Collection differentiation temperature and humidity environment intelligent regulation and control system and regulation and control method

Abstract

The invention relates to the technical field of intelligent environment control, discloses a collection differentiation temperature and humidity environment intelligent regulation and control system and a regulation and control method, and aims to solve the problem that in the prior art, unified temperature and humidity regulation and control cannot meet personalized protection requirements in a multi-material collection coexistence scene. The method comprises the following steps: acquiring identity, material, state and real-time environment data of a collection through a sensing module; dynamically generating a temperature and humidity target interval of each collection based on the material-environment knowledge base and the degradation state; a plurality of independent microclimate areas are constructed in a single space by using micro-scale airflow guidance, a distributed humidity and temperature regulating unit and an airflow barrier, and accurate cooperative control is realized through multi-objective optimization scheduling and closed-loop feedback. The system comprises a collection information sensing module, a differentiation demand modeling module, a partition precise execution module and a central cooperative control module. According to the method, refined environment regulation and control of one object and one strategy are realized, and the pertinence, the adaptability and the energy efficiency ratio of cultural relic protection are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent environmental control technology, and in particular to an intelligent temperature and humidity control system and method for differentiating the temperature and humidity environment of collections. Background Technology

[0002] With the increasing demands for cultural relic preservation and display, collections are placing higher requirements on the stability of temperature and humidity in their preservation environment. Different materials, ages, and types of collections (such as paper artifacts, textiles, metal objects, and organic artifacts) have different optimal temperature and humidity ranges for preservation. A single, constant environmental parameter is insufficient to meet the preservation needs of multiple types of collections coexisting in the same environment. Therefore, there is an urgent need for an intelligent system capable of implementing differentiated and precise temperature and humidity control based on the characteristics of different collections.

[0003] A search revealed a patent, CN105757912B, for a rapid and precise temperature and humidity control system for the microenvironment of a museum display case, published on November 30, 2018. This patent utilizes a compressor-based humidity control and water extraction module, a semiconductor humidity control module, a humidification module, and a temperature control module working in tandem to achieve rapid and precise control of the microenvironment within the display case. However, this system only provides uniform control for the overall environment of a single display case, failing to consider the varying temperature and humidity requirements of different artifacts within the case. It cannot provide suitable local microenvironments for artifacts of different materials within the same space. Furthermore, its control logic is based on fixed threshold feedback, lacking intelligent recognition and adaptive adjustment capabilities based on artifact type, material sensitivity, and historical environmental data, making it difficult to achieve precise protection tailored to each artifact.

[0004] A search revealed a distributed intelligent control system with publication number CN115202419B, published on February 14, 2025. This patent is applied to the storage environment of museum artifacts. It reduces temperature and humidity fluctuations caused by frequent start-ups and shutdowns of constant temperature and humidity units through the linkage of distributed sensors and control equipment. It also introduces air-guiding light pipes to monitor the airflow status of the ducts to improve control stability. Although the system achieves continuous environmental control and anomaly warning functions, it still aims for "overall regional stability" and does not establish a mapping relationship between individual artifact attributes and environmental parameters. It cannot dynamically allocate differentiated temperature and humidity strategies based on artifact category, preservation status, or fragility. Furthermore, the system lacks technical means for independent control of local microclimates in environments with multiple coexisting objectives, making it difficult to achieve parallel optimization of multiple parameters within a limited space.

[0005] The aforementioned problems indicate that while existing museum environmental control technologies have improved in terms of accuracy, stability, and response speed, they still have significant shortcomings in areas such as identifying the differentiated needs of collections, independently controlling multi-objective local microenvironments, and generating adaptive control strategies based on collection characteristics. Therefore, this invention proposes an "Intelligent Temperature and Humidity Environment Control System and Method for Differentiated Collection Environments," aiming to achieve intelligent, personalized, and highly compatible temperature and humidity management of environments where multiple types of collections coexist through collection information perception, differentiated needs modeling, and precise zoning execution, thereby enhancing the scientific nature and system efficiency of cultural relic preservation. Summary of the Invention

[0006] This invention provides an intelligent temperature and humidity control system and method for differentiated collection environments, aiming to solve the technical problem that existing uniform temperature and humidity control strategies in museums or cultural relic preservation environments cannot meet the personalized protection needs of scenarios where multiple types of collections coexist, due to differences in the materials, ages, and types of collections. This invention constructs a three-layer collaborative architecture consisting of a collection information perception layer, a differentiated needs modeling layer, and a zoned precise execution layer. This enables the independent identification, dynamic modeling, and closed-loop control of multiple local microclimate zones within display cases or storage spaces, thereby providing a temperature and humidity environment adapted to the material characteristics of different collections simultaneously within a single physical space.

