Settlement monitoring method and system for power tower

Abstract

The invention discloses a settlement monitoring method and system for a power tower, and relates to the related technical field of settlement monitoring, and the method comprises the steps: obtaining a monitoring data set, executing sensor data anomaly analysis on the monitoring data set, and obtaining an anomaly analysis result. And acquiring an environmental influence factor corresponding to the sensor category, and extracting an environmental parameter of the time corresponding to the abnormal analysis result. Based on the environment parameters and the anomaly analysis result, parameter correction is conducted through a parameter anomaly correction model, and a parameter correction result is obtained. And performing abnormal parameter correction on the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set. And carrying out settlement abnormity identification on the corrected monitoring data set through the settlement trend model to obtain a settlement abnormity identification result, and carrying out settlement abnormity early warning based on the settlement abnormity identification result. The technical problem that in the prior art, data of an electric power tower settlement monitoring sensor is prone to being affected by external environment factors, and consequently tower monitoring is not accurate is solved.

Description

Technical Field

[0001] The present application relates to the field related to settlement monitoring technology, and specifically to a settlement monitoring method and system for power towers. Background Art

[0002] As the main bearing structure of high-altitude transmission lines, power towers are exposed to complex natural environments for a long time and may be affected by various external factors such as wind speed, rainfall, and foundation settlement, causing the towers to settle or tilt, thereby endangering the safe operation of the power grid. In order to effectively monitor the settlement state of power towers, multiple types of sensors are usually deployed on the towers to collect data such as the inclination and strain of the towers in real time, and to manually identify abnormal situations. However, due to the complexity and variability of the external environment, such as climate change and ground vibrations, the sensor data is often interfered with, resulting in abnormal values ​​in the data, which in turn affects the accuracy of settlement monitoring.

[0003] Therefore, in the prior art, the sensor data of power tower settlement monitoring is easily affected by external environmental factors, resulting in inaccurate technical problems in power tower settlement monitoring. Summary of the invention

[0004] This application solves the technical problem in the prior art that the data of the power tower settlement monitoring sensor is easily affected by external environmental factors, resulting in inaccurate tower monitoring, by providing a settlement monitoring method and system for power towers. By analyzing and correcting the abnormalities of the sensor data and predicting the settlement trend, the interference of environmental factors on the sensor data is eliminated, and a more accurate corrected monitoring data set is obtained, ensuring accurate monitoring of the settlement of the power tower and timely identification of abnormal settlement, thus achieving accurate monitoring and early warning of the settlement of the power tower.

[0005] The present application provides a settlement monitoring method for a power tower, the method comprising: collecting monitoring information transmitted by a settlement monitoring sensor group on a power tower, preprocessing the monitoring information and completing time correspondence, and obtaining a monitoring data set. Performing sensor data anomaly analysis on the monitoring data set to obtain an anomaly analysis result. Extracting the sensor category corresponding to the anomaly analysis result, and obtaining the environmental impact factor corresponding to the sensor category. Extracting the environmental parameters corresponding to the time of the anomaly analysis result based on the environmental impact factor. Performing parameter correction based on the environmental parameters and the anomaly analysis result through a parameter anomaly correction model to obtain a parameter correction result. Performing abnormal parameter correction on the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set. Performing settlement anomaly identification of the corrected monitoring data set through a settlement trend model to obtain a settlement anomaly identification result, and performing a settlement anomaly warning based on the settlement anomaly identification result.

[0006] In the implementation method, the monitoring information transmitted by the settlement monitoring sensor group on the power tower is collected, the monitoring information is preprocessed and time correspondence is completed, and the monitoring data set is obtained, including: obtaining the monitoring information transmitted by the settlement monitoring sensor group, the monitoring information includes the monitoring data collected by each sensor according to a preset time interval, and the settlement monitoring sensor group includes multiple sensor categories. The monitoring information is classified and processed according to the sensor category to obtain the monitoring data group of each sensor. The monitoring data group of each sensor is matched with the data time between the groups, and the monitoring data set is obtained based on the corresponding results.

