no more than 0.9 degrees Celsius away from 18 degrees Celsius or 24 degrees Celsius, it is considered to be close to the temperature boundary.

[0066] During the execution of the linkage control command, environmental parameter data is acquired according to a fixed acquisition cycle, which is set to 1 to 5 minutes, preferably 2 minutes. After each acquisition, the temperature and humidity values ​​are compared against the boundary conditions.

[0067] When comparing boundaries, first determine if the current environmental parameter is less than the lower boundary or greater than the upper boundary. If it is within the boundary, then calculate the distance between the current environmental parameter and the adjacent boundary. When the current environmental parameter is close to the upper boundary, the absolute value of the difference between the current value and the upper boundary is used as the upper boundary distance; when the current environmental parameter is close to the lower boundary, the absolute value of the difference between the current value and the lower boundary is used as the lower boundary distance. When the upper boundary distance or lower boundary distance is less than or equal to the corresponding proximity range, the environmental parameter is considered to be in a boundary proximity state.

[0068] When environmental parameters are close to their limits, the current linkage control command is adjusted to meet constraints. If the temperature is close to the upper limit, the control output that may continue to rise is reduced, and the control output in the cooling direction is increased first. If the temperature is close to the lower limit, the control output that may continue to fall is reduced, and the control output in the heating direction is increased first.

[0069] If the humidity is close to the upper limit, reduce the control output that may increase humidity and prioritize increasing the control output in the dehumidification direction; if the humidity is close to the lower limit, reduce the control output that may continue dehumidification and limit the control output in the dehumidification direction. The control output after constraint adjustment must not change the safety boundary corresponding to the warehouse type in the differentiated control strategy, and can only be modified within the range of execution intensity, duration, and response sequence.

[0070] When environmental parameters exceed the boundary values ​​in the minimum constraint set, pause the content in the original linkage control command that is opposite to the recovery direction, and output constraint control command.

[0071] The constraint control command aims to restore a target humidity range, which lies between the upper and lower limits, preferably between 40% and 60%. For example, with a humidity level of 40% to 60% in the teacher's warehouse, the target range is preferably 48% to 52%. If the humidity exceeds 60%, the constraint control command prioritizes dehumidification control and restricts fresh air intake or door opening linkage; if the humidity is below 40%, the constraint control command stops or reduces dehumidification control and prohibits further adjustment towards lower humidity. If the temperature exceeds the upper limit, the constraint control command prioritizes cooling control; if the temperature is below the lower limit, the constraint control command prioritizes heating control.

[0072] After the constraint control command is executed, environmental parameter data continues to be acquired according to the acquisition cycle, and it is determined whether the environmental parameters have returned to the recovery target range.

[0073] If the target recovery range is reached for two consecutive acquisition cycles, the coverage state is lifted, and the system returns to the normal control response corresponding to the original linkage control command or differentiated control strategy. If the system only returns to the boundary range but does not enter the target recovery range, the constraint control command is maintained, and the execution intensity is reduced by one level to continue running. If the system still fails to return to the boundary range for three consecutive acquisition cycles, the coverage state is maintained, and this running state is recorded as a boundary recovery failure record for subsequent feedback verification.

[0074] Through the above processing, the linkage control commands are limited by the minimum set of constraints during execution, preventing environmental parameters from exceeding the safety boundary corresponding to the warehouse type when the network is down or the strategy is abnormal.

[0075] Collect execution results and perform feedback verification. After the network is restored, synchronize and update the running data and policy adjustment results during the network outage period.

[0076] In this embodiment, during the execution of the linkage control command, environmental parameter changes and control response are continuously acquired according to a fixed acquisition cycle. The acquisition cycle can be set from 1 minute to 5 minutes, preferably 2 minutes. The environmental parameter change results include the current temperature value, current humidity value, temperature change range, humidity change range, and whether it is within the control boundary; the control response includes the issuance time of the linkage control command, the target warehouse, the main control object, the cooperating object, the execution intensity, the duration, and the action completion status. After each acquisition, the environmental parameter change results and control response corresponding to the same time point are merged into one execution result item, and arranged according to the acquisition time to form an execution result sequence.

