Preventive cultural relic protection system based on storage environment monitoring and regulation

Abstract

The invention relates to the field of cultural relic protection, and discloses a cultural relic preventive protection system based on storage environment monitoring and regulation, which comprises a data acquisition module, a data fusion module, a trend prediction module, an optimal regulation module and an alarm notification module, the data acquisition module acquires multi-modal data in a cultural relic storage environment; the data fusion module performs weighted fusion on the multi-modal data to generate fusion data representing an environment state; the trend prediction module predicts the fusion data and generates a trend prediction value of a future environment state; the optimal regulation and control module generates a multi-step optimal regulation and control path and outputs a regulation and control instruction to regulate the equipment; and the alarm notification module generates graded alarm information and notifies a manager through a user interface or a mobile terminal. The system effectively improves the stability and protection efficiency of a cultural relic storage environment through precise monitoring, prospective prediction and intelligent regulation and control, and realizes intelligent and automatic preventive protection of cultural relics.

Description

Technical Field

[0001] The present invention relates to the technical field of cultural relics protection, and in particular to a cultural relics preventive protection system based on storage environment monitoring and regulation. Background Art

[0002] In the field of cultural relics protection, the state of the storage environment of cultural relics has a crucial impact on their preservation quality and lifespan. Changes in environmental factors such as temperature, humidity, gas, and light may cause physical, chemical, or biological damage to cultural relics, leading to irreversible damage. At present, many cultural relics storage systems rely on a single or a few environmental parameters for monitoring, which cannot fully and accurately reflect the state changes in complex storage environments. At the same time, existing monitoring systems usually lack data fusion and intelligent prediction technology, and the data collected by sensors are easily affected by noise and accidental factors. Direct use of unprocessed raw data may lead to misjudgment.

[0003] In addition, most existing cultural relics storage systems adopt simple environmental control methods, lacking fine control over the start and stop or adjustment strategies of the equipment, resulting in unsatisfactory protection effects and even unnecessary energy waste. Intelligent prediction and control methods are rarely used in existing technologies, making it difficult to make proactive adjustments in a timely manner according to environmental change trends. Often, the environment is not reacted until it fluctuates significantly or deviates from the safe range of cultural relics. This delayed control method increases the risk of damage to cultural relics. In addition, the alarm mechanism for environmental anomalies is often relatively simple, making it difficult to accurately distinguish the degree of deviation. There is a lack of graded alarms and real-time notifications based on prediction results, which cannot effectively help managers judge and respond to different levels of environmental risks. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a cultural relics preventive protection system based on storage environment monitoring and regulation, which realizes accurate monitoring and preventive protection of the cultural relics storage environment through multimodal environment monitoring, trend prediction and intelligent regulation.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a cultural relics preventive protection system based on storage environment monitoring and regulation, comprising:

[0006] Data acquisition module, used to collect multimodal data in the cultural relics storage environment, including temperature and humidity parameters, gas parameters, and light intensity parameters;

[0007] The data fusion module performs weighted fusion of multimodal data based on the Bayesian estimation method to generate fused data that represents the environmental state;

[0008] A trend prediction module, based on Markov chain Monte Carlo sampling and long short-term memory network model, performs trend prediction on the fused data and generates a prediction result of the future environmental state;

[0009] The optimal control module generates a multi-step optimal control path based on a dynamic programming algorithm and a reinforcement learning strategy, and outputs control instructions for adjusting storage environment equipment;

[0010] The alarm notification module is used to generate alarm information according to the prediction result and send the alarm notification to the management personnel through the user interface or the mobile application.

[0011] Preferably, the data acquisition module includes a temperature and humidity sensor, a gas sensor and a light sensor, wherein:

[0012] The temperature and humidity sensor is used to collect temperature and humidity data in the storage environment;

[0013] Gas sensors are used to collect CO2, SO2, formaldehyde, ozone, PM2.5, PM10, TVOC and atmospheric pressure data in the environment;

[0014] The light sensor is used to collect light intensity data of the environment.

[0015] Preferably, the data fusion module includes:

[0016] Weight calculation unit, used to calculate the dynamic weight w of various sensor data according to the Bayesian formula i , where the weight of each sensor is w i Calculated by the following formula:

[0017]

[0018] Among them, P(E i |D) is sensor E i The credibility under data D, P(D|E i ) is assumed that sensor E i The likelihood value of data D when the data is reliable, P(E i ) is sensor E i The prior confidence of , N represents the total number of sensors;

[0019] Fusion calculation unit, used to calculate the weight w obtained by the weight calculation unit i Perform weighted summation on various sensor data to generate a multimodal data fusion value S t , and its calculation formula is as follows:

[0020]

[0021] Among them, St is the fused data representing the environment state at time t.

