An Internet of Things management method and platform for museum collections

Abstract

The present invention discloses an Internet of Things management method and platform for museum collections, and relates to the technical field of collection management. The method includes: establishing a mapping association between museum collections and Internet of Things devices; connecting the Internet of Things device communication network to obtain a sensor data set; importing each collection sensor data block into an edge node and a cloud server respectively, the edge node performs real-time tracking and early warning according to a preset threshold, and the cloud server performs risk assessment according to the trend prediction result, and provides early warning feedback according to the risk assessment result. The present invention solves the technical problem that it is impossible to effectively combine environmental changes and positioning information for real-time monitoring and risk early warning, resulting in delayed collection management response and inaccurate risk assessment, and achieves the technical effect of realizing real-time monitoring of the collection status through the Internet of Things, and performing risk assessment and early warning feedback, thereby improving the efficiency and safety of collection management.

Description

Technical Field

[0001] The present invention relates to the technical field of collection management, and in particular to an Internet of Things management method and platform for museum collections. Background Art

[0002] Existing museum collection management technology mainly relies on manual records and periodic inspections. This traditional management method has many shortcomings when faced with a large number of collections and complex environmental conditions. First, the frequency of manual inspections is limited, and it is impossible to achieve real-time monitoring of the status of the collections, which can easily miss the risks of collections caused by environmental changes, physical damage and other factors. Secondly, the environmental conditions in the museum (such as temperature, humidity, light, vibration, etc.) directly affect the preservation quality of the collections, and traditional management methods are difficult to respond in time when environmental conditions are abnormal. In addition, for some fragile or valuable collections, it is difficult for traditional management methods to track their specific location and movement status in real time, resulting in the inability to quickly take effective measures in emergencies (such as accidental movement or theft), further increasing the risk of damage or loss of collections. Summary of the invention

[0003] The present application provides an Internet of Things management method and platform for museum collections, which is used to solve the technical problem that it is impossible to effectively combine environmental changes and positioning information for real-time monitoring and risk warning, resulting in delayed collection management response and inaccurate risk assessment.

[0004] In view of the above problems, the present application provides an Internet of Things management method and platform for museum collections.

[0005] In a first aspect of the present application, an Internet of Things management method for museum collections is provided, the method comprising:

[0006] A mapping association is established between museum collections and IoT devices to indicate the monitoring correspondence between museum collections and IoT devices, wherein the IoT devices include positioning sensor devices and environmental sensor devices; the IoT devices are connected to a communication network to obtain a sensor data set, wherein the sensor data set includes a positioning data set and an environmental parameter data set, wherein the sensor data set has a collection timestamp, and each collection sensor data block is constructed according to the mapping relationship; each collection sensor data block is imported into an edge node and a cloud server respectively, wherein the edge node performs real-time tracking and early warning according to a preset threshold value based on the positioning data set and the environmental parameter data set, and the cloud server performs trend prediction based on the positioning data set and the environmental parameter data set, performs risk assessment based on the trend prediction results, and performs early warning feedback based on the risk assessment results.

[0007] The second aspect of the present application provides an Internet of Things management platform for museum collections, the platform comprising:

[0008] A mapping association establishment module, the mapping association establishment module is used to establish a mapping association between museum collections and Internet of Things devices, and is used to represent the monitoring correspondence between museum collections and Internet of Things devices, wherein the Internet of Things devices include positioning sensor devices and environmental sensor devices; a sensor data acquisition module, the sensor data acquisition module is used to connect to the Internet of Things device communication network, obtain a sensor data set, wherein the sensor data set includes a positioning data set and an environmental parameter data set, and the sensor data set has a collection timestamp, and constructs each collection sensor data block according to the mapping relationship; an early warning feedback module, the early warning feedback module is used to import each collection sensor data block into an edge node and a cloud server respectively, the edge node performs real-time tracking and early warning according to the positioning data set and the environmental parameter data set according to a preset threshold, the cloud server performs trend prediction according to the positioning data set and the environmental parameter data set, performs risk assessment according to the trend prediction results, and performs early warning feedback according to the risk assessment results.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] Establish a mapping association between museum collections and IoT devices to indicate the monitoring correspondence between museum collections and IoT devices, wherein the IoT devices include positioning sensor devices and environmental sensor devices; connect the IoT device communication network to obtain sensor data sets, wherein the sensor data sets include positioning data sets and environmental parameter data sets, and the sensor data sets have acquisition timestamps, and construct sensor data blocks for each collection according to the mapping relationship; import the sensor data blocks for each collection into edge nodes and cloud servers respectively, wherein the edge nodes perform real-time tracking and early warning according to preset thresholds based on the positioning data sets and environmental parameter data sets, and the cloud servers perform trend prediction based on the positioning data sets and environmental parameter data sets, perform risk assessment based on the trend prediction results, and provide early warning feedback based on the risk assessment results. The technical effect of achieving real-time monitoring of the collection status through the IoT, performing risk assessment and early warning feedback, and improving the efficiency and safety of collection management is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic diagram of a flow chart of an Internet of Things management method for museum collections provided in an embodiment of the present application;

[0013] Figure 2 A schematic diagram of the structure of an Internet of Things management platform for museum collections provided in an embodiment of the present application.

[0014] Explanation of the accompanying drawings: mapping association establishment module 1, sensor data acquisition module 2, early warning feedback module 3. DETAILED DESCRIPTION

[0015] This application provides an Internet of Things management method and platform for museum collections, which is used to solve the technical problem that it is impossible to effectively combine environmental changes and positioning information for real-time monitoring and risk warning, resulting in delayed collection management response and inaccurate risk assessment.

