A smart station data acquisition method and system based on digital twin

Abstract

The present invention relates to the field of station data collection technology, and specifically to a smart station data collection method and system based on digital twins. The present invention first obtains the process attention of the process link according to the number of visits and access duration of all accounts to the process link in each operation record and the reference weight of the account; obtains the predecessor influence of the target link according to the abnormal situation of the monitoring time series data corresponding to each process link before the target link and the time interval between the process link and the target link; obtains the process monitoring degree of the target link according to the predecessor influence and process attention of the target link; adjusts the preset sampling frequency according to the process monitoring degree of the target link, and obtains the adjusted sampling frequency of the monitoring equipment in the target link. The present invention realizes the reasonable setting of the sampling frequency by deeply digging into the degree of monitoring required for the process link to meet the monitoring needs and ensure the effective use of monitoring resources.

Description

Technical Field

[0001] The present invention relates to the technical field of site data collection, and in particular to a smart site data collection method and system based on digital twins. Background Art

[0002] Smart stations are digital twin platforms that use advanced automation control systems and digital twin technology to build smart stations. The digital twin platform can monitor the status of equipment in real time, collect monitoring equipment data, and achieve remote control and management. Setting the sampling frequency of monitoring equipment in the digital twin platform is a key link. If the sampling frequency is set too high, although more detailed data changes can be captured, it will indeed lead to a waste of resources; if the sampling frequency is set too low, it will lead to the omission of key data, making it difficult to detect potential problems or failures of equipment in a timely manner.

[0003] The existing technology is to collect monitoring equipment data by setting the sampling frequency of monitoring equipment through experience. In the process of sampling smart stations using existing technology, the monitoring demand of monitoring equipment in the process links of smart stations is not deeply explored, resulting in unreasonable setting of the sampling frequency of the equipment, which makes it difficult to meet the monitoring needs while ensuring the effective use of resources. Summary of the invention

[0004] In order to solve the technical problem of unreasonable setting of sampling frequency of equipment in the prior art, the purpose of the present invention is to provide a smart station data acquisition method and system based on digital twins. The technical solutions adopted are as follows:

[0005] A method for collecting data of a smart station based on digital twins, the method comprising:

[0006] Obtain the operation process data set and historical operation record set of the digital twin platform of the smart site; the operation process data set includes the corresponding monitoring time series data of the monitoring equipment in the process link of the operation process to be analyzed; the historical operation record set includes the operation record of the account and the reference weight of the account; the operation record includes the number of visits and the duration of visits to each process link;

[0007] The process link to which the last moment of the running process to be analyzed belongs is taken as the target link; the process attention degree of the target link is obtained according to the number of visits and the visit duration of all accounts to the target link in each operation record and the reference weight of the account; the predecessor influence degree of the target link is obtained according to the abnormal conditions of all monitoring time series data in each process link before the target link and the time interval between the process link and the target link; the process monitoring degree of the target link is obtained according to the predecessor influence degree and the process attention degree of the target link;

[0008] According to the process monitoring degree of the target link, the preset sampling frequency of the monitoring device in the target link is adjusted to obtain the adjusted sampling frequency of the monitoring device in the target link.

[0009] Furthermore, the method for obtaining the process attention degree includes:

[0010] The forward fusion account records the number of visits and the duration of visits to the target link in each operation, and obtains the local attention index of the account to the target link;

[0011] The local attention index is weighted according to the reference weight of each account to obtain the process attention degree of the target link.

[0012] Furthermore, the method for obtaining the local attention index includes:

[0013] The product of the number of visits and the visit duration of each operation record to the target link is calculated, and the cumulative sum of the products of all operation records to the target link is calculated to obtain the local attention index of the account to the target link.

[0014] Furthermore, the method of weighting the local attention index according to the reference weight of each account to obtain the process attention of the target link includes:

[0015] The reference weights of all accounts are used to perform weighted averaging on the local attention index to obtain the process attention degree of the target link.