[0007] In the implementation, the sensor data anomaly analysis is performed on the monitoring data set to obtain the anomaly analysis result, including: obtaining the monitoring data set and extracting the monitoring data group of multiple sensors. Performing data fitting on the monitoring data group of the multiple sensors to obtain multiple fitting curves. Obtaining the deviation ratio of the data in the monitoring data group of the multiple sensors based on the multiple fitting curves. Determining whether the deviation ratio of the data is greater than or equal to a preset threshold, if it is greater than or equal to the preset threshold, marking the corresponding data as abnormal data, if it is less than the preset threshold, marking the corresponding data as normal data, and obtaining the analysis results of the monitoring data group of the multiple sensors according to the data marking. The abnormal data extraction is performed on the analysis results of the monitoring data group of the multiple sensors, the abnormal data includes an abnormal time mark, determining whether multiple abnormal data are included at the same time, and if included, canceling the abnormal mark of the abnormal data at the time, obtaining updated abnormal data, and obtaining the abnormal analysis result according to the updated abnormal data.

[0008] In an implementation, extracting the sensor category corresponding to the abnormal analysis result and obtaining the environmental impact factor corresponding to the sensor category includes: obtaining the sensor category corresponding to the abnormal data in the abnormal analysis result. Obtaining the environmental impact factor corresponding to the sensor category based on big data analysis, the environmental impact factor is an environmental parameter category that affects the sensor category.

[0009] In an implementation, performing parameter correction through a parameter abnormality correction model based on the environmental parameters and the abnormality analysis results to obtain parameter correction results includes: the parameter abnormality correction model includes a parameter classification layer and a parameter processing layer. The environmental parameters and the abnormality analysis results are input into the parameter classification layer to obtain sensor abnormality data and sensor environmental parameters corresponding to each sensor. The sensor abnormality data and the sensor environmental parameters corresponding to each sensor are input into the parameter processing layer, and parameter correction of the sensor abnormality data corresponding to each sensor is performed to obtain the parameter correction results.

[0010] In the implementation, a historical detection and correction record is obtained, wherein the historical detection and correction record includes historical sensor abnormality data, environmental parameters at corresponding moments, and historical abnormality correction parameters. Training data is obtained according to the historical detection and correction record. The training data is used to perform supervised training on a multi-layer feedforward neural network, a trained model is obtained, and a parameter processing layer is obtained.

[0011] In an implementation, performing abnormal parameter correction on the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set includes: obtaining the parameter correction result, wherein the parameter correction result includes update parameters of abnormal parameters of each sensor and corresponding time records. Based on the update parameters of the abnormal parameters of each sensor and the corresponding time records, performing data replacement on the monitoring data set to obtain a corrected monitoring data set.

[0012] The present application also provides a settlement monitoring system for a power tower, comprising:

[0013] The data acquisition module is used to collect the monitoring information transmitted by the settlement monitoring sensor group on the power tower, pre-process the monitoring information and complete the time correspondence to obtain the monitoring data set.

[0014] The anomaly analysis module is used to perform sensor data anomaly analysis on the monitoring data set to obtain anomaly analysis results.

[0015] The influencing factor acquisition module is used to extract the sensor category corresponding to the abnormal analysis result and obtain the environmental influencing factor corresponding to the sensor category.

[0016] The environmental parameter acquisition module is used to extract the environmental parameters of the time corresponding to the abnormal analysis result based on the environmental influencing factors.

[0017] The correction result acquisition module is used to perform parameter correction based on the environmental parameters and the abnormal analysis results through a parameter abnormality correction model to obtain parameter correction results.

[0018] The data correction module is used to correct abnormal parameters of the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set.

[0019] The abnormal warning module is used to identify the settlement anomaly of the modified monitoring data set through a settlement trend model, obtain the settlement anomaly identification result, and issue a settlement anomaly warning based on the settlement anomaly identification result.

