A data acquisition method based on three-dimensional reconstruction system

Abstract

The present invention relates to the field of data storage technology, specifically disclosing a data acquisition method for a three-dimensional reconstruction system, comprising the following steps: S1. Data acquisition: preprocessing the original dataset #imgabs0#; S2. Data preprocessing: preprocessing the real-time received data #imgabs1# and converting it into a standardized dataset #imgabs2#; S3. Importance score calculation: calculating the importance score #imgabs4# for the preprocessed data #imgabs3#; and S4. implementing a dynamic caching strategy based on the importance score #imgabs5#. Blockchain technology enables data traceability and tamper-proofing. Combined with a spatiotemporal prediction model based on a graph convolutional network, this method enhances the objectivity of data credibility assessment and the accuracy of importance prediction. This provides a more reliable basis for decision-making in the three-dimensional reconstruction system and reduces the risk of industrial accidents caused by data distortion.

Description

Technical Field

[0001] The present invention relates to the field of data storage technology, and in particular to a data acquisition method based on a three-dimensional reconstruction system. Background Art

[0002] With the rapid development of the Industrial Internet of Things (IIoT) and intelligent manufacturing technologies, 3D reconstruction systems are widely used in industrial equipment status monitoring, fault diagnosis, and production process optimization. 3D reconstruction relies on the efficient collection and processing of multi-source heterogeneous data (such as sensor data, image information, and equipment logs). However, existing technologies still have the following shortcomings in data collection and cache management:

[0003] 1. Difficulty integrating multi-source heterogeneous data. The types of sensors and equipment deployed on industrial sites are diverse, and the data formats, protocols, and frequencies vary significantly. Traditional methods are unable to efficiently integrate this data, resulting in low data utilization and high analysis latency.

[0004] 2. Insufficient real-time and reliability requirements Three-dimensional reconstruction requires real-time processing of dynamic data, but the existing system lacks effective mechanisms in data capture rate, time alignment and integrity verification, which can easily lead to data missing or misalignment, affecting reconstruction accuracy and decision reliability.

[0005] 3. Static caching strategies lead to low resource utilization. Traditional caching mechanisms are based on fixed rules (such as first-in, first-out (FIFO)) and fail to fully consider differences in data importance. For data that is frequently accessed, has a high rate of change, or is closely related to key business indicators (KPIs), static strategies cannot optimize cache resource allocation, resulting in low cache hit rates and high system latency.

[0006] 4. Data importance assessment is single and has poor adaptability. Existing methods mostly rely on a single indicator (such as access frequency) to evaluate data value, ignoring multi-dimensional factors such as data urgency, predictive importance and credibility, and are difficult to adapt to the dynamic needs of complex industrial scenarios.

[0007] 5. Lack of multi-level caching and intelligent scheduling capabilities. Industrial data is large in scale and highly timely, but traditional systems lack a multi-level caching architecture and intelligent scheduling strategies based on data characteristics, and are unable to strike a balance between access speed and storage costs.

[0008] To address the above problems, the present invention proposes a three-dimensional reconstruction data acquisition method based on dynamic importance scoring. Through innovative data preprocessing, multi-dimensional scoring model and adaptive caching strategy, it significantly improves the processing efficiency and cache resource utilization of multi-source heterogeneous data, and meets the real-time, reliability and intelligence requirements of industrial scenarios. Summary of the Invention

[0009] The purpose of the present invention is to provide a data acquisition method based on a three-dimensional reconstruction system to solve the problems raised in the above background technology.

[0010] To achieve the above object, the present invention provides the following technical solution: a data acquisition method based on a three-dimensional reconstruction system, comprising the following steps:

[0011] S1. Data acquisition: Configuring multiple data sources for the 3D reconstruction system. These data sources are used to collect raw data related to 3D reconstruction.

