A method and system for collecting multi-source heterogeneous data at a construction site

Abstract

The present invention discloses a multi-source heterogeneous data collection method and system for a construction site, and belongs to the field of data processing technology. The method comprises the following steps: obtaining a data set; analyzing a cause influence coefficient; matching monitoring equipment; analyzing a to-sleep index; and controlling the working state of the equipment. The present invention analyzes the cause influence coefficient corresponding to each potential cause of a fault, and simultaneously obtains the matching index between the potential fault factor and the monitoring equipment, and adjusts the to-sleep index of the monitoring equipment in combination with the cause influence coefficient and the matching index, and controls the working state of the monitoring equipment to be monitored to sleep or work according to the to-sleep index and the set threshold, thereby achieving improved resource utilization efficiency and reducing the data processing amount of the monitoring equipment, solving the problem of low resource utilization efficiency in the prior art due to the lack of a dynamic adjustment mechanism for the construction environment and real-time needs.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for collecting multi-source heterogeneous data at a construction site. Background Art

[0002] With the continuous acceleration of urban power grid construction, power grid construction sites have become an important part of urban development. In-depth dismantling and analysis of related factors based on on-site construction violations are of great significance for ensuring on-site safety and improving project quality.

[0003] Currently, data collection at power grid construction sites usually involves real-time data collection of each link, including the working environment, construction team, work content, construction period, etc.

[0004] For example, the invention patent with announcement number: CN104750814B discloses an automatic warehousing method for multi-sensor based heterogeneous data streams, including: a processor receives a multi-sensor heterogeneous data stream collected by at least two sensors; the processor determines whether the sensor corresponding to the multi-sensor heterogeneous data stream exists in the sensor format information table, and if so, determines whether the multi-sensor heterogeneous data matches the sensor format information table, and if so, stores the multi-sensor heterogeneous data in a local stream database; if the sensor corresponding to the multi-sensor heterogeneous data stream does not exist in the sensor format information table, the format information of the sensor corresponding to the multi-sensor heterogeneous data stream is added to the sensor format information table.

[0005] For example, the invention patent with publication number: CN108984603A discloses a heterogeneous data collection method, device, storage medium and system, including: by querying the data field of the preset data directory, it is determined whether the preset data directory contains the corresponding data field; when the preset data directory contains the corresponding data field, the client obtains the interface implementation program of the heterogeneous data to be collected from the server, so that the client can directly obtain the corresponding data from the data provider through the interface implementation program. Therefore, when multiple data collection client systems collect various types of external heterogeneous data, there is no need to develop interfaces, and only the corresponding data interface implementation program needs to be obtained from the server to directly collect the external data.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, in order to achieve all-round real-time monitoring at the current power grid construction site, a large number of monitoring devices are usually deployed, and these devices need to run continuously to ensure construction safety and real-time data. However, in the actual construction process, not all areas or equipment require all-weather real-time monitoring, and not all functions must be turned on all the time. For example, in some low-risk areas, the monitoring frequency can be appropriately reduced; under certain specific conditions, such as unmanned construction or non-peak construction periods, some non-critical functions can be turned off. However, due to the lack of targeted management, current monitoring equipment often runs for a long time at a unified standard, resulting in a large amount of resources being consumed unnecessarily. This situation not only increases the operating costs of energy and equipment, but may also accelerate equipment aging and affect its service life. In summary, due to the lack of a dynamic adjustment mechanism for the construction environment and real-time needs in the prior art, there is a problem of low resource utilization efficiency. Summary of the invention

[0008] The embodiments of the present application provide a multi-source heterogeneous data collection method and system for a construction site, thereby solving the problem of low resource utilization efficiency caused by the lack of a dynamic adjustment mechanism for the construction environment and real-time needs in the prior art, thereby improving resource utilization efficiency and reducing the data processing volume of monitoring equipment.