[0020] The settlement monitoring method and system for power towers proposed in this application collects monitoring information transmitted by the settlement monitoring sensor group on the power tower, pre-processes the monitoring information and completes the time correspondence to obtain a monitoring data set. Perform sensor data anomaly analysis on the monitoring data set to obtain an abnormal analysis result. Extract the sensor category corresponding to the abnormal analysis result, and obtain the environmental impact factor corresponding to the sensor category. Extract the environmental parameters corresponding to the abnormal analysis result based on the environmental impact factor. Perform parameter correction based on the environmental parameters and the abnormal analysis result through the parameter anomaly correction model to obtain the parameter correction result. Perform abnormal parameter correction on the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set. Perform settlement anomaly identification of the corrected monitoring data set through the settlement trend model to obtain a settlement anomaly identification result, and perform settlement anomaly warning based on the settlement anomaly identification result. The technical problem that the settlement monitoring sensor data of the power tower in the prior art is easily affected by external environmental factors, resulting in inaccurate tower monitoring is solved. By analyzing and correcting the abnormalities of sensor data and predicting the settlement trend, the interference of environmental factors on sensor data can be eliminated, and a more accurate corrected monitoring data set can be obtained, ensuring accurate monitoring of power tower settlement and timely identification of abnormal settlement, thus achieving accurate monitoring and early warning of power tower settlement. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0022] Figure 1 A schematic flow chart of a method for monitoring the subsidence of a power tower provided in an embodiment of the present application;

[0023] Figure 2 A schematic diagram of the structure of a settlement monitoring system for a power tower provided in an embodiment of the present application.

[0024] Explanation of the reference numerals: data acquisition module 11 , abnormality analysis module 12 , influencing factor acquisition module 13 , environmental parameter acquisition module 14 , correction result acquisition module 15 , data correction module 16 , abnormality warning module 17 . DETAILED DESCRIPTION

[0025] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0026] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0027] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0028] The present application provides a method and system for monitoring the settlement of a power tower. Figure 1 As shown, the method includes:

[0029] Collect monitoring information transmitted by the settlement monitoring sensor group on the power tower, pre-process the monitoring information and complete time correspondence to obtain a monitoring data set. Perform sensor data anomaly analysis on the monitoring data set to obtain an anomaly analysis result. Extract the sensor category corresponding to the anomaly analysis result and obtain the environmental impact factor corresponding to the sensor category. Extract the environmental parameters of the time corresponding to the anomaly analysis result based on the environmental impact factor.

[0030] Multiple types of sensors are deployed on the power tower to form a settlement monitoring sensor group, and the settlement monitoring sensor group includes an inclination sensor, a displacement sensor, and a strain gauge sensor. Each sensor collects physical quantities related to tower settlement in real time. The monitoring information transmitted by the settlement monitoring sensor group on the power tower is collected, the monitoring information is preprocessed, and the time correspondence between the sensor data is completed to obtain a monitoring data set. Subsequently, sensor data anomaly analysis is performed on the monitoring data set to obtain an anomaly analysis result, which includes abnormal data points and abnormal data time corresponding to each sensor. Further, the sensor category corresponding to the anomaly analysis result is extracted, and the environmental impact factor corresponding to the sensor category is obtained. The environmental impact factor is an environmental parameter category that affects sensor data collection, such as temperature parameters, wind speed parameters, humidity parameters, etc. Further, the environmental parameters corresponding to the time of the anomaly analysis result are extracted based on the environmental impact factor.

[0031] The method provided in the embodiment of the present application also includes: obtaining monitoring information transmitted by a settlement monitoring sensor group, wherein the monitoring information includes monitoring data collected by each sensor in a preset time interval, and the settlement monitoring sensor group includes multiple sensor categories. The monitoring information is classified and processed according to the sensor category to obtain a monitoring data group of each sensor. The monitoring data groups of each sensor are matched with each other in time, and a monitoring data set is obtained based on the matching result.