[0007] As one embodiment of the present invention, the intelligent temperature and humidity control system for differentiated collection environments includes a collection information sensing module, a differentiated demand modeling module, a zoned precise execution module, and a central collaborative control module.

[0008] The collection information sensing module is used to acquire the identity, material category, age information, preservation status assessment data, and real-time environmental parameters of each collection within the display case or storage space. This module consists of a non-contact RFID reader deployed inside the display case, a hyperspectral imaging unit, a distributed temperature and humidity sensor array, and a collection status visual analysis unit. The non-contact RFID reader reads passive RFID tags attached to the collection's shelf to obtain the collection's unique identification code. The hyperspectral imaging unit analyzes the reflectance of the surface material of the collection through continuous spectral reflectance analysis in the visible to near-infrared bands, identifying the material category. The distributed temperature and humidity sensor array consists of multiple miniature temperature and humidity sensors, each corresponding to a preset local microclimate monitoring point, with a sampling frequency of no less than once per second and an accuracy of ±0.1 degrees Celsius for temperature and ±1% for relative humidity. The collection status visual analysis unit uses a high-resolution industrial camera with a polarized light source to acquire images of the collection's surface and quantifies deterioration characteristics such as cracks, mold, and fading based on a convolutional neural network model, generating preservation status assessment data.

[0009] The differentiated demand modeling module is used to generate the target temperature and humidity range and dynamic tolerance boundary for each collection based on the data output by the collection information perception module. This module includes a material-environment mapping knowledge base, a state sensitivity weight calculation unit, and a target range dynamic generation unit. The material-environment mapping knowledge base stores a predefined correspondence table between material categories and standard temperature and humidity storage ranges. Paper artifacts correspond to temperatures of 18-22 degrees Celsius and relative humidity of 45-55%, textiles to 16-20 degrees Celsius and 55-65% relative humidity, metal artifacts to 20-24 degrees Celsius and 30-40% relative humidity, and organic artifacts to 14-18 degrees Celsius and 55-65% relative humidity. The state sensitivity weight calculation unit calculates the sensitivity coefficient of the current artifact to temperature and humidity fluctuations based on the deterioration score output by the artifact's state visual analysis unit. This coefficient ranges from 0.8 to 1.2, with higher deterioration levels resulting in a coefficient closer to 1.2. The target range dynamic generation unit multiplies the standard range in the material-environment mapping knowledge base by the state sensitivity weight coefficient to generate the current artifact's dynamic temperature and humidity target range and upper and lower tolerance boundaries.

[0010] The zonal precision execution module is used to establish multiple isolated or semi-isolated local microclimate control units within a physical space, and independently adjust the temperature and humidity parameters within each unit according to the target range output by the differentiated needs modeling module. This module consists of a microscale airflow guiding structure, a distributed humidity control actuator array, a local temperature control unit array, and an airflow barrier generation device. The microscale airflow guiding structure consists of multiple programmable micro-guide blades, installed at the air duct outlets at the top or bottom of the display case. A servo motor drives the adjustment of the blade angles to guide airflow directionally to cover designated collection areas. The distributed humidity control actuator array consists of several micro-semiconductor dehumidifiers / humidifiers, each actuator corresponding to a local microclimate zone. Its operating current is adjusted in real-time based on the deviation between the target humidity and the measured humidity using a proportional-integral-derivative control algorithm. The local temperature control unit array consists of micro-thermoelectric cooling elements attached to the bottom or side walls of the collection rack. A bidirectional DC power supply controls the switching between the hot and cold ends to achieve localized heating or cooling. The airflow barrier generating device uses an ultrasonic atomizer in conjunction with an electrostatic field generator to form a low-speed air curtain barrier between adjacent collection areas, suppressing air mixing between different microclimate zones. The air curtain wind speed is maintained between 0.3 and 0.5 meters per second.