[0016] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0017] Embodiment 1, as Figure 1 As shown, the present application provides an Internet of Things management method for museum collections, the method comprising:

[0018] A mapping association between museum collections and IoT devices is established to represent the monitoring correspondence between museum collections and IoT devices, wherein the IoT devices include positioning sensor devices and environmental sensor devices.

[0019] Specifically, in order to achieve accurate monitoring of museum collections, a mapping association between museum collections and IoT devices is first established to represent the monitoring correspondence between museum collections and IoT devices. Among them, IoT devices include positioning sensing devices and environmental sensing devices. Positioning sensing devices can use RFID, UWB (ultra-wideband) or Bluetooth beacons, etc., through which the location changes of museum collections can be tracked in real time to ensure accurate monitoring of the movement status of collections. RFID tags can be attached to collections or display cabinets, and use their low power consumption and long-distance communication characteristics to perform accurate positioning and identification; UWB can achieve high-precision position detection, which is suitable for collection monitoring scenarios that require higher positioning accuracy; Bluetooth beacons can be used for large-scale indoor positioning to transmit the location information of collections in real time. Environmental sensing devices include temperature sensors, humidity sensors and light sensors, which are used to monitor environmental parameters in the museum in real time. Temperature sensors can detect temperature changes around collections to ensure that collections are in a suitable storage temperature range; humidity sensors are used to monitor environmental humidity to prevent damage to collections caused by excessively high or low humidity; light sensors can monitor the light intensity around collections to prevent fading or deterioration of collections due to excessive light. Through the comprehensive application of the above-mentioned positioning sensing equipment and environmental sensing equipment, all-round monitoring of the location and environment of museum collections can be achieved, ensuring accurate management and protection of the preservation environment and movement status of the collections.

[0020] Connect to the IoT device communication network to obtain a sensor data set, wherein the sensor data set includes a positioning data set and an environmental parameter data set. The sensor data set has a collection timestamp, and constructs sensor data blocks for each collection according to a mapping relationship.

[0021] Specifically, in order to achieve comprehensive monitoring and management of museum collections, various sensor data sets are obtained in real time by connecting the installed IoT devices to the communication network, including the location data and environmental parameter data of the collections, which are calibrated with correction coefficients. Among them, the location data set includes the real-time location information collected by sensor devices such as RFID, UWB or Bluetooth beacons, and the environmental parameter data set includes environmental parameter information monitored by sensors such as temperature, humidity, and light. In order to ensure the timeliness of the data, the IoT device adds a collection timestamp to all sensor data when collecting data to mark the collection time point of the data. Subsequently, according to the mapping relationship between the museum collections and the IoT devices, the obtained location data set, environmental parameter data set and timestamp are integrated to construct the sensor data blocks of each collection. Each data block includes location data, environmental parameter data and the corresponding timestamp, in the form of {location data, environmental data, timestamp}. Among them, in order to ensure the security of the data, especially for collections with a higher security level, the data blocks are encrypted, and the above data are encrypted using symmetric or asymmetric encryption algorithms. The encrypted data can not only prevent unauthorized access, but also ensure the integrity and confidentiality of the data during transmission. Taking AES symmetric encryption as an example, first select a unique symmetric encryption key. The length of the key can be 128 bits, 192 bits, or 256 bits. Usually, for scenarios with higher security, such as the management of valuable collections, a 256-bit key can be selected. This key needs to be kept confidential because both encryption and decryption rely on it. In order to further protect the sensor data block, the sensor data block is used as the object of encryption to form a complete plaintext sensor data. Subsequently, the AES encryption algorithm is used to process fixed-size blocks (128 bits per block). If the sensor data block length is not an integer multiple of the block size, the sensor data block is automatically padded to ensure that the sensor data block can be completely processed by the AES algorithm. In order to enhance the security of encryption, the CBC (cipher block chaining) mode is used to generate a random initialization vector (IV). This vector is used to encrypt the first block of data and guide the data chain in the subsequent encryption process to ensure that the same data generates different ciphertexts each time it is encrypted to prevent repeated attacks. After that, the plaintext sensor data is encrypted using the preset key and IV. During this process, the sensor data block will be converted into unreadable ciphertext. The encrypted ciphertext can effectively prevent unauthorized access or data leakage, ensuring the security of the sensor data block during transmission or storage. The encrypted sensor data block will be stored in the edge node and cloud server, or transmitted to other related nodes through the communication network. In addition, for collections with extremely high security requirements, blockchain technology can also be used to store each encrypted sensor data block in a distributed manner through blockchain technology.This not only ensures the integrity and tamper-proofness of the data, but also ensures the high security of data during transmission within and outside the museum's network.

[0022] In a possible implementation manner, acquiring the sensor data set includes:

[0023] Obtain an installation relationship between the museum collection and the Internet of Things device, the installation relationship including direct installation and indirect installation, the direct installation means that the sensor device is directly installed on the collection, and the indirect installation means that the sensor device is not directly installed on the collection but installed on an auxiliary structure; if the installation relationship is the indirect installation, locate the relative installation distance and relative installation angle between the museum collection and the auxiliary structure; determine a correction coefficient based on the relative installation distance and relative installation angle combined with the acquisition accuracy of the Internet of Things device; if the installation relationship is the direct installation, determine a correction coefficient based on the acquisition accuracy of the Internet of Things device; use the correction coefficient to correct the sensor data collected by the Internet of Things device and add it to the sensor data set.