[0009] The embodiment of the present application provides a multi-source heterogeneous data collection method for a construction site, comprising the following steps: acquiring a data set: acquiring historical multi-source heterogeneous data of a power grid construction site, fusing the historical multi-source heterogeneous data to form a historical event data set, and preprocessing the historical event data set; analyzing a cause influence coefficient: extracting a potential cause of a fault based on the preprocessed historical event data set, analyzing a cause influence coefficient corresponding to the potential cause of the fault, and the cause influence coefficient is used to measure the degree of influence of the potential cause of the fault on the occurrence of the fault; matching monitoring equipment: matching monitoring equipment according to potential factors of the fault, and analyzing a matching index between the potential factors of the fault and the monitoring equipment, and obtaining a monitoring equipment to be adjusted according to the matching index, and the matching index is used to reflect the probability that the potential factors of the fault are monitored by the monitoring equipment; analyzing a to-sleep index: analyzing a to-sleep index of a to-sleep monitoring equipment to be adjusted according to the cause influence coefficient and the matching index, and the to-sleep index is used to reflect the probability that the to-sleep monitoring equipment to be adjusted is asleep; controlling the working state of the equipment: comparing the to-sleep index with a set threshold in a database, and when the to-sleep index is not lower than the set threshold, controlling the to-sleep monitoring equipment to be adjusted to sleep, and when the to-sleep index is lower than the set threshold, controlling the to-sleep monitoring equipment to be adjusted to work.

[0010] Furthermore, the specific steps of fusing historical multi-source heterogeneous data to form a historical event data set include: obtaining historical multi-source heterogeneous data and standardizing it; temporally aligning the historical multi-source heterogeneous data according to timestamps; spatially aligning the historical multi-source heterogeneous data according to spatial positions; extracting event types from a database as event labels, and grouping the historical multi-source heterogeneous data according to the event labels; using a multimodal fusion algorithm to perform feature fusion on each group of historical multi-source heterogeneous data to obtain a historical event data set; and combining all historical event data groups into a historical event data set.

[0011] Furthermore, the specific steps of extracting potential causes of faults based on the preprocessed historical event data set include: data set division: dividing the preprocessed historical event data set into a training set and a test set; feature recursive splitting: selecting an optimal feature from the historical event data set in the training set to perform an initial split of the historical event data set to obtain multiple subsets, and recursively splitting each subset until a set stop condition is met; based on the result of the recursive splitting, generating a preliminary decision tree model, inputting the test set into the preliminary decision tree model for testing, and generating a decision tree model; and extracting potential causes of faults based on the generated decision tree model.

[0012] Furthermore, the specific method for obtaining the cause influence coefficient is as follows: obtaining the associated characteristic data of each potential cause of the fault from the preprocessed historical event data set, wherein the associated characteristic data includes the fault occurrence frequency, fault duration, temperature, and humidity; obtaining the weight of each data in the associated characteristic data for the cause influence coefficient by an objective weighting method; calculating the cause influence coefficient of each potential cause of the fault by the cause influence coefficient formula according to the associated characteristic data and the weight of each data in the associated characteristic data for the cause influence coefficient; the cause influence coefficient formula is:

[0013]

[0014] In the formula, is the cause influence coefficient of the i-th potential cause of the fault, e is a natural constant, i is the number of the potential cause of the fault, , I is the total number of potential causes of failure, is the failure frequency of the i-th potential cause of failure, for right The weight of is the fault duration of the i-th potential cause of the fault, for right The weight of is the temperature of the ith potential cause of failure, for right The weight of is the humidity of the potential cause of the ith fault, for right The weight of .

[0015] Furthermore, matching monitoring equipment according to potential fault factors specifically refers to: obtaining monitoring equipment labels corresponding to all associated characteristic data in potential fault causes; and matching and searching for corresponding monitoring equipment from a database based on the obtained monitoring equipment labels.

[0016] Furthermore, the specific method of obtaining the matching index is: extracting the monitoring device label contained in each potential cause of the fault, wherein the monitoring device label corresponds to the monitoring device one-to-one; extracting all data type labels that can be monitored by each monitoring device that is matched and found from the parameter library of the monitoring device; and analyzing the matching index between the monitoring device label contained in each potential cause of the fault and all data type labels that can be monitored by the monitoring device through the matching index formula.

[0017] Furthermore, the matching index formula is:

[0018]

[0019] In the formula, is the matching index between the i-th potential cause of the fault and the k-th monitoring device, i is the number of the potential cause of the fault, k is the label number of the monitoring device, that is, the number of the monitoring device, , K is the total number of monitoring device labels, that is, the total number of monitoring devices, The result of whether the g-th data type label in the k-th monitoring device corresponding to the i-th potential cause of the fault can be found in the associated feature data. If it can be detected, then If it cannot be detected, , g is the data type label, , G is the total number of data type labels, The total number of all data type labels that the k-th monitoring device can monitor, is the penalty factor.