[0077] The variation range of environmental parameters in the execution result sequence is calculated based on two consecutive collected values. The temperature variation range is the absolute value of the difference between the current temperature value and the previous temperature value, and the humidity variation range is the absolute value of the difference between the current humidity value and the previous humidity value.

[0078] The action completion status is divided into three categories based on the feedback result: executed, partially executed, and not executed. When both the controlling object and the collaborating object complete the corresponding action, it is recorded as executed; when only part of the action is completed, it is recorded as partially executed; and when no action feedback is received or the feedback is a failure, it is recorded as not executed. The execution result item also records the duration corresponding to the action completion status, facilitating subsequent assessment of whether there is a delay in the control response.

[0079] Subsequently, the sequence of execution results is compared with the control objectives in the differentiated control strategy. The control objectives include the target environmental range, the target recovery time limit, and the target execution intensity.

[0080] The target environment range is derived from the control boundary in the differentiated control strategy. The target recovery time limit is set according to the warehouse type, preferably 20 to 40 minutes for the teacher warehouse and 30 to 60 minutes for the student warehouse. The target execution intensity is represented by three levels: low, medium, and high, corresponding to levels 1, 2, and 3, respectively.

[0081] When comparing, first determine whether the environmental parameters have returned to the target environment range within the target recovery time limit; then determine whether the actual execution intensity is consistent with the target execution intensity; finally determine whether the action completion status has reached the executed state.

[0082] Deviation states are categorized into environmental deviation, response deviation, and execution deviation. Environmental deviation refers to the failure of environmental parameters to return to the target environmental range after the target recovery time limit has expired; response deviation refers to the action completion time exceeding one-third of the target recovery time limit; and execution deviation refers to the action completion status being partially executed or not executed at all.

[0083] If two or more types of deviations exist simultaneously, they are marked as compound deviations. The degree of deviation is determined by the amount of deviation, which is the value by which the current environmental parameter exceeds the boundary of the target environmental range. When the deviation is less than 10% of the width of the target environmental range, it is marked as a slight deviation; when the deviation is greater than or equal to 10% and less than or equal to 25%, it is marked as a moderate deviation; and when the deviation is greater than 25%, it is marked as a severe deviation.

[0084] During the network outage, the execution result sequence and the corresponding strategy adjustment process are sequentially recorded and saved locally. The strategy adjustment process includes the adjustment trigger time, trigger reason, execution intensity before adjustment, execution intensity after adjustment, control boundary before adjustment, control boundary after adjustment, whether the minimum constraint set is called, and deviation status.

[0085] Each strategy adjustment establishes a time correspondence with the adjacent execution result item, enabling subsequent tracking of the adjustment source corresponding to a certain control result.

[0086] To avoid record loss during network outages, local storage uses incrementally increasing numbers in chronological order, and each record includes the number of the previous record. When storage capacity is close to its limit, priority is given to retaining data containing records of moderate deviation, severe deviation, composite deviation, and minimum constraint set calls.

[0087] After network recovery, the integrity of the execution result sequence and policy adjustment results saved during the network outage is first verified. Integrity verification includes checking the continuity of numbers, the time sequence, and the record content. If the numbers are not consecutive, the missing interval is marked; if the time is reversed, they are reordered according to the collection time; if the record content lacks warehouse type, policy fingerprint, or deviation status, the record is marked as incomplete and will not participate in policy updates. Records that pass verification are uploaded uniformly and aggregated according to warehouse type, policy fingerprint, and deviation status.

[0088] After the upload is completed, the differentiated control strategy and corresponding strategy fingerprint are updated based on the strategy adjustment results. If a warehouse experiences more than 3 consecutive environmental deviations under the same control boundary during the network outage, the corresponding response order will be moved forward by 1 position or the execution intensity will be increased by 1 level. If the control results return to the target environment range within the target recovery time limit for 3 consecutive times and the minimum constraint set is not invoked, the original control boundary and execution intensity will remain unchanged.