[0032] Collect monitoring information transmitted by the settlement monitoring sensor group on the power tower, pre-process the monitoring information and complete time correspondence to obtain a monitoring data set, including: obtaining the monitoring data collected by each sensor at a fixed collection frequency in a preset time interval from the settlement monitoring sensor group installed on the power tower, the monitoring data includes the collection time and the data collection result of the sensor, and packaging the monitoring data to obtain the monitoring information. The preset time interval is a pre-set data collection time interval for the sensor. Subsequently, the monitoring information is classified and processed according to the sensor category to obtain a monitoring data group for each sensor. Further, the monitoring data groups of each sensor are time-correlated between groups, and the data in each monitoring data group are time-correlated, so that the data of the monitoring data group of each sensor at different time nodes are corresponding, and the monitoring data set is obtained based on the corresponding results.

[0033] The method provided in the embodiment of the present application also includes: obtaining the monitoring data set, extracting the monitoring data group of multiple sensors. Performing data fitting on the monitoring data group of the multiple sensors to obtain multiple fitting curves. Based on the multiple fitting curves, the deviation ratio of the data in the monitoring data group of the multiple sensors is obtained. It is judged whether the deviation ratio of the data is greater than or equal to a preset threshold value, if it is greater than or equal to the preset threshold value, the corresponding data is marked as abnormal data, if it is less than the preset threshold value, the corresponding data is marked as normal data, and the analysis results of the monitoring data group of the multiple sensors are obtained according to the data marking. The analysis results of the monitoring data group of the multiple sensors are subjected to the abnormal data extraction, the abnormal data includes an abnormal time mark, and it is judged whether multiple abnormal data are included at the same time. If included, the abnormal mark of the abnormal data at that time is cancelled, and the updated abnormal data is obtained, and the abnormal analysis result is obtained according to the updated abnormal data.

[0034] The sensor data anomaly analysis is performed on the monitoring data set to obtain the anomaly analysis result, including: obtaining the monitoring data set and extracting the monitoring data groups of multiple sensors. Subsequently, the monitoring data groups of the multiple sensors are fitted, the data trend of the monitoring data groups of the sensors in the time series is fitted, and multiple fitting curves are obtained, each fitting curve corresponds to the sensor category. Through the fitting curve, the data actually collected by the sensor at each time point of each sensor data group is compared with the expected value of the fitting curve, and the deviation ratio is calculated. The deviation ratio is the ratio of the absolute value of the difference between the actual collection value and the fitting value to the actual collection value. Further, it is determined whether the deviation ratio of the data is greater than or equal to a preset threshold. When the deviation ratio is greater than or equal to the threshold, the data is marked as abnormal, and when the deviation ratio is less than the threshold, it is marked as normal. Among them, the preset threshold is a preset deviation ratio. For example, if the set deviation ratio threshold is 5%, when the deviation ratio of a certain sensor data is 8%, it exceeds the threshold, so the data point will be marked as abnormal data. The analysis results of the monitoring data groups of the multiple sensors are obtained according to the data marking. The abnormal data extraction is performed on the analysis results of the monitoring data groups of the multiple sensors, the abnormal data includes an abnormal time identifier, and it is determined whether multiple abnormal data are included at the same time. If included, the abnormal mark of the abnormal data at that time is cancelled to obtain updated abnormal data, and the abnormal analysis result is obtained according to the updated abnormal data, and the abnormal data point corresponding to each sensor and the abnormal data time are included in the abnormal analysis result.

[0035] The method provided in the embodiment of the present application further includes: obtaining the sensor category corresponding to the abnormal data in the abnormal analysis result, and obtaining the environmental impact factor corresponding to the sensor category based on big data analysis, wherein the environmental impact factor is the environmental parameter category that affects the sensor category.

[0036] Extracting the sensor category corresponding to the abnormal analysis result and obtaining the environmental impact factor corresponding to the sensor category includes: extracting the sensor category corresponding to the abnormal data from the abnormal analysis result. Subsequently, based on big data analysis, historically collected data of different sensor categories are collected from multiple sources for monitoring data and environmental data related to sensors, extracting abnormal monitoring data and corresponding environmental data, and statistically analyzing the environmental data to obtain the proportion of non-standard working environment parameter categories, obtaining one or more environmental parameter categories with a high proportion, preferably three environmental parameter categories, and obtaining environmental impact factors, wherein the environmental impact factors are environmental parameter categories that have a significant impact on the sensor category.