[0020] Furthermore, the method for obtaining the to-sleep index is as follows: extract potential causes of faults whose cause influence coefficient is not higher than the first threshold in the database; extract matching indexes corresponding to the potential causes of faults and each monitoring device based on the extracted potential causes of faults; extract monitoring devices whose matching indexes between the potential causes of faults and the monitoring devices are higher than the second threshold in the database as monitoring devices to be adjusted; obtain the number of personnel at the location of the monitoring device to be adjusted; obtain the current load and the maximum allowable load of the monitoring device to be adjusted, and obtain the current workload ratio by the current load ratio to the maximum allowable load; obtain the weights of the cause influence coefficient, matching index, and current workload ratio for the to-sleep index respectively through an objective weighting method; based on the cause influence coefficient, matching index, number of personnel, and current workload ratio, analyze the to-sleep index of the monitoring device to be adjusted through the to-sleep index formula.

[0021] Furthermore, the sleep waiting index formula is:

[0022]

[0023] In the formula, is the sleep index of the kth monitoring device corresponding to the ith potential cause of the fault, e is a natural constant, k is the label number of the monitoring device, that is, the number of the monitoring device, , K is the total number of monitoring device labels, that is, the total number of monitoring devices, Is the result of judging whether the number of people is 0. If the number of people is 0, then , if the number of people is not 0, then , is the cause influence coefficient of the i-th potential cause of failure, is the matching index between the i-th potential cause of fault and the k-th monitoring device, is the current workload ratio of the i-th potential cause of the fault to the k-th monitoring device, i is the number of the potential cause of the fault, , I is the total number of potential causes of failure, for right The weight of for right The weight of for right The weight of .

[0024] The embodiment of the present application provides a multi-source heterogeneous data acquisition system for a construction site, which is used to implement a multi-source heterogeneous data acquisition method for a construction site, including: a data set acquisition module, a cause influence coefficient analysis module, a monitoring device matching module, a sleep index analysis module, and an equipment working status control module; wherein the data set acquisition module is used to acquire historical multi-source heterogeneous data of a power grid construction site, fuse the historical multi-source heterogeneous data to form a historical event data set, and pre-process the historical event data set; the cause influence coefficient analysis module is used to extract potential causes of faults based on the pre-processed historical event data set, analyze the cause influence coefficients corresponding to the potential causes of faults, and the cause influence coefficients are used to measure the impact of the potential causes of faults on the occurrence of faults. The monitoring device matching module is used to match the monitoring device according to the potential fault factors, and analyze the matching index between the potential fault factors and the monitoring device, and obtain the monitoring device to be adjusted according to the matching index, and the matching index is used to reflect the probability that the potential fault factor is monitored by the monitoring device; the sleep index analysis module is used to analyze the sleep index of the monitoring device to be adjusted according to the cause influence coefficient and the matching index, and the sleep index is used to reflect the probability that the monitoring device to be adjusted is asleep; the device working state control module is used to compare the sleep index with the set threshold value in the database, when the sleep index is not lower than the set threshold value, the monitoring device to be adjusted is controlled to sleep, and when the sleep index is lower than the set threshold value, the monitoring device to be adjusted is controlled to work.

[0025] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0026] 1. By analyzing the cause influence coefficient corresponding to each potential cause of the fault, and obtaining the matching index between the potential factor of the fault and the monitoring equipment, the sleep index of the monitoring equipment to be adjusted is adjusted in combination with the cause influence coefficient and the matching index. According to the sleep index and the set threshold, the working state of the monitored monitoring equipment is controlled to sleep or work, thereby improving the resource utilization efficiency and reducing the data processing volume of the monitoring equipment. This effectively solves the problem of low resource utilization efficiency in the existing technology due to the lack of a dynamic adjustment mechanism for the construction environment and real-time needs.

[0027] 2. By extracting the monitoring equipment label from each potential cause of the fault, and extracting the corresponding data type label from the parameter library of the monitoring equipment, combined with the matching index formula analysis, an accurate match between the potential cause of the fault and the monitoring equipment can be achieved, thereby improving the pertinence and accuracy of equipment monitoring. At the same time, the working status of the monitoring equipment can be dynamically adjusted according to actual needs, avoiding resource waste and ensuring high-quality collection of monitoring data.