[0016] Furthermore, the method for obtaining the preceding influence degree includes:

[0017] Take each process link before the target link in the operation process to be analyzed as each reference link of the target link;

[0018] Accumulate the abnormal score values ​​of all monitoring time series data in each reference link to obtain the reference abnormal index of each reference link;

[0019] According to the time interval between the reference link and the target link, a distance index of the reference link is obtained;

[0020] The preceding influence degree of the target link is obtained according to the reference abnormality index and the distance index of all the reference links of the target link.

[0021] Furthermore, the method for obtaining the abnormality score value includes:

[0022] According to the corresponding monitoring time series data of each monitoring device in the reference link, the CNN neural network is used to perform abnormal scoring processing to determine the abnormal score value of each monitoring time series data in the reference link.

[0023] Furthermore, the method for obtaining the preceding influence of the target link according to the reference abnormality index and the distance index of all the reference links of the target link includes:

[0024] The ratio of the reference abnormality index to the distance index of the reference link is calculated to obtain the abnormality impact local index of the reference link; the cumulative value of the abnormality impact local index of all the reference links of the target link is calculated to obtain the predecessor influence degree of the target link.

[0025] Furthermore, the method for obtaining the process monitoring degree includes:

[0026] The product of the preceding influence degree and the process attention degree of the target link is calculated, and the product is normalized to obtain the process monitoring degree of the target link.

[0027] Furthermore, the method for adjusting the sampling frequency includes:

[0028] Calculate the product of the process monitoring degree of the target link and the preset adjustment parameter to obtain the adjustment parameter of the target link; calculate the product of the adjustment parameter and the preset sampling frequency of the monitoring device in the target link to obtain the adjustment sampling frequency of the monitoring device in the target link.

[0029] The present invention proposes a smart station data acquisition system based on digital twins, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a smart station data acquisition method based on digital twins are implemented.

[0030] The present invention has the following beneficial effects:

[0031] The present invention analyzes the degree to which the process link needs to be monitored, thereby adjusting the sampling frequency according to the degree to which the process link needs to be monitored, and realizing the reasonable setting of the sampling frequency to meet the monitoring requirements and ensure the effective use of monitoring resources. In order to perform real-time process link analysis, considering that the last moment of the running process to be analyzed reflects the current moment, the process link corresponding to the current moment is reflected through the target link. The degree of attention of the station to the target link is reflected by constructing the process attention degree. The higher the process attention degree, the higher the degree of attention to the target link, and the higher the importance of the target link in the station. Considering that the abnormal situation of the process link before the target link will affect the target link, the degree of abnormal influence of the previous process on the target link is reflected through the previous influence degree of the target link. The larger the value of the previous influence degree, the greater the degree of abnormal influence of the previous process, and the more important it is to monitor the target link. The process monitoring degree of the target link is obtained by combining the predecessor influence and process attention of the target link. The larger the value of the process monitoring degree, the more important the target link is to be monitored. Therefore, the preset sampling frequency of the monitoring equipment in the target link is adjusted according to the process monitoring degree, and the adjusted sampling frequency of the monitoring equipment in the target link is obtained, so as to achieve reasonable setting of the sampling frequency to meet the monitoring needs and ensure the effective use of monitoring resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 paying creative work.

[0033] Figure 1 A flowchart of a smart station data collection method based on digital twins provided by an embodiment of the present invention;

[0034] Figure 2 A flow chart of a method for obtaining process attention provided by an embodiment of the present invention;

[0035] Figure 3 A flow chart of a method for obtaining a preceding order influence degree provided by an embodiment of the present invention;

[0036] Figure 4 A structural diagram of a smart station data acquisition system based on digital twins provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a smart station data acquisition method and system based on digital twins proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0038] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0039] The specific scheme of a smart station data collection method and system based on digital twin provided by the present invention is described in detail below with reference to the accompanying drawings.