[0022] Preferably, the trend prediction module includes:

[0023] The state transfer matrix generation unit is used to sample the fusion data through the Markov chain Monte Carlo sampling method to construct the state transfer matrix P of the environment state, where the matrix element P ij Indicates that from state S i Transfer to state S j The transition probability is calculated as:

[0024] P ij =P(S t+1 =S j |S t =S i )

[0025] Among them, S i and S j They represent different states of the environment respectively, and the state transfer matrix P is used to describe the dynamic transfer process of the environment state;

[0026] The time series prediction unit is used to input the state sequence into the long short-term memory network model to generate the predicted value of the environmental state in the next k steps.

[0027] Preferably, the optimal control module includes:

[0028] The dynamic programming unit is used to calculate the optimal control option for the current state based on the Bellman equation according to the prediction results of the future environmental state. The formula is:

[0029]

[0030] Among them, V(s t ) is the current state s t The optimal value function of is used to generate the optimal control strategy for multiple steps in the future by gradually optimizing the control path;

[0031] A reinforcement learning unit is used to dynamically optimize the control strategy based on the Q-learning algorithm, which updates the Q value of each state-action pair by the following formula:

[0032] Q(s t ,a t )=Q(s t ,a t )+α[R t +γmaxa Q (s t+1 ,a)-Q(s t ,at )]

[0033] Among them, s t Indicates the current environment status, a t Indicates the device control action, R t is the reward value, α is the learning rate, γ is the discount factor, and the reinforcement learning unit optimizes the control strategy of the device through multiple iterations.

[0034] Preferably, the optimal control module further includes:

[0035] The objective function setting unit is used to set the minimization target of the energy consumption and environmental deviation of the equipment. The objective function is defined as follows:

[0036]

[0037] Among them, C t represents the energy consumption of the device at time t, u t Indicates the switch status of the device, L t It represents the loss function when the environmental state deviates from the cultural relics protection area, and λ is the trade-off coefficient between energy consumption and protection effect.

[0038] Preferably, the alarm notification module includes:

[0039] The deviation detection unit is used to receive the future environmental state prediction value generated by the trend prediction module, compare it with the preset cultural relics safety protection interval, and calculate the prediction value deviation ΔS t , and its calculation formula is as follows:

[0040]

[0041] in, is the predicted environmental state at future time t, S safe is the center value of the safety interval for cultural relics protection;

[0042] The alarm level determination unit is used to determine the deviation ΔS calculated by the deviation detection unit. t Determine the alarm level, which is divided into minor alarm, significant alarm and severe alarm, where:

[0043] If ΔS t If it is less than the first threshold ∈1, a slight alarm signal is generated;

[0044] If ΔS t If it is greater than or equal to the first threshold ∈1 and less than the second threshold ∈2, a significant alarm signal is generated;

[0045] If ΔS t If it is greater than or equal to the second threshold ∈2, a serious alarm signal is generated;

[0046] The notification generating unit is used to send an alarm message to the user interface or mobile application according to the alarm level generated by the alarm level determining unit, wherein the alarm message includes an environmental deviation value and an alarm level to remind the management personnel to take appropriate intervention measures.

[0047] Preferably, the alarm notification module sends the alarm information to the management personnel via a Web interface, a mobile application or a text message push.

[0048] The present invention also provides a preventive protection method for cultural relics based on storage environment monitoring and regulation, comprising the following steps:

[0049] Acquire multimodal data and collect temperature, humidity, gas concentration and light intensity data of the storage environment through the data acquisition module;

[0050] Use Bayesian estimation to perform weighted fusion on the collected data to generate fused data of the environmental state;

[0051] The state transfer matrix of the fused data is generated based on Markov chain Monte Carlo sampling, and the state sequence is input into the long short-term memory network model to predict the future environment state;

[0052] Based on the prediction results of future environmental states, a multi-step optimal control path is generated through dynamic programming algorithms and reinforcement learning strategies, and control instructions are output to the equipment;

[0053] Generate alarm information based on the deviation of the prediction results, and send alarm notifications to managers through the user interface or mobile application.

[0054] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the above method is implemented.