[0089] After the policy update is completed, the control boundary features, sequence features, and intensity features are re-extracted and combined into an updated policy fingerprint according to a predetermined encoding method, so that the updated differentiated control policy and the policy fingerprint maintain a correspondence.

[0090] Example 2: To verify the effectiveness of the proposed method for coordinated control of a heterogeneous multi-device system in archive storage under network outage conditions, four archive storage rooms from two campuses within the same school were selected for operational testing. The first campus included one teacher storage room and one student storage room, and the second campus also included one teacher storage room and one student storage room. The test was conducted continuously for 6 hours, simulating an external network outage to prevent the cloud from continuously disseminating control strategies. The control method used was the conventional method of calling a unified default energy-saving strategy after network outage. The proposed method employed strategy fingerprint matching, differentiated control strategy reconstruction, coupled dataset determination, and minimum constraint set boundary control. Each storage room was equipped with air conditioning, dehumidification equipment, fresh air equipment, access control record collection units, and temperature and humidity collection units. The temperature and humidity collection cycle was 2 minutes, and the behavioral parameter statistics cycle was 10 minutes.

[0091] Before testing, the differentiated control strategy for the teacher database was set to a temperature of 20°C to 22°C and a humidity of 45% to 55%, with a minimum constraint set of 18°C ​​to 24°C and 40% to 60% humidity. The differentiated control strategy for the student database was set to a temperature of 18°C ​​to 24°C and a humidity of 40% to 60%, with a minimum constraint set of 16°C to 26°C and 35% to 65% humidity. The external ambient temperature was 29°C to 31°C, and the external humidity was 73% to 82%. After a network outage, the control method no longer distinguished between the teacher and student databases and uniformly adopted the energy-saving strategy. This application method first reads the current strategy fingerprint and matches and verifies it against the database type. When a missing strategy or inconsistent strategy fingerprint is detected in the teacher database, the pre-stored strategy fingerprint set for the corresponding database type is retrieved, and the differentiated control strategy is reconstructed by combining the coupled dataset of the current environmental parameter domain and behavioral parameter domain.

[0092] Table 1 Test Objects and Control Boundary Settings

[0093] As shown in Table 1, during the test, from the 80th to the 120th minute, a scenario of concentrated entry and exit of personnel and borrowing operations was set up for the student database, forming a scenario of increased behavioral parameters; from the 160th to the 220th minute, a scenario of external humid air infiltration was set up for the teacher database, but without concentrated entry and exit of personnel, to simulate a continuous humidity shift not caused by behavior. When using the method of this application, the humidity increase of the student database from the 80th to the 120th minute was determined to be a transient disturbance related to behavioral changes, so delayed observation and low-intensity response were adopted; the humidity increase of the teacher database from the 160th to the 220th minute was determined to be a continuous shift unrelated to behavioral changes, so the dehumidification response sequence and execution intensity were increased.

[0094] Table 2. Partial Operational Data During Network Outage

[0095] Table 3 Comparison of test results between the control method and the method of this application

[0096] As can be seen from Table 2 and Table 3, in the control mode, after the network is disconnected, the teacher library is taken over by the unified default energy-saving strategy, and the humidity rises to 66.3% at most, which has exceeded the humidity upper limit of 60% in the minimum constraint set of the teacher library, and the cumulative over-limit time is 74 minutes; when the method of this application is adopted, after the network of the teacher library is disconnected, the strategy fingerprint matching is used to find that the strategy is abnormal, and the differential control strategy is reconstructed according to the warehouse type, the pre-stored strategy fingerprint set and the coupled data set, so that the maximum humidity value is controlled at 58.9%, which does not exceed the limit range of the minimum constraint set. This result shows that when the network is interrupted and the cloud strategy cannot be sent down continuously, the method of this application can still maintain the differential control relationship between the teacher library and the student library, and avoid the teacher library being operated according to the ordinary energy-saving strategy by mistake.