[0011] The central collaborative control module coordinates the data interaction and control command issuance of the three modules mentioned above, ensuring the consistency and stability of the overall system operation. This module includes a multi-objective optimization scheduling engine, a conflict resolution logic unit, and a historical environment database. The multi-objective optimization scheduling engine receives the dynamic temperature and humidity target ranges of all artifacts, constructs a multi-constraint optimization problem, and sets the objective function to minimize the sum of the squared weighted deviations between the actual environmental parameters of each local area and the target range. Constraints include a total energy consumption limit, maximum equipment power limit, and airflow barrier stability requirements. A sequential quadratic programming algorithm is used to solve for the optimal control command set. The conflict resolution logic unit handles situations where the target ranges of adjacent artifacts overlap or contradict each other. When the difference in target humidity between two adjacent areas exceeds 15%, the airflow barrier enhancement mode is automatically activated, and the coverage of the microscale airflow guidance structure is adjusted to expand the isolation distance. The historical environment database continuously records the historical temperature and humidity time-series data, control command sequences, and artifact status change trends of each artifact's environment, used for subsequent correction of state sensitivity weight coefficients and iterative updates of the material-environment mapping knowledge base.

[0012] As one embodiment of the present invention, the intelligent control method for differentiated temperature and humidity environment of the collection includes the following steps: First, the collection information sensing module acquires the identification, material type, age information, preservation status assessment data, and real-time temperature and humidity data of each local microclimate monitoring point for all collections in the display case; Secondly, the identity identifier and material category are input into the material-environment mapping knowledge base to query the corresponding standard temperature and humidity storage range; at the same time, the state sensitivity weight coefficient is calculated based on the storage status assessment data, and the dynamic temperature and humidity target range and tolerance boundary of each collection are generated by combining the standard range. Next, based on the dynamic target range of all collections, a globally optimal partition control instruction set is generated through a multi-objective optimization scheduling engine. The instruction set includes the blade angle of each microscale airflow guide structure, the working current of each distributed humidity control actuator, the direction and magnitude of the thermoelectric power of each local temperature control unit, and the start / stop status and intensity parameters of the airflow barrier generation device. Then, the control instruction set is sent to the partition precision execution module to drive the synchronous action of each execution unit, forming multiple independent and controllable local microclimate zones inside the display case; Finally, real-time temperature and humidity data of each local area are continuously collected and compared with the dynamic target range. If the deviation exceeds the tolerance boundary, the feedback correction mechanism is triggered to recalculate and issue the corrected control command to achieve closed-loop control.

[0013] In a preferred embodiment of the present invention, the hyperspectral imaging unit has a spectral resolution of five nanometers, a spectral range of four hundred to one thousand nanometers, and a spatial resolution of 0.5 millimeters per pixel. By combining principal component analysis and support vector machine classifier, the material category is determined, and the classification accuracy is not less than 95%.

[0014] In a preferred embodiment of the present invention, the spacing between adjacent sensors in the distributed temperature and humidity sensing array is no more than 30 centimeters, and each sensor is equipped with an independent timestamp synchronization circuit to ensure that the time alignment error of all sensing data is less than 10 milliseconds.

[0015] In a preferred embodiment of the present invention, the number of guide vanes in the microscale airflow guiding structure is not less than eight, the rotation angle adjustment range of each vane is from -90 degrees to +90 degrees, the angle positioning accuracy is ±1 degree, and the response time is less than 500 milliseconds.

[0016] In a preferred embodiment of the present invention, the air curtain generated by the airflow barrier generating device has a thickness of five to ten centimeters, covers the top to the bottom of the collection vertically, and the relative humidity gradient inside the air curtain is controlled to be no more than 0.5 percent per centimeter.

[0017] As a preferred embodiment of the present invention, the historical environment database adopts a time-series database architecture, supports efficient querying by collection identity identifier, time window, and environmental parameter type, and regularly performs data compression and archiving, retaining the original data for at least five years.