[0055] The present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0056] The present invention provides a preventive protection system for cultural relics based on storage environment monitoring and regulation. It has the following beneficial effects:

[0057] 1. The present invention uses multimodal data collection and Bayesian estimation data fusion method to weight multiple sensor data such as temperature and humidity, gas concentration and light intensity, which can reduce the measurement error of a single data source, generate more stable and accurate environmental status information, and provide reliable environmental data support for cultural relics protection.

[0058] 2. The present invention adopts a combination of Markov chain Monte Carlo sampling and long short-term memory network (LSTM) model to dynamically predict the environmental status, so that the system can predict the changing trend of the environment in advance, provide timely adjustment basis for the protection of cultural relics, and prevent environmental changes from having potential impacts on cultural relics.

[0059] 3. The system of the present invention uses a control algorithm that combines dynamic programming and reinforcement learning to automatically optimize the control path of the equipment while meeting the needs of cultural relics protection. This not only ensures that the environment operates smoothly within the protection area, but also effectively reduces the energy consumption of the equipment, achieving a balance between protection effect and energy cost.

[0060] 4. The alarm notification module of the present invention can generate alarm information of different levels, such as minor, significant, and severe, according to the degree of deviation of the environmental status, and notify the management personnel in real time through the user interface or mobile terminal to help them take countermeasures in time, further improving the response speed and management efficiency of the system.

[0061] 5. The present invention makes the cultural relics protection system more efficient and accurate through intelligent management of the entire process of monitoring, prediction, regulation and alarm, reduces the frequency of manual intervention, provides intelligent and automated protection means for the long-term safe preservation of cultural relics, and improves the overall quality and efficiency of cultural relics protection work. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the system architecture of the present invention;

[0063] Figure 2 It is a schematic diagram of the method flow of the present invention;

[0064] Figure 3 It is a schematic diagram of the computer device structure of the present invention.

[0065] Among them, 40, computer equipment; 41, processor; 42, memory; 43, storage medium. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Please refer to the attached Figure 1The embodiment of the present invention provides a cultural relic preventive protection system based on storage environment monitoring and regulation. Through multimodal data fusion and intelligent prediction and regulation, accurate monitoring, dynamic prediction and adaptive regulation of the cultural relic storage environment are achieved, thereby extending the preservation life of the cultural relics and improving the cultural relic protection effect. The system includes a data acquisition module, a data fusion module, a trend prediction module, an optimal regulation module and an alarm notification module. The following describes in detail the various modules of the cultural relic preventive protection system based on storage environment monitoring and regulation of the present invention.

[0068] Data acquisition module

[0069] As the basic module of the system, the data acquisition module is responsible for obtaining key parameter information in the environment and providing accurate input data for the data fusion module and subsequent trend prediction and regulation. This module is deployed in the cultural relics storage environment, and collects parameters such as temperature and humidity, light intensity, and gas concentration in real time through sensors to establish the environmental status data foundation and support the subsequent modules to analyze and make decisions on environmental changes.

[0070] In this embodiment, the data acquisition module includes a temperature and humidity sensor, a gas concentration sensor, and a light intensity sensor, which are respectively used to collect the temperature and humidity, specific gas concentration, and light intensity in the cultural relics storage environment.

[0071] Temperature and humidity sensor:

[0072] The temperature and humidity sensor is used to collect the temperature T and humidity H in the storage environment and transmit these data to the data fusion module in real time. Temperature and humidity have a significant impact on the preservation of cultural relics. Higher humidity may cause mold growth or metal oxidation, while lower humidity may cause paper, wood and other materials to dry out and crack. The collection frequency of the temperature and humidity sensor can be set according to actual needs to ensure the real-time and monitoring accuracy of the environmental status.

[0073] Gas concentration sensor:

[0074] Gas concentration sensors are used to monitor specific gas concentrations in the storage environment, including carbon dioxide (CO2), sulfur dioxide (SO2), formaldehyde (HCHO), ozone (O3), PM2.5, PM10, and total volatile organic compounds (TVOC). These gases have potential chemical damage to cultural relics. For example, sulfur dioxide (which can be reflected by the SO2 index in the environment) can cause corrosion of metal materials and acidification of paper; fluctuations in carbon dioxide concentration reflect changes in air quality; and increases in formaldehyde and TVOC concentrations may accelerate the aging of organic materials. Gas concentration sensors can provide input data reflecting air quality to the data fusion module by quantitatively detecting the concentrations of various harmful gases. Assume that the carbon dioxide concentration detected at time t is The sulfur dioxide concentration is The formaldehyde concentration is cHCHO,t , ozone concentration is PM2.5 concentration is c PM2.5,t 、PM10 concentration is c PM10,t .