[0028] 3. By obtaining the associated characteristic data of the potential causes of the fault and matching the corresponding monitoring equipment tags, it ensures that the fault cause is more closely related to the monitoring equipment, reduces the collection and processing of irrelevant data, and improves the pertinence of data collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flowchart of a method for collecting multi-source heterogeneous data at a construction site provided in an embodiment of the present application;

[0030] Figure 2 A schematic diagram of the structure of a multi-source heterogeneous data acquisition system for a construction site provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The embodiment of the present application provides a multi-source heterogeneous data collection method and system for a construction site, which solves the problem of low resource utilization efficiency in the prior art due to the lack of a dynamic adjustment mechanism for the construction environment and real-time needs. The method finds out the potential cause of the fault from the historical event data set based on a decision tree algorithm, and then analyzes the cause influence coefficient corresponding to each potential cause of the fault. At the same time, the matching index between the potential fault factor and the monitoring equipment is obtained to extract the monitoring equipment to be adjusted. The sleep index of the monitoring equipment to be adjusted is calculated based on the cause influence coefficient and the matching index. The working state of the monitoring equipment to be monitored is controlled to sleep or work according to the sleep index and the set threshold, thereby improving the resource utilization efficiency and reducing the data processing volume of the monitoring equipment.

[0032] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0033] like Figure 1As shown, it is a flow chart of a multi-source heterogeneous data collection method for a construction site provided by an embodiment of the present application. The method is applied to a multi-source heterogeneous data collection system for a construction site. The method includes the following steps: obtaining a data set: obtaining historical multi-source heterogeneous data of a power grid construction site, fusing the historical multi-source heterogeneous data to form a historical event data set, and preprocessing the historical event data set; analyzing the cause influence coefficient: extracting the potential cause of the fault based on the preprocessed historical event data set, analyzing the cause influence coefficient corresponding to the potential cause of the fault, and the cause influence coefficient is used to measure the degree of influence of the potential cause of the fault on the occurrence of the fault; matching monitoring equipment: matching the potential factors of the fault according to the potential factors of the fault Monitoring equipment, and analyzing the matching index between the potential fault factor and the monitoring equipment, and obtaining the monitoring equipment to be adjusted according to the matching index, the matching index is used to reflect the probability that the potential fault factor is detected by the monitoring equipment; analyzing the waiting sleep index: analyzing the waiting sleep index of the monitoring equipment to be adjusted according to the cause influence coefficient and the matching index, the waiting sleep index is used to reflect the probability that the current working state of the monitoring equipment to be adjusted can be adjusted to sleep; controlling the working state of the equipment: comparing the waiting sleep index with the set threshold in the database, when the waiting sleep index is not lower than the set threshold, controlling the monitoring equipment to be adjusted to sleep, and when the waiting sleep index is lower than the set threshold, controlling the monitoring equipment to be adjusted to work.

[0034] Furthermore, the specific steps of fusing historical multi-source heterogeneous data to form a historical event data set include: obtaining historical multi-source heterogeneous data and standardizing it; temporally aligning the historical multi-source heterogeneous data according to timestamps; spatially aligning the historical multi-source heterogeneous data according to spatial positions; extracting event types from the database as event labels, and grouping the historical multi-source heterogeneous data according to the event labels; using a multimodal fusion algorithm to perform feature fusion on each group of historical multi-source heterogeneous data to obtain a historical event data set; and combining all historical event data groups into a historical event data set.

[0035] Furthermore, the specific steps of extracting the potential cause of the fault based on the preprocessed historical event data set include: data set division: dividing the preprocessed historical event data set into a training set and a test set; feature recursive splitting: selecting an optimal feature from the historical event data set in the training set to perform the initial split of the historical event data set, obtaining multiple subsets, and recursively splitting each subset until the set stop condition is met. The stop condition includes that the depth of the tree reaches a predetermined value, the purity of the historical event data set after segmentation reaches a predetermined standard, or the size of the subset is lower than a set threshold, and the importance of each feature in distinguishing whether a fault occurs or not can also be reflected by information gain or Gini index.

[0036] Based on the results of recursive splitting, a preliminary decision tree model is generated, in which the decision tree model describes the influence of various characteristics (such as voltage, current, temperature, etc.) in the historical event data on the occurrence of faults through a tree structure.

[0037] The test set is input into the preliminary decision tree model for testing to generate a decision tree model; based on the generated decision tree model, the potential causes of the fault are extracted, where the potential causes of the fault are identified through the feature division of the decision tree nodes. For example, when the voltage, current or temperature exceeds a certain threshold, the probability of a fault occurring is higher.