[0012] S2. Data preprocessing: real-time received data Preprocess and convert into a standardized data set , this preprocessing process covers:

[0013] Data cleaning: Eliminate abnormal, duplicate and erroneous data;

[0014] Format unification: convert data in different formats into a unified standardized format;

[0015] Feature extraction: Use PCA algorithm to extract key feature vectors ;

[0016] Data compression: Implement lossless or lossy compression techniques to reduce storage and transmission costs;

[0017] S3. Importance score calculation: for preprocessed data , calculate its importance score , the calculation formula is as follows:

[0018] ;

[0019] in:

[0020] The frequency of data access;

[0021] The urgency of the data;

[0022] is the data change rate, and the calculation formula is: ;in, is the time interval;

[0023] For the prediction importance based on deep learning prediction model;

[0024] is the correlation between data and key business indicators, and the calculation formula is: ; represents the correlation coefficient, is the eigenvector, Key performance indicators (KPIs);

[0025] Data credibility, reflecting the reliability of the data source and the integrity of the data itself;

[0026] 、 、 、 、 、 is the weight coefficient;

[0027] S4. Based on importance scoring , implement dynamic caching strategy.

[0028] Preferably, in S1, the specific steps of data acquisition are: S11, configuring multiple data sources, the total number of data sources is ,in Is a positive integer, and each data source corresponds to an industrial device or sensor, numbered ;

[0029] S12. For each data source , at a specific time mark , capturing raw data in real time through the Industrial Internet of Things network ,Should Contains the following information:

[0030] Data source identification code , used to uniquely distinguish data sources ;

[0031] Measurement value or data content , representing the data source In time The original information collected;

[0032] Data format or type description , indicating the data source The structure, type, or format of the data produced, to facilitate the identification and processing of data from multiple sources and formats;

[0033] Additional Information , covers additional information related to the data;

[0034] S13, the data from all data sources in step S12 Collected raw data Aggregate, build in time The original dataset ,Right now: ;

[0035] in: Indicates time data collection activities; Indicates taking all data sources In time The raw data generated Collection of;

[0036] S14, the original data set Perform time stamp alignment and data integrity check;

[0037] S15. Calculate data capture rate , to evaluate the time The effectiveness of data collection, the data capture rate is defined as: ;

[0038] S16: Based on the data capture rate and data integrity check results, determine whether the data collection process meets the real-time and reliability standards. If the conditions are met, the original data set is Perform data preprocessing.

[0039] Preferably, predict importance Use a spatiotemporal prediction model based on graph convolutional networks.

[0040] Preferably, data credibility Evaluation is carried out through blockchain technology, and distributed ledgers are used to record data source information to achieve data traceability and tamper-proofing.

[0041] Preferably, based on importance score ,The specific steps to implement the dynamic caching strategy are:

[0042] S41. Cache storage: scoring importance Exceeding the threshold The data is stored in the cache area;

[0043] S42, cache elimination: when the cache capacity Exceeding the threshold When using the LVD algorithm, press Remove data in order from low to high;

[0044] S43, multi-level cache: according to Due to the difference in cache media, data can be stored in different levels of cache media for multi-level caching;

[0045] S44, Intelligent Data Scheduling: Apply reinforcement learning algorithms to intelligently schedule cached data based on system status and network conditions, optimizing data acquisition and transmission paths;

[0046] S45, Adaptive weight update: Using Bayesian optimization technology, the weight coefficient is adjusted in real time according to the cache hit rate H, system latency L and data loss rate D.

[0047] Preferably, the multi-level cache includes at least three levels of cache media: memory cache, solid state drive cache and disk cache, and is scored according to the importance of the data. Allocate data to the corresponding cache level.

[0048] Preferably, the cache level data allocation rules are as follows:

[0049] when When , the data is stored in the memory cache;

[0050] when When storing data in the SSD cache;

[0051] when When , the data is stored in the disk cache;

[0052] in, and The importance score threshold is preset and meets .

[0053] Preferably, the calculation formula for adjusting the weight coefficient in real time is:

[0054]

[0055] in, 、 、 is the weight parameter, and .

[0056] According to the above-mentioned data acquisition method based on the 3D reconstruction system, a data storage system based on Internet technology is proposed, which includes: data acquisition module, data preprocessing module, scoring calculation module, dynamic cache management module, intelligent scheduling module and adaptive optimization module; the data acquisition module obtains real-time data of industrial equipment through sensor networks and IoT devices. ;The data preprocessing module is used to process real-time data Perform preprocessing to obtain preprocessed data ; The scoring calculation module is used to calculate the importance score of the data The dynamic cache management module is used to execute dynamic cache strategies, including data storage, elimination, and multi-level cache management. The intelligent scheduling module uses reinforcement learning algorithms to achieve intelligent data scheduling and transmission path optimization. The adaptive optimization module uses Bayesian optimization to achieve real-time updates of weight coefficients.