[0075] Light intensity sensor:

[0076] The light intensity sensor is used to monitor the light intensity L in the storage environment of cultural relics. Light has a great impact on cultural relics, especially those made of paper, textiles, pigments, etc. Excessive light can cause cultural relics to fade and deteriorate. The light intensity sensor can monitor the changes in ambient light in real time and transmit the light data to the data fusion module. Assume that the light intensity detected at time t is L t .

[0077] In summary, the data acquisition module forms environmental status data at time t through the coordinated work of temperature and humidity sensors, gas concentration sensors and light intensity sensors. Provide basic multimodal data for the data fusion module and subsequent environmental monitoring processes.

[0078] Through the collaborative collection of multiple sensors, the data acquisition module undertakes the basic task of real-time monitoring of environmental status in the entire preventive protection system, ensuring the integrity and timeliness of stored environmental data, and providing accurate environmental input for analysis and regulation of subsequent modules.

[0079] Data fusion module

[0080] The data fusion module receives the multimodal environmental data provided by the data acquisition module, and integrates the data from different sources into a single environmental status data through weighted fusion technology to improve the accuracy and consistency of the monitoring data. Since various sensors have different noise levels and precision, directly using the original data may lead to the accumulation of measurement errors. The data fusion module uses the Bayesian estimation method to dynamically adjust the weights of each data source to generate accurate and stable environmental status information, providing reliable input for subsequent trend prediction and optimal control.

[0081] In this embodiment, the data fusion module uses the Bayesian estimation method to perform weighted fusion on the multimodal data, which specifically includes the following steps:

[0082] 1. Dynamic weight calculation

[0083] In the data fusion process, the system first performs credibility analysis on the data from different sensors and assigns dynamic weights to each sensor data. Assume that the data acquisition module obtains the environmental status data at time t Where T represents temperature, H represents humidity, cHCHO , c PM2.5 、c PM10 are the concentrations of carbon dioxide, sulfur dioxide, formaldehyde, ozone, PM2.5 and PM10 respectively, and L is the light intensity.

[0084] Weight w i The calculation is based on the Bayesian formula, assuming that the data E of each sensor i Under given measurement data D, there are different credibility P(E i |D), weight w i The calculation formula is as follows:

[0085]

[0086] in:

[0087] P(E i |D) indicates that sensor E is i Credibility;

[0088] P(D|E i ) is assumed that sensor E i The likelihood value of data D when the data is reliable;

[0089] P(E i ) is sensor E i The prior confidence of

[0090] N is the total number of sensors.

[0091] Dynamic weight calculation estimates the credibility of each sensor data through the above Bayesian formula, and assigns corresponding weights to each data based on this.

[0092] 2. Weighted fusion calculation

[0093] The data fusion module calculates the weight of each sensor w according to the weight i Perform weighted summation on multimodal data to generate the final fusion environment state S t , and its calculation formula is as follows:

[0094]

[0095] in:

[0096] S t represents the fusion environment state at time t;

[0097] w i is the weight of the corresponding sensor;

[0098] E i is the data value collected by sensor i.

[0099] S after weighted fusion t It represents the environmental state after Bayesian weighting, integrates the data input from different sensors, reduces the uncertainty of a single data source, and can more accurately reflect the actual state of the cultural relics storage environment.

[0100] Through this data fusion processing, the system can obtain more stable environmental status information, reduce the errors caused by sensor data fluctuations, and provide accurate data input for the trend prediction module and the optimal control module.

[0101] Trend Forecast Module

[0102] The trend prediction module receives the fused data provided by the data fusion module, and dynamically predicts the future environmental status through time series analysis, providing forward-looking data support for the optimal control module. Since the cultural relics preservation environment is time-dependent, the trend prediction module uses Markov chain Monte Carlo (MCMC) sampling combined with the long short-term memory network (LSTM) model to predict future environmental change trends through dynamic analysis of the environmental status, ensuring that the system can make timely adjustments before the environment deviates from the protection zone.