[0037] Based on the environmental parameters and the abnormal analysis results, parameter correction is performed through a parameter anomaly correction model to obtain parameter correction results. Based on the parameter correction results, abnormal parameter correction is performed on the monitoring data set to obtain a corrected monitoring data set. Settlement anomaly identification is performed on the corrected monitoring data set through a settlement trend model to obtain a settlement anomaly identification result, and settlement anomaly warning is performed based on the settlement anomaly identification result.

[0038] Based on the abnormal analysis results and environmental parameters, the abnormal data is corrected using a pre-constructed parameter anomaly correction model. The parameter anomaly correction model takes into account the influence of the specific physical characteristics of the sensor and environmental factors on the measurement results, and can effectively correct the deviation of the sensor caused by external conditions. Based on the parameter correction results, the monitoring data set is corrected for abnormal parameters to obtain a corrected monitoring data set. Further, through the pre-established settlement trend model, the settlement trend analysis and anomaly identification are performed on the corrected monitoring data set to obtain a settlement anomaly identification result. The settlement trend model is constructed based on the long short-term memory network model, and the monitoring data of the power tower in the sensor group during the historical monitoring process is obtained. The monitoring data contains the monitoring time, and the abnormal data is manually eliminated to obtain the abnormal identification training data. Subsequently, the abnormal identification training data is input into the long short-term memory network model through multiple iterations until the model reaches the preset number of iterations, and the model after the training is completed is obtained, and the model is verified by the monitoring data until the output accuracy meets the preset accuracy. The trained model is obtained to obtain the settlement trend model. The corrected monitoring data set is input into the settlement trend model to output the sensor prediction value in the future time to obtain the settlement anomaly identification result. Finally, the settlement anomaly identification result is analyzed in the same way as the sensor data anomaly analysis to obtain the corresponding anomaly points, and the settlement anomaly warning is performed based on the anomaly points. The technical problem that the settlement monitoring sensor data of the power tower in the prior art is easily affected by external environmental factors, resulting in inaccurate tower monitoring, is solved. By analyzing the anomaly, correcting and predicting the settlement trend of the sensor data, the interference of environmental factors on the sensor data is eliminated, and a more accurate corrected monitoring data set is obtained, ensuring the accurate monitoring of the settlement of the power tower and the timely identification of abnormal settlement, thereby improving the accurate monitoring and early warning of the settlement of the power tower.

[0039] The method provided in the embodiment of the present application also includes: the parameter anomaly correction model includes a parameter classification layer and a parameter processing layer. The environmental parameters and the abnormal analysis results are input into the parameter classification layer to obtain sensor abnormality data and sensor environmental parameters corresponding to each sensor. The sensor abnormality data and the sensor environmental parameters corresponding to each sensor are input into the parameter processing layer, and parameter correction of the sensor abnormality data corresponding to each sensor is performed to obtain the parameter correction result.

[0040] The parameter anomaly correction model includes a parameter classification layer and a parameter processing layer. The environmental parameters and the abnormal analysis results are input into the parameter classification layer to obtain the sensor abnormal data and sensor environmental parameters corresponding to each sensor. The parameter classification layer is based on rule matching classification. By performing data clustering of each category of sensors on the environmental parameters corresponding to the abnormal analysis results, the data in each clustering cluster corresponds to a sensor, and the sensor abnormal data is classified and the corresponding environmental parameters are obtained. Finally, the sensor abnormal data and the sensor environmental parameters corresponding to each sensor are input into the parameter processing layer, and the parameter correction of the sensor abnormal data corresponding to each sensor is performed to obtain the parameter correction result.

[0041] The method provided in the embodiment of the present application also includes: obtaining a historical detection and correction record, wherein the historical detection and correction record includes historical sensor abnormality data, environmental parameters at corresponding moments, and historical abnormality correction parameters. According to the historical detection and correction record, training data is obtained. The training data is used to perform supervised training on a multi-layer feedforward neural network, a trained model is obtained, and a parameter processing layer is obtained.