[0038] Furthermore, the specific method for obtaining the cause influence coefficient is as follows: obtaining the associated characteristic data of each potential cause of the fault from the preprocessed historical event data set, the associated characteristic data including the fault occurrence frequency, fault duration, temperature, and humidity; obtaining the weight of each data in the associated characteristic data for the cause influence coefficient through the objective weighting method; calculating the cause influence coefficient of each potential cause of the fault through the cause influence coefficient formula according to the associated characteristic data and the weight of each data in the associated characteristic data for the cause influence coefficient; the cause influence coefficient formula is:

[0039]

[0040] In the formula, is the cause influence coefficient of the i-th potential cause of the fault, e is a natural constant, i is the number of the potential cause of the fault, , I is the total number of potential causes of failure, is the failure frequency of the i-th potential cause of failure, for right The weight of is the fault duration of the i-th potential cause of the fault, for right The weight of is the temperature of the ith potential cause of failure, for right The weight of is the humidity of the potential cause of the ith fault, for right The weight of .

[0041] In this embodiment, , , , , , , , , cause influence coefficient The larger the cause influence coefficient, the greater the influence of the potential factor of the fault on the occurrence of the fault, and vice versa.

[0042] Furthermore, matching monitoring equipment according to potential fault factors specifically refers to: obtaining monitoring equipment labels corresponding to all associated characteristic data in the potential fault causes; and matching and searching for corresponding monitoring equipment from a database based on the obtained monitoring equipment labels.

[0043] Furthermore, the specific method of obtaining the matching index is: extracting the monitoring device label contained in each potential cause of the fault, with the monitoring device label and the monitoring device corresponding one to one; extracting all data type labels that can be monitored by each monitoring device that is matched and found from the parameter library of the monitoring device; analyzing the matching index between the monitoring device label contained in each potential cause of the fault and all data type labels that can be monitored by the monitoring device through the matching index formula.

[0044] In this embodiment, by obtaining the associated characteristic data in the potential cause of the fault and matching the corresponding monitoring equipment label, it is ensured that the cause of the fault is more closely related to the monitoring equipment, reducing the collection and processing of irrelevant data, improving the targeted nature of data collection, and being able to more accurately capture data related to potential faults, thereby avoiding waste of resources.

[0045] Furthermore, the matching index formula is:

[0046]

[0047] In the formula, is the matching index between the i-th potential cause of the fault and the k-th monitoring device, i is the number of the potential cause of the fault, k is the label number of the monitoring device, that is, the number of the monitoring device, , K is the total number of monitoring device labels, that is, the total number of monitoring devices, The result of whether the g-th data type label in the k-th monitoring device corresponding to the i-th potential cause of the fault can be found in the associated feature data. If it can be detected, then If it cannot be detected, , g is the data type label, , G is the total number of data type labels, The total number of all data type labels that the k-th monitoring device can monitor, is the penalty factor, which is the value set in the database. The larger the value is, the smaller the match between the monitoring device and the potential cause of the fault is. If the data type label that the monitoring device can monitor corresponds to the associated characteristic data of the potential cause of the fault one by one, the matching index is the highest. For the monitoring device to be adjusted with a high matching index, even if the working mode is adjusted to the sleep mode, it will not affect the monitoring of other data types, and the impact on the overall safety is extremely small.

[0048] In this embodiment, , , , The higher the matching index, the better. The higher the matching index, the higher the degree of exclusive monitoring of the associated characteristic data of the potential cause of the fault by the monitoring device.

[0049] Furthermore, the method for obtaining the to-sleep index is as follows: extract potential causes of faults whose cause influence coefficient is not higher than the first threshold in the database; extract matching indexes corresponding to the potential causes of faults and each monitoring device based on the extracted potential causes of faults; extract monitoring devices whose matching indexes between the potential causes of faults and the monitoring devices are higher than the second threshold in the database as monitoring devices to be adjusted; obtain the number of personnel at the location of the monitoring device to be adjusted; obtain the current load and the maximum allowable load of the monitoring device to be adjusted, and obtain the current workload ratio by the current load ratio to the maximum allowable load; obtain the weights of the cause influence coefficient, matching index, and current workload ratio for the to-sleep index respectively through an objective weighting method; based on the cause influence coefficient, matching index, number of personnel, and current workload ratio, analyze the to-sleep index of the monitoring device to be adjusted through the to-sleep index formula.

[0050] In this embodiment, the sleep index can reflect the sleep possibility of the monitoring device. Therefore, under the premise of ensuring the monitoring needs, the working state of the monitoring device can be dynamically adjusted. This can minimize the energy consumption of the monitoring equipment, extend the service life of the equipment, and improve resource utilization while meeting the monitoring needs of the construction site. In addition, this dynamic control mechanism can reduce unnecessary data processing tasks, effectively alleviate the system load, and improve the overall data processing efficiency.