[0024] Specifically, in order to more accurately monitor the status of museum collections, the installation relationship between museum collections and IoT sensor devices is first determined. The installation relationship is divided into two situations: direct installation and indirect installation. Direct installation means that the sensor device is directly attached to the collection, which is usually applicable to ordinary collections. Indirect installation means that due to the shape, material or its high preciousness of some collections, the sensor device cannot be directly installed, and the sensor must be installed with the help of auxiliary structures (such as brackets, bases, display cabinets, etc.). This method is usually used for collections that are particularly precious or cannot be directly touched. When the sensor device is installed through the auxiliary structure, because there are distance and angle differences between the sensor and the collection itself, these factors will affect the accuracy of the collected data. Therefore, by measuring the relative position of the collection and the auxiliary structure, the relative installation distance and relative installation angle between the collection and the auxiliary structure are obtained, and the relative installation distance and relative installation angle obtained are combined with the collection accuracy of the IoT sensor device arranged on the auxiliary structure, and input into the pre-constructed corresponding correction coefficient calculation formula to calculate the correction coefficient of the indirectly installed IoT sensor device. When the sensor device is directly installed on the collection, since there is no influence of distance or angle, it is only necessary to input the acquisition accuracy of the IoT sensor device into the pre-built corresponding correction coefficient calculation formula to calculate the correction coefficient of the directly installed IoT sensor device. After obtaining the correction coefficients of direct installation and indirect installation, the correction coefficient of direct installation is multiplied by the data collected by the directly installed IoT sensor device, and the correction coefficient of indirect installation is multiplied by the data collected by the indirectly installed IoT sensor device, and then the corrected data is added to the sensor data set. Each data entry in the sensor data set includes the corrected positioning data, the corrected environmental data, and the corresponding acquisition timestamp to ensure the integrity and consistency of the data for subsequent analysis, trend prediction, and early warning.

[0025] In a possible implementation, determining the correction coefficient includes:

[0026] For the direct installation, the correction factor calculation formula is expressed as: ; For the indirect installation, the correction coefficient calculation formula is expressed as: ,in, To install the correction factor directly, is the indirect installation correction factor, is the acquisition accuracy of the sensor equipment, To install the correction factor directly, , d is the relative installation distance, is the relative installation angle, k is the proportional constant used to adjust the proportions of different devices, and k is determined by experimental measurement or device specifications.

[0027] Specifically, for direct installation, the correction factor calculation formula is as follows: ;in, is the correction factor for direct installation. It is the acquisition accuracy of the sensor device, reflecting the inherent error of the sensor device. For indirect installation, the correction coefficient calculation formula is as follows: ;in, is the correction factor for indirect installation. is the distance influence coefficient, which reflects the influence of distance on the sensing data. The coefficient calculation formula is: , k in the formula is a proportional constant used to adjust the differences between different devices. This constant is determined by experimental measurement or device specifications. d is the relative installation distance, that is, the physical distance between the sensor and the collection. The farther the distance, the greater the error in the data collected by the sensor. It is the relative installation angle, that is, the angle difference between the sensor and the collection. The larger the angle, the greater the error in the data collected by the sensor.

[0028] The sensor data blocks of each collection are respectively imported into the edge node and the cloud server. The edge node performs real-time tracking and early warning according to the preset thresholds based on the positioning data set and the environmental parameter data set. The cloud server performs trend prediction based on the positioning data set and the environmental parameter data set, performs risk assessment according to the trend prediction results, and provides early warning feedback according to the risk assessment results.

[0029] Specifically, after obtaining the sensor data blocks of each collection, the horizontal alignment verification and the vertical alignment verification are performed on each collection sensor data block. When the verification is passed, the sensor data blocks of each collection are imported into the edge node and the cloud server respectively. The edge node is located in the local server inside or near the museum, responsible for real-time processing and monitoring. Since the edge node is close to the data source, it has the ability to respond quickly and can analyze the location data and environmental parameters of the collection in real time. The cloud server is used to centrally store and process data, and perform complex data analysis, trend prediction and remote risk assessment. In the edge node, the preset thresholds are set according to international standards and industry specifications. The preset thresholds include environmental parameter thresholds and safe displacement range thresholds. The environmental parameter threshold sets the safe range of environmental conditions such as temperature, humidity, and light. For example, for the storage environment of most collections, the temperature should be maintained at 18°C-22°C, and the humidity should be maintained between 40%-60%. The safe displacement range threshold sets the maximum allowable displacement range of the collection in the display or storage environment. For example, the safe displacement range of an exhibit is ±10 cm, and it is considered abnormal if it exceeds this range. When the edge node detects that the positioning data and environmental parameter data in the sensor data block of each collection exceeds the environmental parameter threshold or the safe displacement range threshold, the early warning mechanism is triggered immediately. The early warning information will be directly sent to the manager responsible for the collection to remind the manager to perform corresponding management operations, such as adjusting the environmental conditions or checking the status of the collection. In the cloud server, the cumulative impact calculation is performed based on the positioning data set and environmental parameter data set in the sensor data block of each collection to determine the cumulative effect of each monitoring parameter. Then, trend prediction is performed based on the obtained cumulative effect to generate trend prediction results. Subsequently, based on the trend prediction results, it is determined whether the collection has potential risks. For example, if the risk probability of the ambient temperature continues to increase and exceeds the set threshold, it will be marked as high risk and added to the risk assessment results. After that, based on the risk assessment results, early warning information is issued, including early warning notifications that the current status exceeds the threshold, to remind managers to take measures, such as adjusting equipment such as air conditioners and humidifiers, and checking whether there is external interference or other factors that cause the position of the collection to change, to prevent greater damage or safety issues. Through this distributed data processing and analysis method, rapid response can be achieved on the edge node, and in-depth analysis can be performed on the cloud server to ensure the safe monitoring and management of the collection.