[0042] Constructing a parameter processing layer includes: obtaining historical detection and correction records, wherein the historical detection and correction records contain historical sensor abnormal data, environmental parameters at the corresponding time of the historical abnormal data, and historical abnormal correction parameters. The historical sensor abnormal data is the abnormal data point that the sensor has occurred in the historical monitoring process. These data mark the abnormal parameter value of the sensor at a specific time point. For example, the tilt angle of the tilt sensor is abnormal, and the stress generated by the strain gauge sensor is abnormal. The environmental parameters at the corresponding time are the environmental conditions when the sensor abnormal data occurs. The historical abnormal correction parameter is the correction value used in the correction process according to the historical data. The correction value is the correction value after the sensor abnormal data is corrected to make it closer to the true value. The historical sensor abnormal data and the corresponding environmental parameters of the historical detection and correction record are used as the input vector of the neural network. The historical abnormal correction parameters of the historical detection and correction record are used as the target output of the training. The output data represents the value after the abnormal data is corrected, that is, the expected correction result, forming a large number of input-output pairs to obtain training data. The neural network is supervised and trained based on the training data. The weights are adjusted multiple times, and the difference between the neural network output and the target correction value is evaluated by the mean square error. The error between the predicted correction value and the historical true correction value is minimized to obtain the trained model and the parameter processing layer.

[0043] The method provided in the embodiment of the present application further includes: obtaining the parameter correction result, wherein the parameter correction result includes the update parameters of each sensor abnormal parameter and the corresponding time record. Based on the update parameters of each sensor abnormal parameter and the corresponding time record, the monitoring data set is replaced to obtain a corrected monitoring data set.

[0044] Obtain the parameter correction results output by the parameter anomaly correction model, which include the updated parameters of each sensor anomaly parameter and the corresponding time records. Obtain the monitoring data set, and replace the corresponding data in the monitoring data set with the corrected values ​​based on the updated parameters of each sensor anomaly parameter and the corresponding time records to obtain the corrected monitoring data set. The corrected monitoring data set obtained after the correction is completed will be used as reliable data for subsequent settlement analysis.

[0045] In the above, refer to Figure 1 The settlement monitoring method for a power tower according to an embodiment of the present invention is described in detail. Figure 2 A settlement monitoring system for a power tower according to an embodiment of the present invention is described.

[0046] The settlement monitoring system for power towers according to the embodiments of the present invention solves the technical problem in the prior art that the settlement monitoring sensor data of power towers is easily affected by external environmental factors, resulting in inaccurate tower monitoring. By analyzing and correcting the abnormalities of sensor data and predicting the settlement trend, the interference of environmental factors on sensor data is eliminated, and a more accurate corrected monitoring data set is obtained, ensuring accurate monitoring of the settlement of power towers and timely identification of abnormal settlement, thereby achieving accurate monitoring and early warning of the settlement of power towers. The settlement monitoring system for power towers includes: a data acquisition module 11, an abnormality analysis module 12, an influencing factor acquisition module 13, an environmental parameter acquisition module 14, a correction result acquisition module 15, a data correction module 16, and an abnormality early warning module 17.

[0047] The data acquisition module 11 is used to collect monitoring information transmitted by the settlement monitoring sensor group on the power tower, pre-process the monitoring information and complete time correspondence to obtain a monitoring data set.

[0048] The abnormality analysis module 12 is used to perform sensor data abnormality analysis on the monitoring data set to obtain abnormality analysis results.

[0049] The impact factor acquisition module 13 is used to extract the sensor category corresponding to the abnormal analysis result and obtain the environmental impact factor corresponding to the sensor category.

[0050] The environmental parameter acquisition module 14 is used to extract the environmental parameters corresponding to the time of the abnormal analysis result based on the environmental impact factors.

[0051] The correction result acquisition module 15 is used to perform parameter correction based on the environmental parameters and the abnormality analysis results through a parameter abnormality correction model to obtain parameter correction results.