[0040] The embodiment of the present invention provides a method and system for collecting data of a smart station based on digital twins. Figure 1 , which shows a flow chart of a smart station data collection method based on digital twins provided by an embodiment of the present invention, the method comprising the following steps:

[0041] Step S1: Obtain the operation process data set and historical operation record set of the digital twin platform of the smart station; the operation process data set includes the corresponding monitoring time series data of the monitoring equipment in the process link of the operation process to be analyzed; the historical operation record set includes the operation records of the account and the reference weight of the account; the operation record includes the number of visits and visit duration to each process link.

[0042] The main purpose of this solution is to reasonably set the sampling frequency of the monitoring equipment in the station. The present invention is explained by taking the reasonable setting of the sampling frequency of the monitoring equipment of the smart station as an example.

[0043] Considering that the operation process of the station is a process of sequentially performing each process link in accordance with the process sequence, in each process link, monitoring equipment in the process link is used to collect monitoring values ​​to achieve the purpose of monitoring the process status. The present invention takes the operation process of a substation as an example. The operation process sequentially performs the power supply process link, the transmission process link, the distribution process link, and the transformation process link. (1) Power supply process link: refers to the transmission of high-voltage electric energy from the power plant to the substation, and the electric energy is transmitted in the form of high voltage during the transmission process. (2) Transmission process link: refers to the transmission of high-voltage electric energy from the substation to distribution stations or users in different regions. (3) Distribution process link: refers to the conversion of high-voltage electric energy from the transmission line into low-voltage electric energy suitable for users. (4) Transformation process link: refers to the increase of low-voltage electric energy to high-voltage electric energy suitable for users. The substation increases the low-voltage electric energy to a suitable voltage through a transformer, and then transmits it to other substations or users. In each process link, the voltage monitoring equipment set in the link is used to collect voltage monitoring values ​​to monitor the power status and ensure the stable transmission of power. For example, in the power supply process link, the incoming circuit breaker is a key device that connects the external power supply and the substation. It is used to control the access and disconnection of the power supply. It is necessary to set up a voltage monitoring device to monitor the power status. In the transmission process link, voltage monitoring equipment is set at the busbar, transmission line interval equipment, etc. to monitor the power status. In the distribution process link, voltage monitoring equipment is set at the distribution transformer to monitor the power status. In the transformation process link, voltage monitoring equipment is set at the main transformer to monitor the power status.

[0044] In order to monitor the status of monitoring equipment in the process link of the station and remotely control the station, digital twin technology is used to build a comprehensive platform for digital mapping and simulation of a physical smart station, and a digital twin platform of the smart station is obtained. The digital twin platform constructs the equipment layout of the station through three-dimensional modeling, and displays the real-time data of the monitoring equipment in each process link through a visual interactive interface. It should be noted that digital twin technology is a technical means well known to those skilled in the art and will not be elaborated here.

[0045] From the database of the digital twin platform of the smart station, a historical operation record set is obtained. Specifically, each staff member corresponds to an account. After the staff member logs in to the account, through the visual interactive interface of the digital twin platform of the station, the operation status of each monitoring device in each process link of the station can be viewed in real time. Each time the account logs in to the account and logs out is regarded as an operation record of the account, and the total duration of accessing the process link corresponding to the visual interactive interface in the operation record is regarded as the access duration of the operation record of the account; the total number of times the process link corresponding to the visual interactive interface in the operation record is accessed is regarded as the number of times the operation record of the account is accessed. In one embodiment of the present invention, the method for obtaining the reference weight of the account includes: the historical operation data set also includes the account level of the account; the account level is normalized to obtain the reference weight of the account. For example, the digital twin smart platform stores the account level of each account, that is, the account level ranges from 1 to 5, where 5 represents the highest level, which means that the staff is most familiar with the operation of the equipment. When evaluating the equipment access mode or predicting equipment failure, the access data of high-level accounts will be given higher importance. In one implementation of the present invention, the operation records of all accounts corresponding to the preset reference time interval and the reference weights of the accounts are taken as a historical operation record set. Among them, the preset reference time interval is 30 days, and the last moment of the preset reference time interval is the current moment, which can be set by the implementer according to the implementation scenario. It should be noted that the normalization method is: normalization is performed using a normalization function to limit the value range to between 0 and 1. Among them, normalization is a technical means well known to those skilled in the art, and the choice of normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0046] From the database of the digital twin platform of the smart station, obtain the operation process data set. Specifically, for each operation process, the corresponding monitoring values ​​of each monitoring device in the operation process are counted in turn to obtain the corresponding monitoring time series data of the monitoring device in the operation process. In order to adjust the sampling frequency in real time, the present invention takes the current operation process as the operation process to be analyzed, that is, the operation process corresponding to the current moment to the start moment of the operation process is taken as the operation process to be analyzed, that is, the last moment of the operation process to be analyzed is the current moment. The operation process data set includes the corresponding monitoring time series data of the monitoring equipment in the process link of the operation process to be analyzed. The database also stores the corresponding monitoring time series data of the monitoring equipment in the process link of each historical operation process.