[0103] In this embodiment, the trend prediction module dynamically predicts the fusion data based on Markov chain Monte Carlo sampling and LSTM model, specifically including the following steps:

[0104] 1. State transfer matrix generation

[0105] In order to analyze the time dependence of the state of the cultural relics storage environment, the trend prediction module first constructs the state transition matrix of the environment state through Markov chain Monte Carlo (MCMC) sampling. Assume that the environment state is divided into n discrete states, denoted as S1, S2, …, S n , in each state S i The system generates the transition probability to the next state. The transition probability P between each state is calculated by multiple sampling. ij , forming the state transfer matrix P:

[0106] P ij =P(S t+1 =S j |S t =S i )

[0107] in:

[0108] P ij Indicates that from state S i Transfer to state S j The probability of

[0109] Si and S j for the current and future state of the environment.

[0110] The state transition matrix P describes the changing trend of the environmental state and provides data support for subsequent time series prediction.

[0111] 2. Time Series Forecasting

[0112] After generating the state transfer matrix, the trend prediction module inputs the state sequence into the LSTM model for multi-step prediction. The LSTM model can capture long-term dependencies and model the historical sequence through its hidden state to predict the future environment state.

[0113] In this embodiment, the hidden layer state h of the LSTM model t The calculation formula is as follows:

[0114] h t =σ(W x x t +W h h t-1 +b)

[0115] in:

[0116] h t is the hidden state of the current time step, representing the predicted information of the environment state;

[0117] W x and W h is the input weight matrix and the hidden layer weight matrix;

[0118] x t The state data entered at the current moment;

[0119] b is the bias.

[0120] By modeling the time series of environmental data through LSTM, the system can predict the environmental state sequence of the next k steps.

[0121] 3. Trend forecast output

[0122] After obtaining the environmental state sequence, the LSTM model outputs the predicted multi-step environmental state The trend prediction value of future environmental changes is formed. The trend prediction result provides a forward-looking basis for the optimal control module, enabling the system to respond in advance according to the predicted future environmental state.

[0123] Through the above steps, the trend prediction module realizes the dynamic prediction of future environmental change trends, providing timely and accurate environmental status reference for preventive protection of cultural relics.

[0124] Optimal control module

[0125] The optimal control module receives the predicted value of the future environmental state provided by the trend prediction module, and generates the optimal control strategy for the environmental control equipment according to the objective function set by the system. The optimal control module adopts a combination of dynamic programming (DP) algorithm and reinforcement learning (RL) to minimize the system energy consumption while ensuring the protection needs of cultural relics by optimizing the control path of environmental equipment. This module can dynamically adjust the control strategy to cope with future environmental changes and provide a stable storage environment for cultural relics.

[0126] In this embodiment, the optimal control module generates the optimal control path based on dynamic programming and reinforcement learning strategy, which specifically includes the following steps:

[0127] 1. Objective function setting

[0128] In order to strike a balance between protection effect and energy consumption, the optimal control module sets an objective function to minimize the deviation between equipment energy consumption and environmental state. The objective function is defined as follows:

[0129]

[0130] in:

[0131] C t To regulate the energy consumption of the equipment at time t;

[0132] u t is the operating state of the device at time t (e.g., on or off);

[0133] L t is the loss function of the environmental state deviating from the cultural relics protection area, reflecting the deviation of the environmental state on the cultural relics preservation conditions;

[0134] λ is the balance coefficient, which is used to balance between energy consumption and protection effect.

[0135] By setting this objective function, the optimal control module can dynamically optimize the control strategy according to the energy consumption and cultural relics protection requirements of different equipment.

[0136] 2. Dynamic programming solution

[0137] After the objective function is set, the optimal control module optimizes the current and future environmental conditions through the dynamic programming (DP) algorithm to generate a multi-step optimal control path. The DP algorithm gradually optimizes the control strategy based on the predicted results of the future environmental conditions and the current equipment status to find the control path with the lowest energy consumption and the best protection effect.

[0138] In this embodiment, the DP algorithm calculates the current state s based on the Bellman equationt The optimal value function V(s t ):

[0139]

[0140] in:

[0141] V(s t ) is the current state s t The optimal value function of

[0142] u t Control actions for devices;

[0143] is the expected value of the future state.

[0144] Through the above Bellman equation, the DP algorithm gradually solves the optimal control strategy for each future time step to ensure that the control equipment always runs on the optimal control path.

[0145] 3. Reinforcement Learning Optimization

[0146] Based on DP, the optimal control module introduces the Q-learning algorithm of reinforcement learning (RL) to cope with random changes in the environment by continuously updating and optimizing the control strategy. Q-learning achieves dynamic optimization of the control strategy by evaluating the reward value of each state-action pair.