[0018] The beneficial effects of this invention are as follows: By constructing a three-layer collaborative mechanism of individual collection information perception, dynamic modeling of differentiated needs, and precise zoning execution, this invention, for the first time, achieves the simultaneous provision of independently adapted local microclimate environments for collections of various materials within the same physical display case or storage space; the system abandons the traditional "regionally constant" control paradigm and instead adopts a refined "one item, one policy" control strategy, significantly improving the pertinence and effectiveness of cultural relic protection; specifically, the fusion of hyperspectral imaging and radio frequency identification enables non-destructive and automatic identification of the material and identity of collections, avoiding errors from manual data entry; and the dynamic adjustment mechanism based on the sensitivity weights of the collection's deterioration state ensures that the target temperature and humidity range is... The invention can adaptively evolve with the health status of the collections, enhancing the foresight of the conservation strategy. The synergistic effect of the microscale airflow guiding structure and the airflow barrier generation device effectively isolates different microclimate zones within a limited space, solving the problem of crosstalk of environmental parameters under the coexistence of multiple objectives. The multi-objective optimization scheduling engine meets the differentiated needs of each collection while taking into account system energy consumption and equipment lifespan, achieving a balance between conservation effectiveness and operational economy. In summary, this invention not only solves the core defect of existing technologies that cannot achieve localized differentiated control, but also significantly improves the environmental management capabilities and scientific conservation level of museums and cultural relic preservation institutions in complex collection coexistence scenarios through intelligent and closed-loop control logic. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall technical solution architecture of the intelligent temperature and humidity control system and control method for differentiated collections proposed in this invention. Figure 2 This is a schematic diagram of the core principle framework of the differentiated demand modeling module in this invention; Figure 3 This is a logical flowchart of the collection information perception and dynamic target interval generation in this invention; Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow between the partitioned precise execution module and the central collaborative control module in this invention. Detailed Implementation

[0020] 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.

[0021] Specific implementation examples are given below.