[0030] In a possible implementation, each of the collection sensor data blocks is imported into the edge node and the cloud server respectively, including:

[0031] Obtain the collection timestamp of the collection sensor data block, and perform timestamp alignment verification on the sensor data received by the edge node according to the collection frequency of the Internet of Things device. The timestamp alignment verification includes horizontal alignment verification and vertical alignment verification. The horizontal alignment verification performs timeline and time interval alignment for the unified sensor device to ensure the collection timing stability of each sensor device; the vertical alignment verification performs synchronous collection timestamp alignment for all sensor data in the data block to ensure the synchronization of the collection of positioning data and environmental parameter data; when the horizontal alignment verification and vertical alignment verification meet the alignment verification conditions, the positioning data and environmental parameter data are imported into the edge node; the positioning data and environmental parameter data obtained by the edge node are transmitted to the cloud server at a regular time, and the timestamp alignment verification is performed on the received data of the cloud server and the transmitted data of the edge node, and the positioning data and environmental parameter data that meet the alignment verification are stored in the cloud server.

[0032] Specifically, when transmitting the sensor data blocks of each collection to the edge node and cloud server, it is crucial to ensure the timing consistency and reliability of the data. Through timestamp alignment verification, it can be ensured that the data has not been tampered with or lost during transmission and storage, and that the positioning data and environmental data can accurately reflect the actual status of the collection at the same time point. Timestamp alignment verification includes horizontal alignment verification and vertical alignment verification. For horizontal alignment verification, the timing consistency verification is mainly performed on the same sensor device. The data collection of each sensor device is carried out according to the set collection frequency. Therefore, when performing horizontal alignment, check whether the continuous data blocks collected by the same sensor device are collected according to the preset time interval. During the verification process, a time interval tolerance will be set. This time interval tolerance is set according to the collection frequency to ensure that the time interval of the data is consistent and there is no data duplication, expiration or invalidity. For example, if the collection frequency of a positioning sensor is to collect data twice every 1 second, it will check whether the difference between the timestamps of two adjacent data is within the time interval tolerance (±40 milliseconds) to ensure that the sensor collects data stably at the set frequency. For the vertical alignment verification, the data synchronization of multiple sensor devices (such as positioning sensors and environmental sensors) in each collection sensor data block is verified. Collection monitoring usually requires the simultaneous acquisition of data from multiple sensor devices, such as location, temperature, humidity, etc. Therefore, the vertical alignment verification can ensure that the data collected by multiple devices at the same time can reflect the comprehensive status of the collection. For example, when the timestamp of a positioning data is 12:00:00, the timestamp of its corresponding environmental parameter data should also be within a time point close to 12:00:00, that is, within the synchronization time tolerance, to verify whether the data of multiple devices are synchronously collected within a reasonable time range. When the horizontal alignment verification and the vertical alignment verification meet the alignment verification conditions, that is, when both meet the time interval tolerance or the synchronization time tolerance, each collection sensor data block is imported into the edge node. On the contrary, if the alignment verification conditions are not met, that is, when the timestamp difference between two adjacent data exceeds the time interval tolerance during horizontal alignment, the data will be marked as abnormal data and re-collection or alarm will be triggered. During the vertical alignment, if the difference between a certain positioning data timestamp and the corresponding environmental parameter data timestamp is not within the synchronization time tolerance, measures will be taken to correct it, such as data interpolation or re-collection. After completing the timestamp alignment verification, the positioning data and environmental parameter data in each collection sensor data block in the edge node will be regularly transmitted to the cloud server, and the cloud server will perform timestamp alignment verification again to ensure the timing consistency of the data. In this timestamp alignment verification, it will be checked whether the data timestamps of each collection sensor data block received by the cloud server are consistent with the data timestamps of each collection sensor data block transmitted by the edge node to prevent the data from being tampered with during the transmission process and ensure the integrity and accuracy of the sensor data.Once the timestamp alignment verification in the cloud server is passed, the positioning data and environmental parameter data in the sensor data block of each collection will be stored in the cloud server for subsequent trend prediction.

[0033] In a possible implementation, performing timestamp alignment verification on the received data of the cloud server and the transmitted data of the edge node includes:

[0034] According to the network transmission delay, an alignment verification tolerance coefficient is configured; and the alignment verification result is determined based on the alignment verification tolerance coefficient.

[0035] Specifically, during the transmission of sensor data from the edge node to the cloud server, network delay may be encountered. This delay may be caused by factors such as network bandwidth, packet processing time, and network path. Therefore, when performing time alignment verification of data transmission, it is necessary to consider network delay and set a reasonable tolerance range as the alignment verification tolerance coefficient. This alignment verification tolerance coefficient is set according to the actual situation of network transmission delay to determine whether the data received by the cloud server is consistent with the data transmitted by the edge node within a reasonable time range. For example, if the maximum value of the network transmission delay is determined to be 300 milliseconds through testing, in order to provide a certain tolerance, the alignment verification tolerance coefficient is set to 400 milliseconds, which can provide buffer space for network jitter or short-term delays. For each piece of sensor data, the timestamps at both ends are recorded during the transmission from the edge node to the cloud server. If the difference between the timestamps at both ends of each piece of sensor data is less than or equal to the alignment verification tolerance coefficient, it means that the sensor data is successfully transmitted within the allowable delay range and the timestamp alignment is valid. On the contrary, if the difference between the timestamps at both ends of each sensor data is greater than the alignment verification tolerance coefficient, it means that the data transmission time exceeds the expected delay range, and there may be network congestion, transmission errors or data loss. In this case, the sensor data will be marked as abnormal data and retransmission or alarm will be triggered to ensure the accuracy and reliability of the sensor data.