[0057] The present invention provides a data storage method based on Internet technology, which addresses the shortcomings of the existing technology and has the following beneficial effects:

[0058] 1. Efficient integration of multi-source heterogeneous data: Through standardized pre-processing processes, format differences in multi-device and multi-protocol data at industrial sites are effectively eliminated, improving data compatibility and analyzability. By combining data source identification codes and additional information, the system's ability to identify and analyze heterogeneous data is enhanced, significantly improving data utilization.

[0059] 2. Real-time and reliability assurance: Time stamp alignment and integrity verification mechanisms ensure accurate matching of data timestamps and the integrity of key data, reducing 3D reconstruction errors caused by data misalignment or loss. Real-time monitoring of the data capture rate (DAR) provides a quantitative basis for the system to dynamically adjust acquisition strategies, meeting the stringent requirements of industrial scenarios for data timeliness.

[0060] 3. Optimized cache resource allocation: A dynamic caching strategy based on multi-dimensional importance scoring overcomes the limitations of traditional static caching. Through the LVD elimination algorithm and a multi-level cache architecture (memory / SSD / disk), data is stored in layers by value, significantly improving the cache hit rate for high-frequency, high-change rate, and KPI-related data, while reducing system latency.

[0061] 4. Adaptive dynamic optimization capability: Bayesian optimization technology adjusts scoring weight coefficients in real time. Combined with reinforcement learning intelligent scheduling, the system can dynamically optimize data transmission paths and storage strategies based on network status, cache hit rate, and other factors. This mechanism ensures access speed to critical data while balancing storage costs and resource utilization, improving system robustness.

[0062] 5. Improved data quality and credibility: Blockchain technology enables data traceability and tamper-proofing. Combined with a spatiotemporal prediction model based on a graph convolutional network, it enhances the objectivity of data credibility assessment and the accuracy of predicted importance. This provides a more reliable basis for decision-making in 3D reconstruction systems and reduces the risk of industrial accidents caused by data distortion.

[0063] 6. Enhanced adaptability to industrial scenarios: The configurable weight coefficient design and dynamically adjustable multi-level cache thresholds enable the system to flexibly adapt to the differentiated needs of different industrial fields (such as real-time priority or storage cost-sensitive scenarios), expanding the application scope of the technical solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a flow chart of a data acquisition method based on a three-dimensional reconstruction system according to the present invention;

[0065] Figure 2 A flow chart illustrating the implementation of the dynamic caching strategy of the present invention;

[0066] Figure 3 This is a principle block diagram of the data storage system based on Internet technology described in the present invention. DETAILED DESCRIPTION

[0067] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0068] like Figure 1 As shown, the present invention provides a technical solution: a data acquisition method based on a three-dimensional reconstruction system, comprising the following steps:

[0069] S1. Data acquisition: Configuring multiple data sources for the 3D reconstruction system. These data sources are used to collect raw data related to 3D reconstruction.

[0070] S11. Configure multiple data sources. The total number of data sources is ,in Is a positive integer, and each data source corresponds to an industrial device or sensor, numbered ;

[0071] S12. For each data source , at a specific time mark , capturing raw data in real time through the Industrial Internet of Things network ,Should Contains the following information:

[0072] Data source identification code , used to uniquely distinguish data sources ;

[0073] Measurement value or data content , representing the data source In time The original information collected;

[0074] Data format or type description , indicating the data source The structure, type, or format of the data produced, to facilitate the identification and processing of data from multiple sources and formats;

[0075] Additional Information , which covers additional information related to the data, such as units, precision, device status, and other necessary descriptive details;

[0076] S13, the data from all data sources in step S12 Collected raw data Aggregate, build in time The original dataset ,Right now: ;

[0077] in: Indicates time data collection activities; Indicates taking all data sources In time The raw data generated Collection of;

[0078] S14, the original data set Perform time stamp alignment and data integrity checks to ensure:

[0079] Time alignment: all data All with a precise time stamp Correlation to meet the needs of real-time data processing;

[0080] Data integrity: Checking key data source collections , ensuring that at every time , the corresponding original data can be obtained from all key data sources;

[0081] S15. Calculate data capture rate , to evaluate the time The effectiveness of data collection, the data capture rate is defined as: ;

[0082] S16: Based on the data capture rate and data integrity check results, determine whether the data collection process meets the real-time and reliability standards. If the conditions are met, the original data set is Perform data preprocessing.