[0022] This invention provides an intelligent temperature and humidity control system and method for differentiating the environment of collections. Its core lies in the independent local microclimate control of multiple types of collections within a single physical display case or storage space through a three-layer collaborative architecture: collection information perception layer, differentiated demand modeling layer, and zoned precise execution layer. The following will describe in detail the specific implementation of the method, focusing on the internal logic, data flow, parameter processing mechanism, and closed-loop control strategy of each step. The intelligent temperature and humidity control method for the collection includes the following steps: S1, acquiring the identification, material category, age information, preservation status assessment data, and real-time temperature and humidity data of each local microclimate monitoring point for all collections in the display case; S2, querying the standard temperature and humidity preservation range based on the identification and material category, and generating the dynamic temperature and humidity target range and tolerance boundary for each collection by combining the preservation status assessment data; S3, generating the globally optimal partition control instruction set through a multi-objective optimization scheduling engine based on the dynamic target range of all collections; S4, sending the control instruction set to the partition precision execution module to drive the synchronous action of each execution unit, forming multiple independent and controllable local microclimate areas inside the display case; S5, continuously collecting real-time temperature and humidity data of each local area, and triggering a feedback correction mechanism if the deviation exceeds the tolerance boundary, recalculating and sending the corrected control instruction to achieve closed-loop control; In step S1, the system first activates the collection information sensing module to complete comprehensive data collection of all collections inside the display case and their surrounding environment. This process is completed collaboratively by four sub-units: a non-contact RFID reader, a hyperspectral imaging unit, a distributed temperature and humidity sensor array, and a collection status visual analysis unit. The non-contact RFID reader is deployed on the top or side wall of the display case, operating at a frequency of 13.56 MHz, with a reading distance covering the entire interior space of the display case. When a collection is placed inside the display case, the passive RFID tag attached to its holder is activated, and the reader captures the unique identification code of the collection stored in the tag. This code is a 16-digit hexadecimal string used for subsequent data indexing and association. The spectral imaging unit is mounted above the display case, with its optical lens pointing vertically downwards at the bottom of the case. It covers a spectral range of 400 to 1000 nanometers, with a spectral resolution of 5 nanometers and a spatial resolution of 0.5 millimeters per pixel. This unit continuously acquires hyperspectral cubic data of the artifact's surface at a rate of two frames per second and transmits the raw data to a local image processing unit. The image processing unit first performs radiometric calibration and dark current correction on the hyperspectral data, then extracts the average reflectance spectral curve for each region of the artifact and inputs it into a pre-trained support vector machine classifier. This classifier constructs a decision boundary based on the first ten principal components after dimensionality reduction using principal component analysis and outputs a material category determination, including paper artifacts, textiles, metal objects, or organic materials. The cultural relics are categorized into four classes, with a classification accuracy of no less than 95%. The distributed temperature and humidity sensor array consists of no fewer than twelve miniature temperature and humidity sensors, each with a package size of 5 mm x 5 mm x 2 mm. The temperature measurement range is 0 to 50 degrees Celsius with an accuracy of ±0.1 degrees Celsius, and the humidity measurement range is 10% to 90% relative humidity with an accuracy of ±1% relative humidity. The sensors are evenly distributed in a grid pattern on the bottom of the display case, with the distance between adjacent sensors no greater than 30 cm, ensuring that each artifact corresponds to at least one dedicated monitoring point. All sensors are equipped with independent timestamp synchronization circuits. These circuits receive synchronization pulse signals from the central collaborative control module, ensuring that the data sampling of each sensor is synchronized. Alignment is achieved with a time error of less than ten milliseconds. The visual analysis unit for the collection's condition consists of a five-megapixel industrial camera and a ring-polarized light source. The camera lens has a focal length of 25 mm, an aperture of 2.8, and an exposure time of 20 milliseconds. This unit triggers an image acquisition before each adjustment cycle to obtain a high-resolution image of the collection's front. After preprocessing with denoising, contrast enhancement, and edge sharpening, the image is input into a convolutional neural network model. This model is a five-layer deep residual network that has been trained on a dataset containing 100,000 labeled images. It can quantify indicators such as crack density, mold area ratio, and color saturation decay, and output a degradation score between zero and one hundred. The higher the score, the worse the preservation condition. In step S2, the system inputs the artifact identification, material category, and deterioration score obtained in step S1 into the differentiated demand modeling module to generate the dynamic temperature and humidity target range for each artifact. This process consists of three sub-steps. First, the system retrieves the historical control records from the historical environment database based on the artifact identification. If it is the first time the artifact is placed in the cabinet, the material category is used as the query key. The material-environment mapping knowledge base is stored in read-only memory. Its content is a structured table with four rows and four columns: the first column is the material category name, the second and third columns are the lower and upper limits of the standard temperature, respectively, and the fourth and fifth columns are the lower and upper limits of the standard humidity, respectively. Specifically, paper artifacts correspond to a temperature of 18 to 22 degrees Celsius and a relative humidity of 45 to 55 percent; textiles correspond to a temperature of 16 to 20 degrees Celsius and a relative humidity of 55 to 65 percent; metal artifacts correspond to a temperature of 20 to 24 degrees Celsius and a relative humidity of 30 to 40 percent; and organic artifacts correspond to a temperature of 14 to 18 degrees Celsius and a relative humidity of 55 to 65 percent. The system matches the corresponding row based on the material category. First, the standard temperature and humidity range is extracted. Second, the state sensitivity weight calculation unit receives the degradation score and linearly maps it to the sensitivity coefficient. The mapping relationship is defined as follows: when the degradation score is zero, the coefficient is 0.8; when the score is 100, the coefficient is 1.2. The intermediate value is calculated by linear interpolation. This coefficient is used to scale the tolerance range of the standard range, reflecting the current tolerance of the