[0047] It should be noted that, in order to facilitate calculation, all indicator data involved in the calculation in the embodiment of the present invention are preprocessed to eliminate the dimension effect. The specific means of eliminating the dimension effect are technical means well known to those skilled in the art and are not limited here.

[0048] Step S2: Take the process link to which the last moment of the running process to be analyzed belongs as the target link; obtain the process attention degree of the target link according to the number of visits and visit duration of all accounts to the target link in each operation record and the reference weight of the account; obtain the predecessor influence degree of the target link according to the abnormal conditions of all monitoring time series data in each process link before the target link and the time interval between the process link and the target link; obtain the process monitoring degree of the target link according to the predecessor influence degree and process attention degree of the target link.

[0049] The present invention analyzes the degree of monitoring required for the process link, thereby adjusting the sampling frequency according to the degree of monitoring required for the process link, and realizing the reasonable setting of the sampling frequency to meet the monitoring requirements and ensure the effective use of monitoring resources. In order to perform real-time process link analysis, considering that the last moment of the running process to be analyzed reflects the current moment, the process link corresponding to the current moment is reflected through the target link. The degree of attention of the station to the target link is reflected by constructing the process attention degree. The higher the process attention degree, the higher the degree of attention to the target link, and the higher the importance of the target link in the station. Considering that the abnormal situation of the process link before the target link will affect the target link, the degree of abnormal influence of the previous process on the target link is reflected by the previous influence degree of the target link. The larger the value of the previous influence degree, the greater the degree of abnormal influence of the previous process, and the more the target link needs to be monitored. Combined with the previous influence degree and the process attention degree of the target link, the process monitoring degree of the target link is obtained. The larger the value of the process monitoring degree, the higher the importance of the target link to be monitored.

[0050] In order to reasonably set the sampling frequency corresponding to the process link corresponding to the current moment, considering that the last moment of the running process to be analyzed reflects the current moment, the process link corresponding to the current moment is reflected through the target link.

[0051] To analyze the attention level of the target link, please refer to Figure 2 , which shows a flow chart of a method for obtaining process attention in one embodiment of the present invention. Preferably, in one embodiment of the present invention, considering that the staff can view the operation status of each monitoring device in the process link of the station in real time through the visual interactive interface of the digital twin platform, when the staff often clicks on the visual interactive interface of the process link to view and the viewing time of the process link is long, it means that the staff has a high degree of attention to the process link, and the process link is more important, and considering that the reference weight of the account represents the account's familiarity with the operation of the equipment, it reflects the importance of the account's access data. The process attention of the target link is obtained through the number of visits and access duration of all accounts to the target link in each operation record and the reference weight of the account. The method for obtaining process attention includes:

[0052] Step S201: Obtain the account's local attention index to the target link based on the number of visits and visit duration of the account to the target link in each operation record.