[0097] At the same time, it can be seen from the test results of the student library that when the humidity rises short-term due to the concentrated entry and exit of personnel and borrowing operations, the control mode only triggers forced dehumidification based on the humidity value, and there are 5 misforced controls; the method of this application combines the change characteristics of behavior parameters, identifies this type of humidity change as a transient disturbance, and only retains 1 low-intensity control after a delay, reducing unnecessary linkage control. Thus, the coupled data set can be used to distinguish the short-term fluctuations caused by behavior from the continuous offset caused by non-behavior, and make the linkage control instruction match the operating state of the warehouse.

[0098] After the network is restored, the method of this application uploads the execution result sequence, strategy adjustment result, deviation state and minimum constraint set call record formed during the network disconnection period in a unified manner, and updates the differential control strategy and the corresponding strategy fingerprint. During the test, the integrity rate of the effective operation records of the method of this application is 100%, and it can trace every process of strategy reconstruction, execution intensity adjustment and boundary constraint control during the network disconnection period; due to the lack of continuous numbering and strategy adjustment records in the control mode, some operation processes cannot be restored.

[0099] The above data shows that the method of this application can keep the environmental parameters of different warehouses within the corresponding safe range through strategy fingerprint verification, differential control strategy reconstruction, distinction between transient disturbance and continuous offset, and boundary control of the minimum constraint set under the condition of network disconnection, and provide traceable data for the strategy update after the network is restored.

[0100] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application.

Claims

1. A method for coordinated control of a multi-device heterogeneous system in an archive storage facility, characterized in that, include: The archives of each campus are classified, and the types of archives and corresponding environmental control parameters are obtained to generate differentiated control strategies. The differentiated control strategies are then feature-encoded to form corresponding strategy fingerprints. Collect environmental parameter data and behavioral parameter data from each warehouse, and construct a coupled dataset of environmental parameter domain and behavioral parameter domain; Under normal network conditions, the differentiated control strategy and its strategy fingerprint are issued. When a network outage is detected, the current strategy fingerprint is read and matched with the warehouse type for verification. When matching fails or a strategy is missing, the corresponding warehouse's differentiated control strategy is reconstructed based on the pre-stored strategy fingerprint and the coupled dataset of the environmental parameter domain and the behavioral parameter domain. Based on the reconstructed differentiated control strategy and the coupled dataset, the current warehouse operating status is determined, transient disturbances and continuous offsets are distinguished, and corresponding linkage control commands are determined. During the execution of the linkage control command, the minimum constraint set corresponding to the warehouse type is invoked to perform boundary constraint control on the environmental parameters; Collect execution results and perform feedback verification. After the network is restored, synchronize and update the running data and policy adjustment results during the network outage period.

2. The method for coordinated control of a multi-device heterogeneous system in an archive storage facility according to claim 1, characterized in that, The archives storage rooms of each campus are classified, including: Extract the confidentiality level, usage frequency within a unit statistical period, and temperature and humidity control range of each warehouse to form a warehouse attribute set; determine the warehouse type based on the warehouse attribute set; retrieve the corresponding environmental control parameters according to the warehouse type as the basic data for generating differentiated control strategies.

3. The method for coordinated control of a multi-device heterogeneous system in an archive storage facility according to claim 2, characterized in that, Generate differentiated control strategies, including: Environmental control parameters are segmented according to the sampling time sequence to form stable intervals, slowly changing intervals, and abrupt change intervals; Based on the warehouse type, a control priority order is assigned to each segmented area, and the corresponding control boundaries are determined; The execution intensity is determined based on the degree of deviation of environmental parameters from the center value of the control boundary, forming a differentiated control strategy that includes control boundary, response sequence and execution intensity.