collection to environmental fluctuations. Finally, the target range dynamic generation unit performs range adjustment calculations. For the temperature target range, the new lower limit is equal to the original lower limit minus the original upper limit minus the original lower limit multiplied by the coefficient minus one, and then divided. The new upper limit is equal to the original upper limit plus the original upper limit minus the original lower limit multiplied by the coefficient minus one, and then divided by two. The adjustment method for the humidity target range is the same. The dynamic temperature and humidity target range generated in this way has an asymmetric tolerance boundary. The higher the degree of deterioration, the narrower the range and the stricter the control requirements. For example, a paper cultural relic with a deterioration score of 80 has a sensitivity coefficient of 1.12. The standard temperature range is 18 to 22 degrees Celsius. The new range after adjustment is 17.56 to 22.44 degrees Celsius. The total tolerance width is only 4.88 degrees Celsius, which is 12% narrower than the standard range. In step S3, the central collaborative control module receives the dynamic temperature and humidity target ranges of all collections and activates the multi-objective optimization scheduling engine to generate a globally optimal control instruction set. This engine models the problem as a constrained nonlinear optimization problem. Assume there are N collections in the display case, and the target temperature of the i-th collection is... The target humidity is The measured temperature was The measured humidity was The objective function J is defined as the weighted sum of squares of the environmental deviations of each collection: ; Where w_i is the weight of the i-th item, and its value is the reciprocal of its sensitivity coefficient, reflecting that items with more severe deterioration should receive higher control priority; α and β are the dimensionless normalization coefficients of temperature and humidity, respectively, with values ​​of 0.01 and 0.0001, so that the two contributions are of the same order of magnitude; the constraint condition includes: the total humidity control power does not exceed P max The total temperature control power does not exceed Q. max The wind speed v, which is the airflow barrier between any two adjacent local microclimate regions. j Satisfying 0.3 m / s ≤ v j ≤0.5 m / s; Optimization variables include: blade angles of M microscale airflow guide structures. , The operating current of L distributed humidity control actuators , Thermoelectric power of R local temperature control units , ; and the atomization intensity of the S airflow barrier generating devices , The multi-objective optimization scheduling engine uses a sequential quadratic programming algorithm to solve this problem. The algorithm initializes all variables to zero and iteratively updates them until the objective function converges or the maximum number of iterations (fifty) is reached. In each iteration, the engine calls the physical simulation submodule to predict the temperature and humidity field distribution inside the display case under the current variable combination. This submodule is based on a simplified computational fluid dynamics model, dividing the display case into a three-dimensional mesh. The temperature and humidity changes in each mesh cell are jointly determined by the convection term, diffusion term, and source term. The source term is determined by the position and power of the humidity control actuator and the temperature control unit, while the convection term is jointly determined by the airflow guiding structure and the airflow barrier. The simulation results are used to calculate the objective function J and the degree of constraint violation, guiding the direction of variable updates. The final output control instruction set contains the precise control parameters of all actuators. In step S4, the zoned precision execution module receives and executes the control command set synchronously; each guide vane in the microscale airflow guiding structure is driven by an independent servo motor. After receiving the angle command, the motor rotates the vane to the designated position within 500 milliseconds, with an angle positioning accuracy of ±1 degree; the vane angle determines the direction vector of the outflow airflow, ensuring that the airflow precisely covers the target collection area; each semiconductor dehumidifier / humidifier in the distributed humidity control actuator array adjusts its condensation or evaporation rate according to the working current command; the current value is calculated in real time using a proportional-integral-derivative control algorithm, with the formula as follows: ; in, The difference between the target humidity and the measured humidity in the l-th region. , , The proportional, integral, and differential gains are respectively set to 0.5, 0.05, and 0.1. Each thermoelectric cooler in the local temperature control unit array receives a bidirectional DC power supply command. The polarity of the power supply determines the switching between the hot and cold ends, thereby achieving heating or cooling. The power level is directly set by the thermoelectric power command, ranging from zero to five watts. The airflow barrier generating device adjusts the vibration frequency of the ultrasonic atomizer and the voltage of the electrostatic field generator according to the atomization intensity command to generate a vertical air curtain with a thickness of five to ten centimeters. The air curtain covers the boundary between adjacent collection areas, and the relative humidity gradient inside is controlled to be no more than 0.5% per centimeter, effectively suppressing lateral air mixing. In step S5, the system enters the closed-loop monitoring stage; the distributed temperature and humidity sensor array continuously collects data at a frequency of once per second, and the central collaborative control module compares the measured values ​​with the dynamic target range in real time; if the measured temperature or humidity of any collection exceeds its tolerance boundary, the feedback correction mechanism is immediately triggered; this mechanism first determines whether the duration of the deviation exceeds a preset threshold of ten seconds to eliminate instantaneous disturbances; if it is confirmed to be a continuous deviation, the optimization process of step S3 is re-executed, but this optimization uses the current environmental state as the initial condition and adds a penalty term to accelerate convergence; the corrected control instruction set is issued, and the execution module performs fine-tuning; at the same time, the historical environment database records the deviation event, control response time, and final stable state for subsequent state sensitivity weight coefficient correction; for example, if a collection frequently exceeds the humidity limit, the system will automatically increase its sensitivity coefficient in the next modeling to further narrow the target range; The above method achieves differentiated and refined temperature and humidity control of the coexistence environment of multiple types of collections through strict step synchronization, precise parameter calculation and closed-loop feedback mechanism; the system constructs multiple isolated local microclimate zones at the physical level, establishes a dynamic mapping relationship between collection attributes and environmental parameters at the logical level, and achieves multi-objective collaborative optimization at the control level, thereby solving the fundamental defect of existing technologies that cannot take into account both diversity and accuracy.