[0083] Through S11 to S16, real-time collection and integration of data from multiple sources and formats are achieved, providing a comprehensive and reliable raw data foundation for subsequent data preprocessing, analysis and decision-making; by introducing data format or type description and additional information , which enhances the system's ability to identify and process heterogeneous data; through time mark alignment and data integrity verification, it ensures the real-time and reliability of data, meeting the requirements of the industrial Internet of Things environment for efficient data processing; data capture rate The calculation of helps to monitor and evaluate the effectiveness of the data collection process in real time, and provides a basis for the adaptive adjustment and optimization of the system.

[0084] S2. Data preprocessing: real-time received data Preprocess and convert into a standardized data set , this preprocessing process covers:

[0085] Data cleaning: Eliminate abnormal, duplicate and erroneous data;

[0086] Format unification: convert data in different formats into a unified standardized format;

[0087] Feature extraction: Use PCA algorithm to extract key feature vectors ;

[0088] Data compression: Implement lossless or lossy compression techniques to reduce storage and transmission costs;

[0089] Through these preprocessing steps, the original data Cleaned, standardized, featurized, and compressed into a more consistent and usable standardized dataset , providing strong support for subsequent data analysis and decision-making.

[0090] S3. Importance score calculation: for preprocessed data , calculate its importance score , the calculation formula is as follows:

[0091] ;

[0092] in:

[0093] Representation data Frequency of access: Frequently accessed data is usually of high importance in the system. Prioritizing caching of such data can improve system response speed and user experience. The number of data accesses within the statistical period can be used to measure this metric.

[0094] The urgency of the data; that is, the necessity of timely data processing. Data that requires real-time response or is associated with emergency events should be assigned a higher urgency value. This indicator can be set according to business needs, event levels, or preset rules;

[0095] The data change rate is used to measure the magnitude of data change in unit time. The calculation formula is: ;in, The data with a high rate of change indicates that the system status or environment has changed significantly, and it is important to obtain and process such data in a timely manner.

[0096] To improve the prediction accuracy of deep learning prediction models, we use a spatiotemporal prediction model based on graph convolutional networks to consider the temporal and spatial correlation of data.

[0097] is the correlation between data and key business indicators, and the calculation formula is: ; represents the correlation coefficient, is the eigenvector, Key Performance Indicator (KPI) reflects the impact of data on key business indicators. The higher the relevance, the more important the data is, and the data should be cached first.

[0098] To ensure data credibility, the reliability of the data source and the integrity of the data itself are evaluated through blockchain technology; distributed ledgers are used to record data source information, achieve data traceability and tamper-proofing, and ensure the authenticity and reliability of the data;

[0099] 、 、 、 、 、 is the weight coefficient; =1, and each coefficient is greater than or equal to zero. Each weight coefficient is used to balance the influence of different evaluation indicators on the importance of data;

[0100] Through the comprehensive calculation of the above indicators, the overall importance score of the data is obtained In actual applications, the weight coefficients can be adjusted according to specific business requirements and system conditions. 、 、 、 、 、 To optimize the cache strategy. For example, for a system with high real-time requirements, you can increase and For applications with high data credibility requirements, the weight of The weight of

[0101] Importance Rating It is the basis for the subsequent implementation of dynamic caching strategies. In step S4, the system will determine the data caching, elimination and scheduling strategies based on the scoring results to ensure efficient acquisition of key data and optimal utilization of system resources.

[0102] S4. Based on importance scoring ,implement dynamic caching strategies;

[0103] like Figure 2 As shown in the figure, the specific steps to implement the dynamic caching strategy are:

[0104] S41. Cache storage: scoring importance Exceeding the threshold The data is stored in the cache; that is, Data ,For the data that meets the requirements, it will be saved in the cache first to ensure the ,fast access and processing of key data and improve the ,system response speed.