[0147] In this embodiment, the Q-value update formula of the Q-learning algorithm is as follows:

[0148] Q(s t ,a t )=Q(s t ,a t )+α[R t +γmax a Q(s t+1 ,a)-Q(s t ,a t )]

[0149] in:

[0150] Q(s t ,a t ) is in state s t Next, perform action a t Q value;

[0151] R t is the reward value of the current time step;

[0152] α is the learning rate, which controls the step size of Q value update;

[0153] γ is a discount factor that weighs the impact of current rewards and future rewards.

[0154] Through the dynamic learning of Q-learning, the optimal control module can gradually optimize the control strategy during the actual operation process and adapt to changes in different environments.

[0155] Through the above steps, the optimal control module combines the advantages of dynamic programming and reinforcement learning to achieve precise control of the cultural relics storage environment, ensuring that the system optimizes energy consumption while achieving cultural relics protection effects.

[0156] Alarm notification module

[0157] The alarm notification module receives the future environmental status prediction data provided by the trend prediction module and compares it with the safe range of cultural relics protection in real time, so as to generate an alarm signal and notify the management personnel in time when the environment deviates from the set standard. This module can classify the abnormal deviation of the environmental status into levels, and help the management personnel identify the severity of the risk through multi-level alarm, so as to facilitate timely intervention and ensure that the storage environment of cultural relics is maintained within the appropriate range.

[0158] In this embodiment, the alarm notification module implements the alarm function through the steps of deviation detection, alarm level determination and notification generation:

[0159] 1. Deviation Detection

[0160] The alarm notification module first calculates the deviation between the future environmental state prediction value generated by the trend prediction module and the preset safety interval to determine whether the environmental state deviates from the protection requirements. Assume that the environmental state prediction value at the future time t is The central value of the safety interval for cultural relics protection is S safe , then the deviation value ΔS t The calculation is as follows:

[0161]

[0162] in:

[0163] ΔS t Indicates the deviation between the predicted environmental state and the safe interval;

[0164] is the predicted value of the environmental state at the future time t;

[0165] S safe It is the center value of the safety interval required for cultural relics protection.

[0166] Through the above deviation calculation, the system can quantify the distance between the environmental status and the cultural relics protection requirements, and provide a basis for the judgment of the alarm level.

[0167] 2. Alarm level determination

[0168] The alarm notification module calculates the deviation value ΔS t Determine the alarm level so that managers can take corresponding measures according to different alarm levels. Specifically, the system presets two thresholds ∈1 and ∈2 (∈1<∈2), and divides the alarm level according to the size of the deviation value:

[0169] When ΔS t When <∈1, the environmental deviation is small and the system generates a slight alarm signal;

[0170] When ∈1≤ΔS t <∈2, the environmental deviation reaches a moderate level, and the system generates a significant alarm signal;

[0171] When ΔS t When ≥∈2, the environmental deviation is large and the system generates a serious alarm signal.

[0172] Through this hierarchical judgment mechanism, the alarm notification module can generate alarms of corresponding levels according to the different degrees of deviation, making it easier for managers to quickly identify the severity of environmental risks.

[0173] 3. Notification generation

[0174] According to the alarm level determination result, the alarm notification module generates a notification including deviation information and alarm level, and sends it to the management personnel through the user interface or mobile application. In this embodiment, the notification content of the alarm information includes:

[0175] Deviation value ΔS t , used to indicate the gap between the environmental state and the safety interval;

[0176] Alert level (minor, significant, severe);

[0177] The specific values ​​of relevant environmental parameters can help managers have a clearer understanding of the current or future environmental status.

[0178] After the notification is generated, the system can push alarms through the web interface or mobile applications to ensure that managers can receive alarm information in a timely manner when environmental risks occur.

[0179] Through the above steps, the alarm notification module realizes real-time monitoring and graded alarm of environmental status, which can ensure that alarm reminders are issued in time when abnormal deviations occur in the cultural relics storage environment, and provides effective support for managers' daily monitoring and emergency handling.