[0053] Taking into account that the longer the access time is, the longer the time the account has accessed the corresponding visual interactive interface of the target link in the operation record in the historical process; the higher the number of visits is, the more frequently the account has accessed the corresponding visual interactive interface of the target link in the operation record in the historical process. The local attention index of the account to the target link is obtained. The larger the local attention index of the account to the target link is, the higher the account's attention to the target link is.

[0054] Preferably, in one embodiment of the present invention, the method for obtaining the local attention index includes:

[0055] The product of the number of visits and the visit duration of each operation record to the target link is calculated, and the cumulative sum of the products of all operation records to the target link is calculated to obtain the local attention index of the account to the target link. In one embodiment of the present invention, the formula for obtaining the local attention index includes:

[0056]

[0057] in, For account Target link Local attention indicators; For account No. Operation records for target links Number of visits; For account No. Operation records for target links Length of visit; For account The total number of all operation records.

[0058] Step S202: Obtain the process attention degree of the target link according to the local attention index of each account to the target link and the reference weight of each account.

[0059] In order to better analyze the importance of the target link in the station, considering that the reference weight of the account represents the account's familiarity with the operation of the equipment, reflecting the importance of the account's access data, the larger the account's local attention index to the target link, the higher the account's attention to the target link, and the process attention to the target link is obtained. The higher the process attention, the higher the attention to the target link, reflecting the higher importance of the target link in the station.

[0060] Preferably, in one embodiment of the present invention, the product of the account's local attention index to the target link and the account's reference weight is calculated to obtain the account's local account parameters to the target link; the mean of all accounts' local account parameters to the target link is calculated to obtain the process attention of the target link.

[0061] To analyze the degree to which the target link is affected by the previous process link, please refer to Figure 3 , which shows a flow chart of a method for obtaining a predecessor influence degree in one embodiment of the present invention. Preferably, considering that the operation process of the station is a process of performing each process link in sequence according to the process sequence, there is a propagation effect between the process links, that is, an abnormality in the predecessor process link will often affect the subsequent process links along the production process. The predecessor influence degree of the target link is used to reflect the degree to which the target link is affected by the abnormality of the predecessor process. The larger the value of the predecessor influence degree, the greater the degree of influence of the abnormality of the predecessor process, and the more important it needs to be monitored. For example, the more abnormal the corresponding monitoring time series data of the predecessor process link is, the more difficult it is to guarantee the data of the predecessor link, which affects the processing effect of the subsequent process link. At this time, the importance of monitoring the subsequent process link is higher. By increasing the sampling frequency of the subsequent process link, the potential impact caused by the predecessor process link can be discovered in time, so as to take measures to reduce losses. In one embodiment of the present invention, the method for obtaining the predecessor influence degree includes:

[0062] Step S211: each process link before the target link in the running process to be analyzed is used as each reference link of the target link.

[0063] Considering that during the operation of the station, each process link is carried out in a certain order. Therefore, in order to analyze the impact of the target link on the previous process link, all the process links before the target link in the operation process to be analyzed are first taken as reference links.

[0064] Step S212: Obtain a reference abnormality index for each reference link according to the abnormal conditions of all monitoring time series data in each reference link.

[0065] The reference abnormality index reflects the abnormal situation of the reference link. The larger the reference abnormality index is, the more abnormal the reference link is.

[0066] In one embodiment of the present invention, a method for obtaining a reference abnormality index includes: performing abnormality scoring processing on the monitoring time series data corresponding to each monitoring device in the reference link through a CNN neural network to determine the abnormality score value of each monitoring time series data in the reference link; calculating the cumulative value of the abnormality score values ​​of all monitoring time series data in the reference link to obtain a reference abnormality index for the reference link.