[0036] In a possible implementation, the cloud server performs trend prediction based on the positioning data set and the environmental parameter data set, and performs risk assessment according to the trend prediction result, including:

[0037] The positioning data set and the environmental parameter data set are arranged in time series according to the acquisition timestamp, and a multi-monitoring parameter time series chain is constructed, which includes at least positioning monitoring data, temperature monitoring data, humidity monitoring data, and light monitoring data; according to each monitoring parameter time series chain, a cumulative impact calculation is performed to determine the cumulative effect of each monitoring parameter, wherein the cumulative effect calculation expression of the positioning monitoring data is: , is the positioning monitoring data at the i-th time point, is the initial position, is the starting time point of accumulation, t is the current time point, and the cumulative effect calculation expression of environmental parameter monitoring data is: , is the cumulative effect of environmental parameters, is the environmental parameter value at the i-th time point, is the safety threshold of environmental parameters, is the indicator function, when 1 if yes, otherwise 0.

[0038] Specifically, in order to ensure that all monitoring data can be analyzed in chronological order, the positioning data set and the environmental parameter data set are first arranged in time sequence according to the acquisition timestamps in the positioning data set and the environmental parameter data set, that is, the sensor data at different time points are organized in chronological order to ensure the time correlation between the data, thereby forming a multi-monitoring parameter time series chain. This multi-monitoring parameter time series chain includes at least the positioning monitoring data time series chain, the temperature monitoring data time series chain, the humidity monitoring data time series chain, and the light monitoring data time series chain. The positioning monitoring data time series chain records the location information of the collection at different time points. The temperature monitoring data time series chain records the changes in the ambient temperature around the collection. The humidity monitoring data time series chain records the changes in ambient humidity. The light monitoring data records the changes in light intensity. Subsequently, based on these time series chains, the cumulative impact calculation of each monitoring parameter is performed according to the cumulative effect calculation expression of the positioning monitoring data and the cumulative effect calculation expression of the environmental parameter monitoring data to determine the cumulative effect of each parameter over time. The purpose of the cumulative effect calculation is to capture the trend changes of environmental parameters or positioning data over a period of time. Although the threshold is not exceeded at a certain moment, the accumulation process has begun to have a potential impact on the collection. Among them, the cumulative effect calculation expression of positioning monitoring data is: ;in, is the cumulative effect of the positioning data, i.e., from the start time The cumulative change of the positioning data up to the current time point t. It is the positioning monitoring data at the i-th time point, that is, the actual position of the collection measured by the sensor at the time point i. is the initial position, that is, the position of the collection at the beginning of the cumulative effect calculation. is the starting time point of accumulation, that is, the starting time point for calculating the cumulative effect. t is the current time point, that is, the current moment when the cumulative effect is calculated. This expression compares the positioning data at each moment with the initial position and accumulates the cumulative change in the positioning data. The size of the cumulative effect reflects the total displacement of the collection in the time dimension, which can help monitor the displacement trend of the collection. Even if the displacement at a certain moment is not significant, the cumulative effect can reflect whether the long-term displacement will pose a threat to the safety of the collection. The calculation expression for the cumulative effect of environmental parameter monitoring data is: ;in, is the cumulative effect of environmental parameters, i.e., from the start time The part of an environmental parameter (such as temperature, humidity, light, etc.) that exceeds the safety threshold from the current time point t. is the environmental parameter value at the i-th time point, including temperature, humidity or light and other parameter values. It is the safety threshold of environmental parameters, that is, the upper or lower limit of the environmental conditions that the collection can withstand. is the indicator function, when The cumulative effect of environmental parameters reflects the long-term impact of environmental factors such as temperature, humidity, and light on the collection by accumulating the changes that exceed the safety threshold. If a parameter exceeds the safety range for a long time, even if the single excess value is not large, if the cumulative effect is too large, it may cause irreversible damage to the collection.

[0039] The risk trend of museum collections is predicted based on the cumulative effect to obtain a time series risk prediction result, which is the risk probability of each predicted time series corresponding to a risk threshold. When the risk probability reaches the threshold, risk warning information is generated.

[0040] Specifically, after obtaining the cumulative effect, the cumulative effect is input into the pre-built risk trend prediction model, which is built based on the long short-term memory network (LSTM). The risk trend prediction model analyzes the cumulative effect through the learned mapping relationship to generate a time series risk prediction result. This time series risk prediction result includes the risk probability of each predicted time series corresponding to the risk threshold. This risk threshold is set according to the sensitivity, material and preservation conditions of each collection, and is used to determine when potential risks that may endanger the collection may occur in the trend prediction, including environmental parameter risk thresholds, positioning data risk thresholds, etc. The environmental parameter risk thresholds include temperature thresholds, humidity thresholds, light thresholds, etc. Taking the temperature threshold as an example, the temperature threshold can be set to a certain period of time in the future. When the predicted temperature rise exceeds 2°C, it may have a negative impact on the preservation environment of the collection, which is considered a potential risk. When the risk probability of the risk trend prediction reaches or exceeds the set threshold, a risk warning information will be automatically generated. For example, if the probability of a temperature rise of 2°C in the future is predicted to reach 90%, a temperature warning will be sent to notify managers to take timely actions, such as adjusting the air conditioning system or protecting the collection. If the probability of a displacement of more than 3 mm in the future is predicted to reach 90%, a positioning warning will be triggered to remind managers to check whether there are unstable factors in the support or environment to avoid damage to the collection.