[0051] Furthermore, the sleep waiting index formula is:

[0052]

[0053] In the formula, is the sleep index of the kth monitoring device corresponding to the ith potential cause of the fault, Is the result of judging whether the number of people is 0. If the number of people is 0, then , if the number of people is not 0, then , is the current workload ratio of the kth monitoring device corresponding to the ith potential cause of the fault, for right The weight of for right The weight of for right The weight of .

[0054] In this embodiment, if the number of people is not 0, the monitoring device to be adjusted is not allowed to be adjusted to sleep mode to ensure the safety of the people at the location where the monitoring device to be adjusted is located. For the monitoring device to be adjusted, the higher the current workload ratio, the more energy consumption is saved by adjusting the monitoring device to be adjusted to sleep mode. , , , , , , hour, The lower the sleep index is, the less recommended it is to adjust the monitoring model to be adjusted to the sleep mode. On the contrary, the higher the sleep index is, the more recommended it is to adjust the monitoring model to be adjusted to the sleep mode.

[0055] like Figure 2As shown, it is a structural schematic diagram of a multi-source heterogeneous data acquisition system for a construction site provided by an embodiment of the present application. The embodiment of the present application provides a multi-source heterogeneous data acquisition system for a construction site, which is used to implement a multi-source heterogeneous data acquisition method for a construction site, including: a data set acquisition module, a cause influence coefficient analysis module, a monitoring equipment matching module, a sleep index analysis module, and an equipment working status control module; wherein the data set acquisition module is used to acquire historical multi-source heterogeneous data of a power grid construction site, fuse the historical multi-source heterogeneous data to form a historical event data set, and pre-process the historical event data set; the cause influence coefficient analysis module is used to extract potential causes of faults based on the pre-processed historical event data set, analyze the cause influence coefficients corresponding to the potential causes of the faults, and the cause influence coefficients corresponding to the potential causes of the faults. The number is used to measure the degree of influence of the potential cause of the fault on the occurrence of the fault; the monitoring equipment matching module is used to match the monitoring equipment according to the potential factors of the fault, and analyze the matching index between the potential factors of the fault and the monitoring equipment, and obtain the monitoring equipment to be adjusted according to the matching index, and the matching index is used to reflect the probability that the potential factors of the fault are monitored by the monitoring equipment; the sleep index analysis module is used to analyze the sleep index of the monitoring equipment to be adjusted according to the cause influence coefficient and the matching index, and the sleep index is used to reflect the probability that the monitoring equipment to be adjusted is asleep; the equipment working state control module is used to compare the sleep index with the set threshold value in the database, when the sleep index is not lower than the set threshold value, the monitoring equipment to be adjusted is controlled to sleep, and when the sleep index is lower than the set threshold value, the monitoring equipment to be adjusted is controlled to work.

[0056] In summary, the embodiment of the present application analyzes the cause influence coefficient corresponding to each potential cause of the fault, and obtains the matching index between the potential fault factor and the monitoring device, and combines the cause influence coefficient and the matching index to adjust the sleep index of the monitoring device to be adjusted, and controls the working state of the monitoring device to be monitored to sleep or work according to the sleep index and the set threshold, thereby achieving improved resource utilization efficiency and reducing the data processing amount of the monitoring device.