[0067] It should be noted that CNN neural network is a technical means well known to those skilled in the art, and will not be elaborated here. Taking the monitoring time series data corresponding to any monitoring device in the reference link as an example, a brief process of using CNN neural network to determine the abnormal score value of monitoring time series data is briefly described: Considering that the database of the digital twin platform includes the monitoring time series data corresponding to the monitoring device of the historical operation process. The monitoring time series data corresponds to the monitoring device as the target device, and the monitoring time series data of all historical operation processes of the target device is used as the reference data set. The CNN neural network is trained using the reference data set. Since the monitoring time series data in the reference data set reflects the monitoring situation information, the staff scores the monitoring time series data for abnormalities based on experience, so that each monitoring time series data has a corresponding abnormal score value. Then the annotated reference data set is divided into a training set and a validation set according to a preset ratio, and the reference data set is input into the CNN neural network for training. The loss function is cross entropy, and the gradient descent method is used until the loss function converges to obtain the trained CNN neural network; the monitoring time series data is input into the trained CNN neural network, and the CNN neural network automatically scores abnormalities to determine the abnormal score value of the monitoring time series data. In the embodiment of the present invention, the preset ratio is 7:3, and the implementer can set it according to the implementation scenario.

[0068] For the above steps, the anomaly score value reflects the degree of anomaly of the monitored time series data in the reference link, and the reference anomaly index of the reference link reflects the overall degree of anomaly of the reference link.

[0069] Step S213: Obtain the distance index of the reference link according to the time interval between the reference link and the target link.

[0070] The smaller the value of the distance index is, the smaller the time interval between the reference link and the target link is, which means the greater the influence of the reference link on the target link is.

[0071] In one embodiment of the present invention, the time interval corresponding to the start time of the reference link and the start time of the target link is used as the distance indicator of the reference link. In other embodiments of the present invention, the process links to be analyzed can be numbered in sequence according to the process sequence, and the absolute value of the difference between the reference link and the target link is used as the distance indicator of the reference link.

[0072] Step S214: Obtain the predecessor influence of the target link according to the reference abnormality index and distance index of all reference links of the target link.

[0073] The predecessor impact of the target link reflects the degree to which the target link is affected by the abnormality of the predecessor process. The larger the predecessor impact value is, the greater the degree of impact of the abnormality of the predecessor process is.

[0074] In one embodiment of the present invention, the ratio of the reference abnormality index to the distance index of the reference link is calculated to obtain the abnormal impact local index of the reference link; the cumulative value of the abnormal impact local index of all reference links of the target link is calculated to obtain the predecessor influence degree of the target link.

[0075] For the above steps, the attenuation effect of the abnormal degree and time interval of the reference link on the abnormal impact degree is considered to obtain the abnormal impact local index of the reference link. The abnormal impact local index reflects the influence degree of the reference link on the target link. Finally, the abnormal impact local index of all reference links of the target link is combined to obtain the predecessor impact degree of the target link. The predecessor impact degree comprehensively reflects the comprehensive degree of the target link being affected by the abnormalities of all its predecessor process links.

[0076] Preferably, considering that the higher the process attention is, the more important the process link is in the station, and the more important it is to monitor the process link; the larger the value of the predecessor influence is, the greater the degree of abnormal influence of the predecessor process is, and the more important it is to monitor the process link, the process monitoring degree of the target link is obtained by combining the predecessor influence and process attention of the target link. The larger the value of the process monitoring degree is, the more important it is to monitor the target link. In one embodiment of the present invention, the method for obtaining the process monitoring degree is:

[0077] The product of the pre-order influence and process attention of the target link is calculated, and the product is normalized to obtain the process monitoring degree of the target link. It should be noted that the normalization method is: normalization is performed using a normalization function to limit the value range to between 0 and 1. Among them, normalization is a technical means well known to those skilled in the art, and the selection of the normalization function can be standard normalization, etc. The specific normalization method is not limited here.

[0078] Step S3: According to the process monitoring degree of the target link, the preset sampling frequency of the monitoring device in the target link is adjusted to obtain the adjusted sampling frequency of the monitoring device in the target link.