[0041] For the risk trend prediction model, first obtain the historical positioning data and historical environmental parameter data, as well as the historical risk probability of each set of historical positioning data and historical environmental parameter data, and then use the same method as above to calculate the cumulative impact of each set of historical positioning data and historical environmental parameter data to obtain the historical cumulative effect of each set of historical positioning data and historical environmental parameter data. By combining the historical cumulative effect with the corresponding historical risk probability, a training data set is constructed. Subsequently, the risk trend prediction model structure is constructed based on the LSTM model, including the input layer, output layer, LSTM layer, etc., and then the risk trend prediction model is initialized based on random numbers, and the weight and bias of each layer are set. After that, the training data set is input into the risk trend prediction model, and the time series features are extracted through the LSTM layer to generate the risk probability of each future time point, and then the Sigmoid activation function is used to output the probability value between 0 and 1. Then, the binary cross entropy loss function is used to calculate the difference between the risk probability output by the risk trend prediction model and the historical risk probability. When the risk probability output by the risk trend prediction model is close to the historical risk probability, the loss is small; when the difference is large, the loss is large. Then use the back propagation algorithm to calculate the gradient of the loss function for each parameter in the risk trend prediction model, and use the optimizer Adam to update the weights and biases of the risk trend prediction model, gradually reducing the loss value. Repeat this process until the training reaches the maximum number of iterations. At this time, use the data not used for training to evaluate the prediction performance of the risk trend prediction model, and calculate the accuracy and recall of the risk trend prediction model. If the calculated accuracy and recall meet the expected expectations, the current risk trend prediction model is output. When using the risk trend prediction model, by inputting the cumulative effect into the risk trend prediction model, the risk probability of each prediction time series at each time point can be calculated, and then the calculated risk probability of each prediction time series is compared with the risk threshold to determine whether to generate risk warning information. If the risk probability of each prediction time series is less than or equal to the corresponding risk threshold, no risk warning will be generated. On the contrary, if there is a risk probability greater than the corresponding risk threshold, risk warning information will be generated.

[0042] In a possible implementation, the risk threshold includes:

[0043] Construct a multi-factor interaction model, take multiple environmental parameters and positioning data as input variables, capture the relationship between variables through the interaction model, and fit linear or nonlinear relationships; determine the change in the size of the collection based on the positioning data, take the change in the size of the collection as the target, and fit the interactive influence relationship of the multiple environmental parameters according to the timestamp alignment relationship between the environmental parameters and the positioning data; according to the risk threshold of the collection, analyze the environmental parameters according to the interactive influence relationship, and determine the risk threshold of each environmental parameter.

[0044] Specifically, we first build a multi-factor interaction model, taking multiple environmental parameters (such as temperature, humidity, light, etc.) and positioning data (such as displacement or position change) as input variables to capture the relationship between the variables. The core purpose of the interaction model is to fit linear or nonlinear relationships to reflect how these variables affect each other. Fitting linear relationships is achieved through the linear fitting formula in the multi-factor interaction model. The linear fitting formula is as follows: ;in, It is the interactive influence relationship result obtained through linear fitting. is the linear influence coefficient of the environmental parameter. is the linear influence coefficient of the positioning data. is the interaction term between two environmental parameters. is the temperature data, It is humidity data. is the nth environmental data. For fitting nonlinear relationships, an LSTM model trained based on the corresponding sample data is used for fitting. The training method of this LSTM model is the same as the training method of the aforementioned risk trend prediction model. Subsequently, in order to determine the size change of the collection, the size change of the collection is obtained through multiple positioning sensors arranged on the contour or edge of the collection. Each positioning sensor is arranged at a key position of the collection contour, and the relative distance between adjacent sensors can be collected in real time. Based on the change in these distances, it is determined whether the size of the collection has changed. The size change is then calculated by the change in distance between the positioning sensors as a goal of the interaction model. For example, the initial positions of the positioning sensors are and , the position at time t is and , then the distance change at time point t is .

[0045] Afterwards, in order to ensure that the model can accurately capture the relationship between environmental parameters and changes in the size of the collection, the environmental parameter data and the positioning data are time-stamp aligned to ensure synchronization between the two. With the data after time-stamp alignment, the interactive model can further fit the interaction between multiple environmental parameters and positioning data. For example, the combined change of temperature and humidity may cause a slight change in the size of the collection. Then, in the interactive model, the interactive impact of multiple environmental parameters is analyzed in combination with the risk thresholds of different collections to determine the risk thresholds of these environmental parameters under specific circumstances. During the analysis process, the corresponding safety thresholds can be set according to the materials, preservation requirements and thresholds of environmental parameters of different collections. For example, the temperature threshold of a certain collection is 2°C and the humidity threshold is 60%. Based on the relationship fitted by the interactive model, it is determined when these parameters will approach or exceed their risk thresholds. Then, by analyzing the interactive relationship between different environmental parameters and positioning data, the risk thresholds of certain environmental parameters are determined. For example, if the cumulative effect of the simultaneous increase in temperature and humidity reaches a certain critical point, a risk warning may be triggered.