[0023] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A differentiated temperature and humidity environment intelligent control system for collections, characterized in that: include: The collection information sensing module is used to acquire the identity, material category, age information, preservation status assessment data, and real-time temperature and humidity data of each collection in the display case or storage space. The differentiated demand modeling module is used to input the identity identifier and material category into the material-environment mapping knowledge base to query the corresponding standard temperature and humidity storage range, and calculate the state sensitivity weight coefficient based on the storage state evaluation data, and generate the dynamic temperature and humidity target range and tolerance boundary for each collection by combining the standard temperature and humidity storage range. The zoned precision execution module is used to receive the zoned control instruction set and drive the microscale airflow guiding structure, distributed humidity control actuator array, local temperature control unit array and airflow barrier generation device to operate synchronously, forming multiple isolated local microclimate zones inside the display case or storage space. The central collaborative control module is used to construct and solve a multi-constraint optimization problem based on the dynamic temperature and humidity target range of all collections through a multi-objective optimization scheduling engine to generate the zonal control instruction set. When the measured value of any local area exceeds the tolerance boundary, a feedback correction mechanism is triggered to regenerate and issue the corrected control instruction.

2. The intelligent temperature and humidity control system for differentiated environments of collections according to claim 1, characterized in that, The collection information sensing module includes: A non-contact RFID reader is used to read passive RFID tags attached to collection racks to obtain the unique identification code of the collection; The hyperspectral imaging unit is used to acquire hyperspectral cubic data of the surface of the artifacts and output the material category determination results. A distributed temperature and humidity sensor array is used to acquire temperature and humidity data at each local microclimate monitoring point at a sampling frequency of no less than once per second. The visual analysis unit for the condition of artifacts is used to acquire high-resolution images of the artifact surface and output a deterioration score as data for assessing the preservation status.

3. A method for intelligent control of differentiated temperature and humidity environment for collections, characterized in that, include: The collection information sensing module obtains the identification, material category, age information, preservation status assessment data, and real-time temperature and humidity data of each collection in the display case or storage space. The identity identifier and material category are input into the material-environment mapping knowledge base to query the corresponding standard temperature and humidity storage range. The state sensitivity weight coefficient is calculated based on the storage status assessment data. The dynamic temperature and humidity target range and tolerance boundary of each collection are generated by combining the standard temperature and humidity storage range. Based on the dynamic temperature and humidity target range of all collections, a multi-constraint optimization problem is constructed and solved through a multi-objective optimization scheduling engine to generate a globally optimal zonal control instruction set. The zonal control instruction set includes the blade angle of the microscale airflow guide structure, the working current of the distributed humidity control actuator, the direction and magnitude of the thermoelectric power of the local temperature control unit, and the start / stop status and intensity parameters of the airflow barrier generation device. The zonal control instruction set is sent to the zonal precision execution module to drive the microscale airflow guiding structure, the distributed humidity control actuator array, the local temperature control unit array and the airflow barrier generation device to operate synchronously, forming multiple isolated local microclimate zones inside the display case or storage space. Real-time temperature and humidity data of each local microclimate region are continuously collected. If the measured value of any region exceeds the tolerance boundary of its corresponding dynamic temperature and humidity target range, the feedback correction mechanism is triggered, multi-objective optimization scheduling is re-executed, and the corrected control command is issued to achieve closed-loop control.

4. The method for intelligent control of differentiated temperature and humidity environment of collections according to claim 1, characterized in that, The collection information sensing module acquires the identification, material type, age information, preservation status assessment data, and real-time temperature and humidity data of each collection item within the display case or storage space, including: The passive RFID tag attached to the collection rack is read by a non-contact RFID reader to obtain the unique identification code of the collection; Hyperspectral cubic data of the surface of the artifacts were acquired by a hyperspectral imaging unit. After radiometric calibration and dark current correction, the average reflectance spectral curve was extracted and input into a pre-trained support vector machine classifier to output the material category determination result. Temperature and humidity data of each local microclimate monitoring point are acquired by a distributed temperature and humidity sensor array at a sampling frequency of no less than once per second, wherein the distance between adjacent sensors is no more than 30 centimeters and the timestamp synchronization error is less than 10 milliseconds. High-resolution images of the artifact's surface are acquired through a visual analysis unit for artifact condition. After image preprocessing, the images are input into a convolutional neural network model to output a deterioration score as data for assessing the artifact's preservation status.

5. The method for intelligent control of differentiated temperature and humidity environment of collections according to claim 3, characterized in that, The identity identifier and material category are input into the material-environment mapping knowledge base to query the corresponding standard temperature and humidity storage range. Based on the storage status assessment data, a state sensitivity weight coefficient is calculated. Combined with the standard temperature and humidity storage range, a dynamic temperature and humidity target range and tolerance boundary are generated for each artifact, including: Based on the material category, the corresponding standard temperature range and standard humidity range are extracted from the material-environment mapping knowledge base. Among them, paper cultural relics correspond to a temperature of 18 to 22 degrees Celsius and a relative humidity of 45 to 55%, textiles correspond to a temperature of 16 to 20 degrees Celsius and a relative humidity of 55 to 65%, metal objects correspond to a temperature of 20 to 24 degrees Celsius and a relative humidity of 30 to 40%, and organic cultural relics correspond to a temperature of 14 to 18 degrees Celsius and a relative humidity of 55 to 65%. The degradation score is linearly mapped to the state sensitivity weight coefficient, with a mapping range of 0.8 to 1.