[0079] Through the above steps, the process monitoring degree is constructed to reflect the degree to which the process link needs to be monitored, so as to adjust the preset sampling frequency of the monitoring equipment in the target link according to the process monitoring degree, obtain the adjusted sampling frequency of the monitoring equipment in the target link, and realize the reasonable setting of the sampling frequency to meet the monitoring needs and ensure the effective use of monitoring resources.

[0080] Preferably, in one embodiment of the present invention, the method for adjusting the sampling frequency includes:

[0081] Calculate the product of the process monitoring degree of the target link and the preset adjustment parameter to obtain the adjustment parameter of the target link; calculate the product of the adjustment parameter and the preset sampling frequency of the monitoring equipment in the target link to obtain the adjusted sampling frequency of the monitoring equipment in the target link. In one embodiment of the present invention, the preset adjustment parameter is 1.2, and the implementer can set it according to the implementation scenario. In one embodiment of the present invention, the method for setting the preset sampling frequency of the monitoring equipment in the target link is: obtain the standard sampling frequency of the monitoring equipment from the database of the digital twin platform of the smart site, and use the standard sampling frequency as the preset sampling frequency of the monitoring equipment in the target link. In other embodiments of the present invention, the preset sampling frequency of the monitoring equipment can also be set based on empirical values.

[0082] For the above steps, the adjustment parameter determines the specific adjustment range of the sampling frequency according to the degree of monitoring required in the process link. If the adjustment parameter is greater than 1, the preset sampling frequency will be increased to achieve more frequent monitoring; if the adjustment parameter is not greater than 1, the preset sampling frequency will be reduced, indicating that the monitoring frequency can be appropriately reduced to save resources and improve operating efficiency; in order to reasonably set the adjustment sampling frequency of the monitoring equipment in the target link, meet the monitoring needs and ensure the effective use of monitoring resources.

[0083] Specifically, the frequency of the monitoring device in the target link is set by utilizing the adjusted sampling frequency of the monitoring device in the target link.

[0084] The present invention also proposes a smart station data acquisition system based on digital twins, please refer to Figure 4 , which shows a structural diagram of a smart station data acquisition system based on digital twins provided by an embodiment of the present invention. The system includes: a data acquisition module 101, a process monitoring analysis module 102, and a frequency adjustment module 103.

[0085] The data acquisition module 101 is used to obtain the operation process data set and historical operation record set of the digital twin platform of the smart station; the operation process data set includes the corresponding monitoring time series data of the monitoring equipment in the process link of the operation process to be analyzed; the historical operation record set includes the operation records of the account and the reference weight of the account; the operation records include the number of visits and visit duration to each process link.

[0086] The process monitoring degree analysis module 102 is used to take the process link to which the last moment of the running process to be analyzed belongs as the target link; obtain the process attention degree of the target link according to the number of visits and visit duration of all accounts to the target link in each operation record and the reference weight of the account; obtain the predecessor influence degree of the target link according to the abnormal conditions of all monitoring time series data in each process link before the target link and the time interval between the process link and the target link; obtain the process monitoring degree of the target link according to the predecessor influence degree and process attention degree of the target link.

[0087] The frequency adjustment module 103 is used to adjust the preset sampling frequency of the monitoring device in the target link according to the process monitoring degree of the target link, and obtain the adjusted sampling frequency of the monitoring device in the target link.

[0088] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the smart station data acquisition system based on digital twins and the smart station data acquisition method based on digital twins provided in the above embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0089] In summary, the embodiment of the present invention provides a method and system for collecting data of a smart station based on digital twins. First, the process attention of the process link is obtained according to the number of visits and the access duration of all accounts to the process link in each operation record and the reference weight of the account; the predecessor influence of the target link is obtained according to the abnormal situation of the monitoring time series data corresponding to each process link before the target link and the time interval between the process link and the target link; the process monitoring degree of the target link is obtained according to the predecessor influence and process attention of the target link; according to the process monitoring degree of the target link, the preset sampling frequency of the monitoring equipment in the target link is adjusted to obtain the adjusted sampling frequency of the monitoring equipment in the target link. In the embodiment of the present invention, by deeply digging into the degree of monitoring required for the process link, the reasonable setting of the sampling frequency is achieved to meet the monitoring needs and ensure the effective use of monitoring resources.