[0105] S42, cache elimination: when the cache capacity Exceeding the threshold When using the LVD algorithm, press Remove data from low to high order; the specific steps are as follows:

[0106] S421. Score the data in the cache according to its importance Sort in ascending order;

[0107] S422, starting with the data with the lowest score, remove them in sequence until the cache capacity is reached. ;

[0108] S423. Ensure that data with high importance scores is always retained in the cache to ensure efficient access to key data by the system.

[0109] S43, multi-level cache: according to The data can be stored in different levels of cache media for multi-level caching. The multi-level cache includes at least three levels of cache media: memory cache, solid-state drive cache and disk cache, and is scored according to the importance of the data. Allocate data to the corresponding cache level;

[0110] Level 1 cache (memory cache): used to store data with the highest importance score and provide the fastest access speed;

[0111] Secondary cache (SSD cache): used to store data with medium importance scores, taking into account both access speed and storage capacity;

[0112] Level 3 cache (disk cache): used to store data with lower importance scores and provides larger storage capacity;

[0113] The cache level data allocation rules are as follows:

[0114] when When , the data is stored in the first-level cache;

[0115] when When , the data is stored in the secondary cache;

[0116] when When , the data is stored in the third-level cache;

[0117] in, and The importance score threshold is preset and meets .

[0118] S44. Intelligent Data Scheduling: Apply reinforcement learning algorithms to intelligently schedule cached data based on system status and network conditions, optimizing data acquisition and transmission paths. Specific implementation methods are as follows:

[0119] S441, state space definition: The system's cache status, network bandwidth, latency and other parameters constitute the state space to monitor the system environment in real time;

[0120] S442, action space definition: including operations such as migration, replication, and deletion of cached data, and selection of data transmission paths;

[0121] S443, Reward Function Design: Design a reward function based on system performance indicators such as cache hit rate, data access latency, and network load to guide the optimization direction of the reinforcement learning model;

[0122] S444, Strategy Update: Using reinforcement learning algorithms such as Q-learning, we continuously update the scheduling strategy to select the optimal scheduling action under different system states and improve overall performance;

[0123] Through adaptive adjustment of reinforcement learning, the system can dynamically optimize data scheduling strategies in complex and changing network environments, thereby maximizing data access efficiency and network resource utilization.

[0124] S45, Adaptive Weight Update: Using Bayesian optimization technology, the weight coefficient is adjusted in real time according to the cache hit rate H, system latency L and data loss rate D;

[0125] The calculation formula for real-time adjustment of weight coefficient is:

[0126]

[0127] Cache miss rate : reflects the probability of cache miss, H is the cache hit rate;

[0128] System latency L: refers to the average latency time for data access;

[0129] Data loss rate D: refers to the proportion of data loss caused by cache elimination or network problems;

[0130] in, 、 、 is the weight parameter, and , used to balance the importance of different performance indicators.

[0131] Bayesian optimization steps:

[0132] Establishing a proxy model: Using Gaussian processes or other regression models, establish a mapping relationship between weight coefficients and performance indicators;

[0133] Collect sample data: Measure the corresponding performance index values ​​under different weight coefficient combinations to enrich the training data of the agent model;

[0134] Select the optimal point: Through the prediction of the surrogate model, find the weight combination that minimizes the objective function in the weight coefficient space;

[0135] Update weight coefficients: Apply new optimal weight coefficients to importance score calculations and implement new caching strategies;

[0136] Through the above process, the system can adaptively adjust the weight coefficient of each factor, so that the cache strategy always remains in the optimal state, adapts to real-time changing system requirements and network environment, improves cache hit rate, and reduces system latency and data loss rate.

[0137] Through the implementation of the above dynamic caching strategy, the system can realize multi-level cache storage and intelligent data scheduling according to the importance score of the data, make full use of cache resources, and improve the efficiency and reliability of data access; at the same time, the adaptive weight update mechanism based on Bayesian optimization ensures the dynamic adjustment of the importance score, so that the caching strategy can adapt to the complex changes in the industrial Internet environment and continuously optimize system performance.