[0046] Embodiment 2 is based on the same inventive concept as the IoT management method for museum collections in the above embodiment. Figure 2 As shown, the present application provides an Internet of Things management platform for museum collections, and the platform and method embodiments in the present application embodiments are based on the same inventive concept. The platform includes:

[0047] Mapping association establishment module 1: The mapping association establishment module 1 is used to establish a mapping association between museum collections and Internet of Things devices, and is used to represent the monitoring correspondence between museum collections and Internet of Things devices. The Internet of Things devices include positioning sensor devices and environmental sensor devices; sensor data acquisition module 2: The sensor data acquisition module 2 is used to connect to the Internet of Things device communication network and obtain a sensor data set. The sensor data set includes a positioning data set and an environmental parameter data set. The sensor data set has a collection timestamp, and each collection sensor data block is constructed according to the mapping relationship; early warning feedback module 3: The early warning feedback module 3 is used to import each collection sensor data block into the edge node and the cloud server respectively. The edge node performs real-time tracking and early warning according to the positioning data set and the environmental parameter data set according to the preset threshold value. The cloud server performs trend prediction based on the positioning data set and the environmental parameter data set, performs risk assessment according to the trend prediction results, and performs early warning feedback according to the risk assessment results.

[0048] Furthermore, the sensor data acquisition module 2 also includes:

[0049] Obtain an installation relationship between the museum collection and the Internet of Things device, the installation relationship including direct installation and indirect installation, the direct installation means that the sensor device is directly installed on the collection, and the indirect installation means that the sensor device is not directly installed on the collection but installed on an auxiliary structure; if the installation relationship is the indirect installation, locate the relative installation distance and relative installation angle between the museum collection and the auxiliary structure; determine a correction coefficient based on the relative installation distance and relative installation angle combined with the acquisition accuracy of the Internet of Things device; if the installation relationship is the direct installation, determine a correction coefficient based on the acquisition accuracy of the Internet of Things device; use the correction coefficient to correct the sensor data collected by the Internet of Things device and add it to the sensor data set.

[0050] Furthermore, the sensor data acquisition module 2 also includes:

[0051] For the direct installation, the correction factor calculation formula is expressed as: ; For the indirect installation, the correction coefficient calculation formula is expressed as: ,in, To install the correction factor directly, is the indirect installation correction factor, is the acquisition accuracy of the sensor equipment, is the distance influence coefficient, , d is the relative installation distance, is the relative installation angle, k is the proportional constant used to adjust the proportions of different devices, and k is determined by experimental measurement or device specifications.

[0052] Furthermore, the early warning feedback module 3 also includes:

[0053] Obtain the collection timestamp of the collection sensor data block, and perform timestamp alignment verification on the sensor data received by the edge node according to the collection frequency of the Internet of Things device. The timestamp alignment verification includes horizontal alignment verification and vertical alignment verification. The horizontal alignment verification performs timeline and time interval alignment for the unified sensor device to ensure the collection timing stability of each sensor device; the vertical alignment verification performs synchronous collection timestamp alignment for all sensor data in the data block to ensure the synchronization of the collection of positioning data and environmental parameter data; when the horizontal alignment verification and vertical alignment verification meet the alignment verification conditions, the positioning data and environmental parameter data are imported into the edge node; the positioning data and environmental parameter data obtained by the edge node are transmitted to the cloud server at a regular time, and the timestamp alignment verification is performed on the received data of the cloud server and the transmitted data of the edge node, and the positioning data and environmental parameter data that meet the alignment verification are stored in the cloud server.

[0054] Furthermore, the early warning feedback module 3 also includes:

[0055] According to the network transmission delay, an alignment verification tolerance coefficient is configured; and the alignment verification result is determined based on the alignment verification tolerance coefficient.

[0056] Furthermore, the early warning feedback module 3 also includes:

[0057] The positioning data set and the environmental parameter data set are arranged in time series according to the acquisition timestamp, and a multi-monitoring parameter time series chain is constructed, which includes at least positioning monitoring data, temperature monitoring data, humidity monitoring data, and light monitoring data; according to each monitoring parameter time series chain, a cumulative impact calculation is performed to determine the cumulative effect of each monitoring parameter, wherein the cumulative effect calculation expression of the positioning monitoring data is: , is the positioning monitoring data at the i-th time point, is the initial position, is the starting time point of accumulation, t is the current time point, and the cumulative effect calculation expression of environmental parameter monitoring data is: , is the cumulative effect of environmental parameters, is the environmental parameter value at the i-th time point, is the safety threshold of environmental parameters, is the indicator function, when The cumulative effect is 1 when the risk trend of the museum collection is predicted, and a time series risk prediction result is obtained. The time series risk prediction result is the risk probability of each predicted time series corresponding to the risk threshold. When the risk probability reaches the threshold, risk warning information is generated.

[0058] Furthermore, the early warning feedback module 3 also includes:

[0059] Construct a multi-factor interaction model, take multiple environmental parameters and positioning data as input variables, capture the relationship between variables through the interaction model, and fit linear or nonlinear relationships; determine the change in the size of the collection based on the positioning data, take the change in the size of the collection as the target, and fit the interactive influence relationship of the multiple environmental parameters according to the timestamp alignment relationship between the environmental parameters and the positioning data; according to the risk threshold of the collection, analyze the environmental parameters according to the interactive influence relationship, and determine the risk threshold of each environmental parameter.