[0057] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0058] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0059] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0060] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0061] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0062] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for collecting multi-source heterogeneous data at a construction site, characterized in that: The following steps are involved: Acquire data set: acquire historical multi-source heterogeneous data from the power grid construction site, fuse and process the historical multi-source heterogeneous data to form a historical event data set, and pre-process the historical event data set; Analyze the cause influence coefficient: extract the potential cause of the fault based on the preprocessed historical event data set, and analyze the cause influence coefficient corresponding to the potential cause of the fault. The cause influence coefficient is used to measure the influence degree of the potential cause of the fault on the occurrence of the fault; Matching monitoring equipment: Matching monitoring equipment according to potential fault factors, analyzing the matching index between the potential fault factors and the monitoring equipment, and obtaining the monitoring equipment to be adjusted according to the matching index, which is used to reflect the probability that the potential fault factors are detected by the monitoring equipment; Analyze the waiting-to-sleep index: Analyze the waiting-to-sleep index of the monitoring device to be adjusted according to the cause influence coefficient and the matching index. The waiting-to-sleep index is used to reflect the probability of the monitoring device to be adjusted sleeping. The method for obtaining the waiting-to-sleep index is as follows: Extracting potential causes of faults whose cause influence coefficient is not higher than the first threshold in the database; Extracting matching indexes corresponding to the potential causes of the faults and the various monitoring devices based on the extracted potential causes of the faults; Extracting monitoring devices whose matching index between the potential cause of the fault and the monitoring device is higher than a second threshold in the database as monitoring devices to be adjusted; Obtain the number of personnel at the location where the monitoring equipment to be adjusted is located; Obtain the current load and maximum allowable load of the monitoring equipment to be adjusted, and the current load is divided into the maximum allowable load to obtain the current working load ratio; The weights of the cause influence coefficient, matching index, and current workload ratio for the waiting-to-sleep index are obtained through the objective weighting method; Based on the cause influence coefficient, matching index, number of personnel, and current workload ratio, the waiting-to-sleep index of the monitoring equipment to be adjusted is analyzed by the waiting-to-sleep index formula; the waiting-to-sleep index formula is: ； In the formula, is the sleep index of the kth monitoring device corresponding to the ith potential cause of the fault, e is a natural constant, k is the label number of the monitoring device, that is, the number of the monitoring device, , K is the total number of monitoring device labels, that is, the total number of monitoring devices, Is the result of judging whether the number of people is 0. If the number of people is 0, then , if the number of people is not 0, then , is the cause influence coefficient of the i-th potential cause of failure, is the matching index between the i-th potential cause of fault and the k-th monitoring device, is the current workload ratio of the i-th potential cause of the fault to the k-th monitoring device, i is the number of the potential cause of the fault, , I is the total number of potential causes of failure, for right The weight of for right The weight of for right The weight of Control the working status of the device: compare the sleep index with the set threshold in the database. When the sleep index is not lower than the set threshold, control the monitoring device to be adjusted to sleep; when the sleep index is lower than the set threshold, control the monitoring device to be adjusted to work.

2. A method for collecting multi-source heterogeneous data at a construction site as claimed in claim 1, characterized in that: The specific steps of fusing historical multi-source heterogeneous data to form a historical event data set include: Obtain historical multi-source heterogeneous data and standardize it; Time-align historical multi-source heterogeneous data according to timestamps; Spatial alignment of historical multi-source heterogeneous data according to spatial location; Extract event types from the database as event labels, and group historical multi-source heterogeneous data according to the event labels; Use a multimodal fusion algorithm to perform feature fusion on each set of historical multi-source heterogeneous data to obtain a historical event data set; All historical event data sets are grouped together into a historical event data set.

3. A method for collecting multi-source heterogeneous data at a construction site as claimed in claim 1, characterized in that: The specific steps of extracting the potential cause of the fault based on the preprocessed historical event data set include: Dataset division: Divide the preprocessed historical event dataset into training set and test set; Feature recursive splitting: Select an optimal feature from the historical event data set in the training set to perform the initial split of the historical event data set, obtain multiple subsets, and recursively split each subset until the set stop condition is met; Based on the result of recursive splitting, a preliminary decision tree model is generated, and the test set is input into the preliminary decision tree model for testing to generate a decision tree model; Based on the generated decision tree model, the potential causes of the fault are extracted.

4. A method for collecting multi-source heterogeneous data at a construction site as claimed in claim 1, characterized in that: The specific method for obtaining the cause influence coefficient is: Acquire associated feature data of potential causes of each fault from the preprocessed historical event data set, wherein the associated feature data includes fault occurrence frequency, fault duration, temperature, and humidity; Through the objective weighting method, the weight of each data in the associated feature data for the cause influence coefficient is obtained; According to the associated characteristic data and the weight of each data in the associated characteristic data for the cause influence coefficient, the cause influence coefficient of each potential cause of the fault is calculated by the cause influence coefficient formula; The cause influence coefficient formula is: ； In the formula, is the cause influence coefficient of the i-th potential cause of the fault, e is a natural constant, i is the number of the potential cause of the fault, , I is the total number of potential causes of failure, is the failure frequency of the i-th potential cause of failure, for right The weight of is the fault duration of the i-th potential cause of the fault, for right The weight of is the temperature of the ith potential cause of failure, for right The weight of is the humidity of the potential cause of the ith fault, for right The weight of .

5. A method for collecting multi-source heterogeneous data at a construction site as claimed in claim 1, characterized in that: The specific meaning of matching monitoring equipment according to potential failure factors is: Obtain the monitoring device labels corresponding to all associated characteristic data in the potential causes of the fault; Based on the acquired monitoring device labels, the corresponding monitoring devices are matched and searched from the database.