2. The standard temperature range and standard humidity range are asymmetrically scaled. The scaling formula is as follows: the lower limit of the new range is equal to the original lower limit minus the original upper limit minus the original lower limit multiplied by the coefficient minus one and then divided by two. The upper limit of the new range is equal to the original upper limit plus the original upper limit minus the original lower limit multiplied by the coefficient minus one and then divided by two, thus generating the dynamic temperature and humidity target range and tolerance boundary.

6. The method for intelligent control of differentiated temperature and humidity environment of collections according to claim 5, characterized in that, Based on the dynamic temperature and humidity target ranges of all collections, a multi-constraint optimization problem is constructed and solved using a multi-objective optimization scheduling engine, generating a globally optimal set of zonal control instructions, including: The objective function is to minimize the sum of squared weighted deviations between the actual environmental parameters of each local area and the dynamic target interval, with the weights being the reciprocals of the state sensitivity weight coefficient. The constraints include that the total humidity control power does not exceed the preset upper limit, the total temperature control power does not exceed the preset upper limit, and the wind speed of the airflow barrier between any adjacent local microclimate areas is maintained between 0.3 meters and 0.5 meters per second; The sequential quadratic programming algorithm is used to solve the problem iteratively. In each iteration, the physical simulation submodule based on the simplified computational fluid dynamics model is called to predict the temperature and humidity field distribution under the current combination of control parameters, until the objective function converges or the maximum number of iterations is reached.

7. The method for intelligent control of differentiated temperature and humidity environment of collections according to claim 6, characterized in that, The zonal control instruction set is sent to the zonal precision execution module, driving the microscale airflow guiding structure, distributed humidity control actuator array, local temperature control unit array, and airflow barrier generation device to operate synchronously, forming multiple mutually isolated local microclimate zones within the display case or storage space, including: The programmable micro-guide blades in the micro-scale airflow guiding structure are controlled to rotate to a specified angle with an angle positioning accuracy of ±1 degree and a response time of less than 500 milliseconds, so as to guide the airflow to cover the specified collection area. The condensation or evaporation rate of each semiconductor dehumidifier / humidifier plate in the distributed humidity control actuator array is adjusted according to the working current command. The working current is calculated in real time through a proportional-integral-derivative control algorithm. The polarity and power of the bidirectional DC power supply of each thermoelectric cooling element in the local temperature control unit array are controlled to achieve local heating or cooling, with a power range of zero to five watts; Adjust the vibration frequency of the ultrasonic atomizer and the voltage of the electrostatic field generator of the airflow barrier generating device to generate a vertical air curtain with a thickness of five to ten centimeters. The relative humidity gradient inside the air curtain is controlled to be no more than 0.5 percent per centimeter.

8. The method for intelligent control of differentiated temperature and humidity environment of collections according to claim 7, characterized in that, Real-time temperature and humidity data of various local microclimate zones are continuously collected. If the measured value of any zone exceeds the tolerance boundary of its corresponding dynamic temperature and humidity target range, a feedback correction mechanism is triggered. Multi-objective optimization scheduling is re-executed and corrected control commands are issued to achieve closed-loop control, including: Determine if the duration of the deviation exceeds ten seconds to eliminate instantaneous disturbances; If the deviation is confirmed to be persistent, the multi-objective optimization scheduling is re-executed with the current environmental state as the initial condition, and a penalty term is added to the objective function to accelerate convergence. The revised control instructions are sent to the precise execution module of the partition for fine-tuning, and the deviation event, control response time and stable state are recorded in the historical environment database for subsequent correction of the state sensitivity weight coefficient.

9. The method for intelligent control of differentiated temperature and humidity environment of collections according to claim 8, characterized in that, The hyperspectral imaging unit has a spectral resolution of five nanometers, a spectral range of four hundred to one thousand nanometers, and a spatial resolution of 0.5 millimeters per pixel. It uses principal component analysis and support vector machine classifier to jointly determine the material category, achieving a classification accuracy of no less than 95%.

10. The method for intelligent control of differentiated temperature and humidity environment of collections according to claim 9, characterized in that, The microscale airflow guiding structure has no fewer than eight guide vanes, and the rotation angle adjustment range of each vane is from -90 degrees to +90 degrees.

Images