[0060] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0062] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An Internet of Things management method for museum collections, characterized in that: The method comprises: Establishing a mapping association between museum collections and IoT devices to indicate a monitoring correspondence between museum collections and IoT devices, wherein the IoT devices include positioning sensor devices and environmental sensor devices; Connecting to the IoT device communication network, acquiring a sensor data set, wherein the sensor data set includes a positioning data set and an environmental parameter data set, wherein the sensor data set has a collection timestamp, and constructing sensor data blocks for each collection according to a mapping relationship; The sensor data blocks of each collection are respectively imported into the edge node and the cloud server. The edge node performs real-time tracking and early warning according to the positioning data set and the environmental parameter data set according to the preset threshold value. The cloud server performs trend prediction according to the positioning data set and the environmental parameter data set, performs risk assessment according to the trend prediction results, and provides early warning feedback according to the risk assessment results. The acquiring of the sensor data set comprises: Acquire the installation relationship between the museum collection and the IoT device, where the installation relationship includes direct installation and indirect installation. The direct installation is that the sensor device is directly installed on the collection, and the indirect installation is that the sensor device is not directly installed on the collection but installed on an auxiliary structure. If the installation relationship is the indirect installation, locate the relative installation distance and relative installation angle between the museum collection and the auxiliary structure; Determine a correction coefficient according to the relative installation distance, the relative installation angle and the collection accuracy of the Internet of Things device; If the installation relationship is the direct installation, a correction coefficient is determined according to the acquisition accuracy of the IoT device; The sensor data collected by the IoT device is corrected using the correction coefficient and then added to the sensor data set.

2. The Internet of Things management method for museum collections according to claim 1, characterized in that: The determining of the correction coefficient comprises: For the direct installation, the correction factor calculation formula is expressed as: ; For the indirect installation, the correction factor calculation formula is expressed as: ,in, To install the correction factor directly, is the indirect installation correction factor, is the acquisition accuracy of the sensor equipment, is the distance influence coefficient, , d is the relative installation distance, is the relative installation angle, k is the proportional constant used to adjust the proportions of different devices, and k is determined by experimental measurement or device specifications.

3. The Internet of Things management method for museum collections according to claim 1, characterized in that: Importing the sensor data blocks of each collection into the edge node and the cloud server respectively includes: Obtain the collection timestamp of the collection sensor data block, and perform timestamp alignment verification on the sensor data received by the edge node according to the collection frequency of the IoT device. The timestamp alignment verification includes horizontal alignment verification and vertical alignment verification. The horizontal alignment verification performs timeline and time interval alignment for the unified sensor device to ensure the collection timing stability of each sensor device; the vertical alignment verification performs synchronous collection timestamp alignment for all sensor data in the data block to ensure the synchronization of the collection of positioning data and environmental parameter data; When the horizontal alignment verification and the vertical alignment verification meet the alignment verification conditions, the positioning data and the environmental parameter data are imported into the edge node; The positioning data and environmental parameter data acquired by the edge node are transmitted to the cloud server at regular intervals, and timestamp alignment verification is performed on the received data of the cloud server and the transmitted data of the edge node. The positioning data and environmental parameter data that meet the alignment verification are stored in the cloud server.

4. The Internet of Things management method for museum collections according to claim 3, characterized in that: Performing timestamp alignment verification on the received data of the cloud server and the transmitted data of the edge node, including: Configure the alignment verification tolerance coefficient according to the network transmission delay; An alignment verification result is determined based on the alignment verification tolerance coefficient.

5. The Internet of Things management method for museum collections according to claim 1, characterized in that: The cloud server performs trend prediction based on the positioning data set and the environmental parameter data set, and performs risk assessment according to the trend prediction results, including: Arrange the positioning data set and the environmental parameter data set in time sequence according to the acquisition timestamp, and construct a multi-monitoring parameter time sequence chain, which includes at least positioning monitoring data, temperature monitoring data, humidity monitoring data, and light monitoring data; According to the time series chain of each monitoring parameter, the cumulative impact calculation is performed respectively to determine the cumulative effect of each monitoring parameter. Among them, the cumulative effect calculation expression of the positioning monitoring data is: , is the positioning monitoring data at the i-th time point, is the initial position, is the starting time point of accumulation, t is the current time point, and the cumulative effect calculation expression of environmental parameter monitoring data is: , is the cumulative effect of environmental parameters, is the environmental parameter value at the i-th time point, is the safety threshold of environmental parameters, is the indicator function, when 1 when , otherwise 0; The risk trend of museum collections is predicted based on the cumulative effect to obtain a time series risk prediction result, which is the risk probability of each predicted time series corresponding to a risk threshold. When the risk probability reaches the threshold, risk warning information is generated.

6. The Internet of Things management method for museum collections according to claim 5, characterized in that: The method further comprises: Construct a multi-factor interaction model, taking multiple environmental parameters and location data as input variables, and use the interaction model to capture the relationship between variables and fit linear or nonlinear relationships; Determine the change in the size of the collection according to the positioning data, and fit the interactive influence relationship of the multiple environmental parameters according to the alignment relationship between the environmental parameters and the timestamps of the positioning data, taking the change in the size of the collection as the target; According to the risk threshold of the collection, the environmental parameters are analyzed according to the interactive influence relationship to determine the risk threshold of each environmental parameter.

7. An Internet of Things management platform for museum collections, characterized in that: The platform is used to execute the Internet of Things management method for museum collections according to any one of claims 1 to 6, and the platform includes: Mapping association establishment module: establishes a mapping association between museum collections and IoT devices, which is used to indicate the monitoring correspondence between museum collections and IoT devices, wherein the IoT devices include positioning sensor devices and environmental sensor devices; Sensor data acquisition module: connects to the IoT device communication network to acquire sensor data sets, including positioning data sets and environmental parameter data sets. The sensor data sets have acquisition timestamps, and construct sensor data blocks for each collection according to the mapping relationship. Early warning feedback module: The sensor data blocks of each collection are imported into the edge node and the cloud server respectively. The edge node performs real-time tracking and early warning according to the preset thresholds based on the positioning data set and the environmental parameter data set. The cloud server performs trend prediction based on the positioning data set and the environmental parameter data set, performs risk assessment according to the trend prediction results, and provides early warning feedback according to the risk assessment results.

Images