[0180] In general, the present invention obtains environmental data through sensors such as temperature and humidity, gas concentration and light, and uses Bayesian estimation for data fusion to generate stable environmental status information; then, Markov chain Monte Carlo sampling and long short-term memory network (LSTM) model are used to predict future environmental change trends; according to the prediction results, the optimal control module generates equipment control strategies through a combination of dynamic programming and reinforcement learning to achieve a balance between energy consumption and protection effects; when the environmental state deviates from the protection interval, the alarm notification module generates a graded alarm to notify the management personnel in time. The system effectively improves the monitoring accuracy and control efficiency of the cultural relics storage environment, and provides an intelligent and automated protection method for the long-term preservation of cultural relics.

[0181] Please refer to the attached Figure 2 The present invention also provides a preventive protection method for cultural relics based on storage environment monitoring and regulation, comprising the following steps:

[0182] S1. Multimodal data acquisition: Through the data acquisition module, the temperature and humidity, gas concentration and light intensity data in the cultural relics storage environment are obtained in real time to comprehensively characterize the various state information of the environment.

[0183] S2. Data fusion processing: The collected multimodal data are input into the data fusion module, and the data are weighted and fused using Bayesian estimation to generate a single environmental status data to eliminate the measurement bias of various data sources and improve the overall accuracy of the environmental status.

[0184] S3. Environmental state prediction: Generate the transfer matrix of the environmental state through Markov chain Monte Carlo sampling, input the state sequence into the long short-term memory network (LSTM) model, predict the future environmental state, and provide data support for subsequent regulation.

[0185] S4. Generation of optimal control strategy: Based on the predicted future environmental state, the optimal environmental control path is generated through a dynamic programming algorithm, and the equipment control scheme is dynamically optimized in combination with a reinforcement learning strategy to determine the equipment's control command output.

[0186] S5. Alarm information generation and notification: Based on the predicted results of the environmental status and the safety requirements for the preservation of cultural relics, the deviation of the predicted value is calculated, and alarm information of different levels is generated and sent to the management personnel through the user interface or mobile application so that protective measures can be taken in time.

[0187] Through multi-step processing, this method realizes intelligent monitoring, dynamic prediction and optimal regulation of the cultural relics storage environment, thereby improving the safety and protection efficiency of cultural relics preservation.

[0188] Please refer to the attached Figure 3The present invention further provides a computer device 40, comprising: a processor 41 and a memory 42, wherein the memory 42 stores a computer program executable by the processor, and when the computer program is executed by the processor, the above method is executed.

[0189] The present invention further provides a storage medium 43 on which a computer program is stored. When the computer program is run by the processor 41, the above method is executed.

[0190] Among them, the storage medium 43 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0191] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cultural relic preventive protection system based on storage environment monitoring and regulation, characterized in that: include: Data acquisition module, used to collect multimodal data in the cultural relics storage environment, including temperature and humidity parameters, gas parameters, and light intensity parameters; The data fusion module performs weighted fusion of multimodal data based on the Bayesian estimation method to generate fused data that represents the environmental state; A trend prediction module, based on Markov chain Monte Carlo sampling and long short-term memory network model, performs trend prediction on the fused data and generates a prediction result of the future environmental state; The optimal control module generates a multi-step optimal control path based on a dynamic programming algorithm and a reinforcement learning strategy, and outputs control instructions for adjusting storage environment equipment; The alarm notification module is used to generate alarm information according to the prediction result and send the alarm notification to the management personnel through the user interface or the mobile application.

2. The cultural relics preventive protection system based on storage environment monitoring and regulation according to claim 1 is characterized in that: The data acquisition module includes a temperature and humidity sensor, a gas sensor and a light sensor, wherein: The temperature and humidity sensor is used to collect temperature and humidity data in the storage environment; Gas sensors are used to collect CO2, SO2, formaldehyde, ozone, PM2.5, PM10, TVOC and atmospheric pressure data in the environment; The light sensor is used to collect light intensity data of the environment.

3. The cultural relics preventive protection system based on storage environment monitoring and regulation according to claim 1 is characterized in that: The data fusion module comprises: Weight calculation unit, used to calculate the dynamic weight w of various sensor data according to the Bayesian formula i , where the weight of each sensor is w i Calculated by the following formula: Among them, P(E i |D) is sensor E i The credibility under data D, P(D|E i ) is assumed that sensor E i The likelihood value of data D when the data is reliable, P(E i ) is sensor E i The prior confidence of , N represents the total number of sensors; Fusion calculation unit, used to calculate the weight w obtained by the weight calculation unit i Perform weighted summation on various sensor data to generate a multimodal data fusion value S t , and its calculation formula is as follows: Among them, S t is the fused data representing the environment state at time t.