[0090] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. 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.

[0091] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A smart station data collection method based on digital twins, characterized in that: The method comprises: Obtain the operation process data set and historical operation record set of the digital twin platform of the smart site; the operation process data set includes the corresponding monitoring time series data of the monitoring equipment in the process link of the operation process to be analyzed; the historical operation record set includes the operation record of the account and the reference weight of the account; the operation record includes the number of visits and the duration of visits to each process link; The process link to which the last moment of the running process to be analyzed belongs is taken as the target link; the process attention degree of the target link is obtained according to the number of visits and the visit duration of all accounts to the target link in each operation record and the reference weight of the account; the predecessor influence degree of the target link is obtained according to the abnormal conditions of all monitoring time series data in each process link before the target link and the time interval between the process link and the target link; the process monitoring degree of the target link is obtained according to the predecessor influence degree and the process attention degree of the target link; According to the process monitoring degree of the target link, a preset sampling frequency of the monitoring device in the target link is adjusted to obtain an adjusted sampling frequency of the monitoring device in the target link; The method for obtaining the reference weight of the account includes: the historical operation data set also includes the account level of the account; the account level is normalized to obtain the reference weight of the account; The method for obtaining the process attention degree includes: The forward fusion account records the number of visits and the duration of visits to the target link in each operation, and obtains the local attention index of the account to the target link; Weighting the local attention indicator according to the reference weight of each account to obtain the process attention degree of the target link; The method for obtaining the local attention index includes: Calculate the product of the number of visits and the visit duration of each operation record to the target link, calculate the cumulative sum of the products of all operation records to the target link, and obtain the local attention index of the account to the target link; The method for obtaining the preceding influence degree includes: Take each process link before the target link in the operation process to be analyzed as each reference link of the target link; Accumulate the abnormal score values ​​of all monitoring time series data in each reference link to obtain the reference abnormal index of each reference link; According to the time interval between the reference link and the target link, a distance index of the reference link is obtained; Acquire the preceding influence degree of the target link according to the reference abnormality index and the distance index of all the reference links of the target link; The method for obtaining the process monitoring degree includes: Calculate the product of the preceding influence degree and the process attention degree of the target link, normalize the product, and obtain the process monitoring degree of the target link; The method of weighting the local attention index according to the reference weight of each account to obtain the process attention degree of the target link includes: Using the reference weights of all accounts to perform weighted averaging on the local attention index, the process attention of the target link is obtained; The method for obtaining the abnormality score value includes: According to the monitoring time series data corresponding to each monitoring device in the reference link, the abnormality scoring process is performed through the CNN neural network to determine the abnormality score value of each monitoring time series data in the reference link; The method for obtaining the preceding influence of a target link according to the reference abnormality index and the distance index of all the reference links of the target link comprises: Calculate the ratio of the reference abnormality index to the distance index of the reference link to obtain the abnormality impact local index of the reference link; calculate the cumulative value of the abnormality impact local index of all the reference links of the target link to obtain the pre-order influence degree of the target link; The method for adjusting the sampling frequency includes: Calculate the product of the process monitoring degree of the target link and the preset adjustment parameter to obtain the adjustment parameter of the target link; calculate the product of the adjustment parameter and the preset sampling frequency of the monitoring device in the target link to obtain the adjustment sampling frequency of the monitoring device in the target link; The digital twin platform constructs the equipment layout of the site through three-dimensional modeling, and displays the real-time data of the monitoring equipment in each process link through a visual interactive interface.

2. A digital twin-based smart station data acquisition system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, it implements the steps of a smart station data collection method based on digital twins as described in claim 1.

Images