4. The method for coordinated control of a multi-device heterogeneous system in an archive storage facility according to claim 3, characterized in that, The formation of a strategy fingerprint includes: Extract the control boundaries, response sequence, and execution intensity from the differentiated control strategies; The control boundary is formed into a boundary feature, the response order is formed into a sequence feature, and the execution intensity is formed into an intensity feature; The boundary features, sequence features, and intensity features are combined and encoded to form a policy fingerprint corresponding to the differentiated control strategy.

5. The method for coordinated control of a multi-device heterogeneous system in an archive storage facility according to claim 1, characterized in that, The construction of the coupled dataset includes: Temperature and humidity are collected at a fixed sampling period, and the variation between adjacent sampling points is extracted to form an environmental parameter sequence. Record the frequency of personnel entry and exit, borrowing operations, and warehouse door opening and closing according to the statistical time window, and form a sequence of behavioral parameters based on the intensity of the behavior; The environmental parameter sequence and the behavioral parameter sequence are aligned according to a unified time period to form a correspondence between environmental changes and behavioral changes.

6. The method for coordinated control of a multi-device heterogeneous system in an archive storage facility according to claim 5, characterized in that, The construction of the coupled dataset also includes: Based on the magnitude of environmental changes and the intensity of behavior within a continuous time period, the degree of correlation between environmental changes and behavioral changes is determined; based on the degree of correlation, environmental changes caused by behavior and environmental changes not caused by behavior are distinguished; and the time period, environmental change state, behavioral change state, and correlation results are combined to form a coupled dataset.

7. The method for coordinated control of a multi-equipment heterogeneous system in an archive storage facility according to claim 4, characterized in that, Read the current policy fingerprint and match it with the warehouse type for verification, including: Under normal network conditions, the differentiated control strategy, strategy fingerprint, and warehouse type are bound and stored. When a network outage is detected, the currently executing strategy fingerprint and corresponding warehouse type are read. The boundary features, sequence features, and intensity features in the current strategy fingerprint are compared with the pre-stored strategy fingerprint of the corresponding warehouse type. If the comparison is inconsistent, it is marked as a strategy anomaly.

8. The method for coordinated control of a multi-device heterogeneous system in an archive storage facility according to claim 7, characterized in that, Reconstruct the differentiated control strategy for the corresponding warehouses, including: Retrieve the pre-stored policy fingerprint set corresponding to the current warehouse type and filter out policy fingerprints that are inconsistent with the warehouse type; based on the current environment parameter domain and behavior parameter domain, select feature fingerprints that meet similar requirements from the pre-stored policy fingerprint set; parse the control boundary, response order and execution intensity corresponding to the feature fingerprints, and make corrections in combination with the coupled dataset to obtain the reconstructed differentiated control strategy.

9. A method for coordinated control of a multi-equipment heterogeneous system in an archive storage facility according to claim 8, characterized in that, The process of determining the current warehouse operating status and identifying linkage control instructions includes: extracting the environmental change magnitude, behavioral intensity, and correlation results of the current time period and several consecutive time periods preceding it from the coupled dataset to form a state change sequence; comparing the state change sequence with the control boundary in the reconstructed differentiated control strategy to distinguish between transient disturbances and continuous offsets; selecting the corresponding control response mode based on the determination result, and generating linkage control instructions that include the target warehouse, the main control object, the collaborating object, the execution intensity, the response order, and the duration.

10. A method for coordinated control of a multi-device heterogeneous system in an archive storage facility according to claim 9, characterized in that, Boundary constraint and feedback updates after executing the linkage control command include: Call the minimum set of constraints corresponding to the warehouse type, adjust the current linkage control command when the environmental parameters are close to the boundary value, and output the constraint control command when the environmental parameters exceed the boundary value; Continuously collect environmental parameter changes and control response data to form an execution result sequence, and compare it with the control objectives in the differentiated control strategy to mark the deviation status; After the network is restored, upload the execution result sequence and policy adjustment results during the network outage, and update the differentiated control policy and corresponding policy fingerprint based on the deviation status.

Images