6. A method for collecting multi-source heterogeneous data at a construction site as claimed in claim 5, characterized in that: The specific method of obtaining the matching index is as follows: Extracting a monitoring device label contained in each potential cause of a fault, wherein the monitoring device label corresponds to the monitoring device in a one-to-one manner; Extract all data type labels that can be monitored by each monitoring device found to be matched from the parameter library of the monitoring device; The matching index formula is used to analyze the matching index between the monitoring device label contained in each potential cause of the fault and all data type labels that can be monitored by the monitoring device.

7. A method for collecting multi-source heterogeneous data at a construction site as claimed in claim 6, characterized in that: The matching index formula is: ； In the formula, is the matching index between the i-th potential cause of the fault and the k-th monitoring device, i is the number of the potential cause of the fault, k is the label number of the monitoring device, that is, the number of the monitoring device, , K is the total number of monitoring device labels, that is, the total number of monitoring devices, The result of whether the g-th data type label in the k-th monitoring device corresponding to the i-th potential cause of the fault can be found in the associated feature data. If it can be detected, then If it cannot be detected, , g is the data type label, , G is the total number of data type labels, The total number of all data type labels that the k-th monitoring device can monitor, is the penalty factor.

8. A multi-source heterogeneous data acquisition system for a construction site, used to implement the multi-source heterogeneous data acquisition method for a construction site according to any one of claims 1 to 7, characterized in that: include: Data set acquisition module, cause influence coefficient analysis module, monitoring equipment matching module, waiting sleep index analysis module, equipment working status control module; The data set acquisition module is used to acquire historical multi-source heterogeneous data from the power grid construction site, fuse the historical multi-source heterogeneous data to form a historical event data set, and pre-process the historical event data set; The cause influence coefficient analysis module is used to extract the potential cause of the fault based on the preprocessed historical event data set, and analyze the cause influence coefficient corresponding to the potential cause of the fault, wherein the cause influence coefficient is used to measure the influence degree of the potential cause of the fault on the occurrence of the fault; The monitoring device matching module is used to match the monitoring device according to the potential fault factor, and analyze the matching index between the potential fault factor and the monitoring device, and obtain the monitoring device to be adjusted according to the matching index, wherein the matching index is used to reflect the probability that the potential fault factor is monitored by the monitoring device; The waiting-to-sleep index analysis module is used to analyze the waiting-to-sleep index of the monitoring device to be adjusted according to the cause influence coefficient and the matching index. The waiting-to-sleep index is used to reflect the probability of the monitoring device to be adjusted sleeping. The waiting-to-sleep index is obtained in the following manner: Extracting potential causes of faults whose cause influence coefficient is not higher than the first threshold in the database; Extracting matching indexes corresponding to the potential causes of the faults and the various monitoring devices based on the extracted potential causes of the faults; Extracting monitoring devices whose matching index between the potential cause of the fault and the monitoring device is higher than a second threshold in the database as monitoring devices to be adjusted; Obtain the number of personnel at the location where the monitoring equipment to be adjusted is located; Obtain the current load and maximum allowable load of the monitoring equipment to be adjusted, and the current load is divided into the maximum allowable load to obtain the current working load ratio; The weights of the cause influence coefficient, matching index, and current workload ratio for the waiting-to-sleep index are obtained through the objective weighting method; Based on the cause influence coefficient, matching index, number of personnel, and current workload ratio, the sleep index of the monitoring equipment to be adjusted is analyzed through the sleep index formula; The sleep waiting index formula is: ； In the formula, is the sleep index of the kth monitoring device corresponding to the ith potential cause of the fault, e is a natural constant, k is the label number of the monitoring device, that is, the number of the monitoring device, , K is the total number of monitoring device labels, that is, the total number of monitoring devices, Is the result of judging whether the number of people is 0. If the number of people is 0, then , if the number of people is not 0, then , is the cause influence coefficient of the i-th potential cause of failure, is the matching index between the i-th potential cause of fault and the k-th monitoring device, is the current workload ratio of the i-th potential cause of the fault to the k-th monitoring device, i is the number of the potential cause of the fault, , I is the total number of potential causes of failure, for right The weight of for right The weight of for right The weight of The device working status control module is used to compare the sleep index with the set threshold in the database. When the sleep index is not lower than the set threshold, the monitoring device to be adjusted is controlled to sleep; when the sleep index is lower than the set threshold, the monitoring device to be adjusted is controlled to work.

Images