4. The cultural relics preventive protection system based on storage environment monitoring and regulation according to claim 1 is characterized in that: The trend prediction module includes: The state transfer matrix generation unit is used to sample the fusion data through the Markov chain Monte Carlo sampling method to construct the state transfer matrix P of the environment state, where the matrix element P ij Indicates that from state S i Transfer to state S j The transition probability is calculated as: P ij =P(S t+1 =S j |S t =S i ) Among them, S i and S j They represent different states of the environment respectively, and the state transfer matrix P is used to describe the dynamic transfer process of the environment state; The time series prediction unit is used to input the state sequence into the long short-term memory network model to generate the predicted value of the environmental state in the next k steps.

5. The cultural relics preventive protection system based on storage environment monitoring and regulation according to claim 1 is characterized in that: The optimal control module includes: The dynamic programming unit is used to calculate the optimal control option for the current state based on the Bellman equation according to the prediction results of the future environmental state. The formula is: Among them, V(s t ) is the current state s t The optimal value function of is used to generate the optimal control strategy for multiple steps in the future by gradually optimizing the control path; A reinforcement learning unit is used to dynamically optimize the control strategy based on the Q-learning algorithm, which updates the Q value of each state-action pair by the following formula: Q(s t ,a t )=Q(s t ,a t )+α[R t +γmax a Q(s t+1 ,a)-Q(s t ,a t )] Among them, s t Indicates the current environment status, a t Indicates the device control action, R t is the reward value, α is the learning rate, γ is the discount factor, and the reinforcement learning unit optimizes the control strategy of the device through multiple iterations.

6. The cultural relics preventive protection system based on storage environment monitoring and regulation according to claim 5 is characterized in that: The optimal control module also includes: The objective function setting unit is used to set the minimization target of the energy consumption and environmental deviation of the equipment. The objective function is defined as follows: Among them, C t represents the energy consumption of the device at time t, u t Indicates the switch status of the device, L t It represents the loss function when the environmental state deviates from the cultural relics protection area, and λ is the trade-off coefficient between energy consumption and protection effect.

7. The cultural relics preventive protection system based on storage environment monitoring and regulation according to claim 1 is characterized in that: The alarm notification module includes: The deviation detection unit is used to receive the future environmental state prediction value generated by the trend prediction module, compare it with the preset cultural relics safety protection interval, and calculate the prediction value deviation ΔS t , and its calculation formula is as follows: in, is the predicted environmental state at future time t, S safe is the center value of the safety interval for cultural relics protection; The alarm level determination unit is used to determine the deviation ΔS calculated by the deviation detection unit. t Determine the alarm level, which is divided into minor alarm, significant alarm and severe alarm, where: If ΔS t If it is less than the first threshold ∈1, a slight alarm signal is generated; If ΔS t If it is greater than or equal to the first threshold ∈1 and less than the second threshold ∈2, a significant alarm signal is generated; If ΔS t If it is greater than or equal to the second threshold ∈2, a serious alarm signal is generated; The notification generating unit is used to send an alarm message to the user interface or mobile application according to the alarm level generated by the alarm level determining unit, wherein the alarm message includes an environmental deviation value and an alarm level to remind the management personnel to take appropriate intervention measures.

8. The cultural relics preventive protection system based on storage environment monitoring and regulation according to claim 1 is characterized in that: The alarm notification module sends the alarm information to the management personnel through the web interface, mobile application or SMS push.

9. A method for preventive protection of cultural relics based on storage environment monitoring and regulation, applied to the preventive protection system for cultural relics based on storage environment monitoring and regulation as claimed in any one of claims 1 to 8, characterized in that: The following steps are involved: Acquire multimodal data and collect temperature, humidity, gas concentration and light intensity data of the storage environment through the data acquisition module; Use Bayesian estimation to perform weighted fusion on the collected data to generate fused data of the environmental state; The state transfer matrix of the fused data is generated based on Markov chain Monte Carlo sampling, and the state sequence is input into the long short-term memory network model to predict the future environment state; Based on the prediction results of future environmental states, a multi-step optimal control path is generated through dynamic programming algorithms and reinforcement learning strategies, and control instructions are output to the equipment; Generate alarm information based on the deviation of the prediction results, and send alarm notifications to managers through the user interface or mobile application.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the preventive protection method of cultural relics based on storage environment monitoring and regulation as described in claim 9.

Images