[0138] like Figure 3 As shown in the figure, the data storage system based on Internet technology is applied to a data acquisition method based on a three-dimensional reconstruction system, including: a data acquisition module, a data preprocessing module, a scoring calculation module, a dynamic cache management module, an intelligent scheduling module and an adaptive optimization module; the data acquisition module obtains real-time data of industrial equipment through sensor networks and IoT devices. ;The data preprocessing module is used to process real-time data Perform preprocessing to obtain preprocessed data ; The scoring calculation module is used to calculate the importance score of the data The dynamic cache management module is used to execute dynamic cache strategies, including data storage, elimination, and multi-level cache management. The intelligent scheduling module uses reinforcement learning algorithms to achieve intelligent data scheduling and transmission path optimization. The adaptive optimization module uses Bayesian optimization to achieve real-time updates of weight coefficients.

[0139] The dynamic cache management module is used to implement dynamic cache strategies, including data storage, elimination, and multi-level cache management:

[0140] Cache storage: scoring importance Exceeding the threshold The data is stored in the cache area; in this way, key data can be cached first, improving the system's response speed and data processing efficiency;

[0141] Cache eviction: When cache capacity Exceeding the threshold When using the LVD (lowest value data) algorithm, the importance score is Remove data in descending order. This strategy ensures that high-importance data is cached while freeing up space for new data.

[0142] Multi-level cache: scoring by importance Due to the differences in cache media, data can be stored in different levels of cache media to achieve multi-level cache. Multi-level cache includes at least three levels of cache media: memory cache, solid-state drive cache, and disk cache. Specifically:

[0143] Memory cache: used to store data with the highest importance score and provide the fastest access speed.

[0144] SSD cache: used to store data with medium importance scores, balancing speed and capacity.

[0145] Disk cache: used to store data with lower importance scores and provides larger storage capacity.

[0146] Based on a three-dimensional reconstruction system data acquisition device, a three-dimensional reconstruction system data acquisition method program is stored in the three-dimensional reconstruction system data acquisition device. When the three-dimensional reconstruction system data acquisition method program is executed by a processor, the steps of a three-dimensional reconstruction system data acquisition method are implemented.

[0147] Data acquisition and preprocessing: Through program instructions, control the data acquisition module and data preprocessing module to complete the acquisition and preprocessing of real-time data;

[0148] Importance score calculation: Call the program of the score calculation module to realize the importance score of the data Calculation of

[0149] Dynamic cache policy execution: Use program control to control the dynamic cache management module to perform cache storage, elimination and multi-level cache operations;

[0150] Intelligent data scheduling: Through the program scheduling intelligent scheduling module, reinforcement learning algorithm is applied to achieve intelligent scheduling and transmission optimization of cached data;

[0151] Adaptive weight adjustment: The program that calls the adaptive optimization module uses Bayesian optimization technology to adjust the weight coefficients in the score calculation in real time.

[0152] To sum up, the data acquisition method based on the three-dimensional reconstruction system provided in this embodiment realizes the dynamic optimization and adaptive adjustment of the data caching strategy through the organic combination and coordinated operation of various functional modules. It can efficiently integrate multi-source heterogeneous data, improve data compatibility and utilization, ensure data real-time and reliability, reduce three-dimensional reconstruction errors, optimize cache resource allocation, improve cache hit rate, reduce system latency, have adaptive dynamic optimization capabilities, balance storage costs and resource utilization, improve data quality and credibility, provide a reliable decision-making basis for three-dimensional reconstruction, and can flexibly adapt to the needs of different industrial scenarios and expand the scope of application.

[0153] While 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 these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A data acquisition method based on a three-dimensional reconstruction system, characterized in that: The following steps are involved: S1. Data acquisition: Configuring multiple data sources for the 3D reconstruction system. These data sources are used to collect raw data related to 3D reconstruction. S2. Data preprocessing: real-time received data Preprocess and convert into a standardized data set , this preprocessing process covers: Data cleaning: Eliminate abnormal, duplicate and erroneous data; Format unification: convert data in different formats into a unified standardized format; Feature extraction: Use PCA algorithm to extract key feature vectors ; Data compression: Implement lossless or lossy compression techniques to reduce storage and transmission costs; S3. Importance score calculation: for preprocessed data , calculate its importance score , the calculation formula is as follows: ； in: The frequency of data access; The urgency of the data; is the data change rate, and the calculation formula is: ;in, is the time interval; For the prediction importance based on deep learning prediction model; is the correlation between data and key business indicators, and the calculation formula is: ; represents the correlation coefficient, is the eigenvector, Key performance indicators (KPIs); Data credibility, reflecting the reliability of the data source and the integrity of the data itself; 、 、 、 、 、 is the weight coefficient; S4. Based on importance scoring ,implement dynamic caching strategies; In S1, the specific steps of data acquisition are: S11. Configure multiple data sources. The total number of data sources is ,in Is a positive integer, and each data source corresponds to an industrial device or sensor, numbered ; S12. For each data source , at a specific time mark , capturing raw data in real time through the Industrial Internet of Things network ,Should Contains the following information: Data source identification code , used to uniquely distinguish data sources ; Measurement value or data content , representing the data source In time The original information collected; Data format or type description , indicating the data source The structure, type, or format of the data produced, to facilitate the identification and processing of data from multiple sources and formats; Additional Information , covers additional information related to the data; S13, the data from all data sources in step S12 Collected raw data Aggregate, build in time The original dataset ,Right now: ; in: Indicates time data collection activities; Indicates taking all data sources In time The raw data generated Collection of; S14, the original data set Perform time stamp alignment and data integrity check; S15. Calculate data capture rate , to evaluate the time The effectiveness of data collection, the data capture rate is defined as: ; S16: Based on the data capture rate and data integrity check results, determine whether the data collection process meets the real-time and reliability standards. If the conditions are met, the original data set is Perform data preprocessing; Based on importance scoring ,The specific steps to implement the dynamic caching strategy are: S41. Cache storage: scoring importance Exceeding the threshold The data is stored in the cache area; S42, cache elimination: when the cache capacity Exceeding the threshold When using the LVD algorithm, press Remove data in order from low to high; S43, multi-level cache: according to Due to the difference in cache media, data can be stored in different levels of cache media for multi-level caching; S44, Intelligent Data Scheduling: Apply reinforcement learning algorithms to intelligently schedule cached data based on system status and network conditions, optimizing data acquisition and transmission paths; S45, Adaptive weight update: Using Bayesian optimization technology, the weight coefficient is adjusted in real time according to the cache hit rate H, system latency L and data loss rate D.

2. The data acquisition method based on a three-dimensional reconstruction system according to claim 1, characterized in that: Predicting Importance Use a spatiotemporal prediction model based on graph convolutional networks.

3. The data acquisition method based on a three-dimensional reconstruction system according to claim 2, characterized in that: Data credibility Evaluation is carried out through blockchain technology, and distributed ledgers are used to record data source information to achieve data traceability and tamper-proofing.

4. The data acquisition method based on a three-dimensional reconstruction system according to claim 3, characterized in that: The multi-level cache includes at least three levels of cache media: memory cache, solid state drive cache and disk cache, and is scored according to the importance of the data. Allocate data to the corresponding cache level.

5. The data acquisition method based on a three-dimensional reconstruction system according to claim 4, characterized in that: The cache level data allocation rules are as follows: when When , the data is stored in the memory cache; when When storing data in the SSD cache; when When , the data is stored in the disk cache; in, and The importance score threshold is preset and meets .

6. The data acquisition method based on a three-dimensional reconstruction system according to claim 5, characterized in that: The calculation formula for real-time adjustment of weight coefficient is: ； in, 、 、 is the weight parameter, and .

7. A data acquisition system based on a three-dimensional reconstruction system, which is applied to the data acquisition method based on a three-dimensional reconstruction system according to claim 6, characterized in that: include: Data acquisition module: obtain real-time data of industrial equipment through sensor networks and IoT devices ; Data preprocessing module is used to process real-time data Perform preprocessing to obtain preprocessed data ; Score calculation module, used to calculate the importance score of data ; Dynamic cache management module, used to implement dynamic cache strategies, including data storage, elimination and multi-level cache management; Intelligent scheduling module, using reinforcement learning algorithms to achieve intelligent data scheduling and transmission path optimization; The adaptive optimization module uses Bayesian optimization to achieve real-time update of weight coefficients.

Images