[0052] The data correction module 16 is used to correct abnormal parameters of the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set.

[0053] The abnormal warning module 17 is used to identify the settlement anomaly of the modified monitoring data set through a settlement trend model, obtain the settlement anomaly identification result, and issue a settlement anomaly warning based on the settlement anomaly identification result.

[0054] The specific configuration of the data acquisition module 11 will be described in detail below. The data acquisition module 11 may further include: collecting monitoring information transmitted by the settlement monitoring sensor group on the power tower, preprocessing the monitoring information and completing time correspondence, and obtaining a monitoring data set, including: obtaining monitoring information transmitted by the settlement monitoring sensor group, the monitoring information includes monitoring data collected by each sensor according to a preset time interval, and the settlement monitoring sensor group includes multiple sensor categories. The monitoring information is classified and processed according to the sensor category to obtain the monitoring data group of each sensor. The monitoring data group of each sensor is subjected to inter-group data time correspondence, and the monitoring data set is obtained based on the correspondence result.

[0055] The specific configuration of the abnormal analysis module 12 will be described in detail below. The abnormal analysis module 12 further includes: performing sensor data abnormal analysis on the monitoring data set, obtaining abnormal analysis results, including: obtaining the monitoring data set, extracting monitoring data groups of multiple sensors. Performing data fitting on the monitoring data groups of the multiple sensors, obtaining multiple fitting curves. Obtaining the deviation ratio of the data in the monitoring data groups of the multiple sensors based on the multiple fitting curves. Determining whether the deviation ratio of the data is greater than or equal to a preset threshold, if it is greater than or equal to the preset threshold, marking the corresponding data as abnormal data, if it is less than the preset threshold, marking the corresponding data as normal data, and obtaining the analysis results of the monitoring data groups of the multiple sensors according to the data marking. The abnormal data extraction is performed on the analysis results of the monitoring data groups of the multiple sensors, the abnormal data includes an abnormal time mark, determining whether multiple abnormal data are included at the same time, and if included, canceling the abnormal mark of the abnormal data at that time, obtaining updated abnormal data, and obtaining the abnormal analysis results according to the updated abnormal data.

[0056] The specific configuration of the influencing factor acquisition module 13 will be described in detail below. The influencing factor acquisition module 13 may further include: extracting the sensor category corresponding to the abnormal analysis result, and acquiring the environmental influencing factor corresponding to the sensor category, including: acquiring the sensor category corresponding to the abnormal data in the abnormal analysis result. The environmental influencing factor corresponding to the sensor category is acquired based on big data analysis, and the environmental influencing factor is the category of environmental parameters that affect the sensor category.

[0057] The specific configuration of the correction result acquisition module 15 will be described in detail below. The correction result acquisition module 15 further includes: performing parameter correction through a parameter abnormality correction model based on the environmental parameters and the abnormality analysis results to obtain parameter correction results, including: the parameter abnormality correction model includes a parameter classification layer and a parameter processing layer. The environmental parameters and the abnormality analysis results are input into the parameter classification layer to obtain sensor abnormality data and sensor environmental parameters corresponding to each sensor. The sensor abnormality data and the sensor environmental parameters corresponding to each sensor are input into the parameter processing layer, and parameter correction of the sensor abnormality data corresponding to each sensor is performed to obtain the parameter correction results.

[0058] The specific configuration of the correction result acquisition module 15 will be described in detail below. The correction result acquisition module 15 further includes: obtaining historical detection correction records, wherein the historical detection correction records contain historical sensor abnormality data, environmental parameters at corresponding moments, and historical abnormality correction parameters. According to the historical detection correction records, training data is obtained. The training data is used to perform supervised training on a multi-layer feedforward neural network, and a trained model is obtained to obtain a parameter processing layer.

[0059] The specific configuration of the data correction module 16 will be described in detail below. The data correction module 16 further includes: performing abnormal parameter correction on the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set, including: obtaining the parameter correction result, the parameter correction result includes the update parameters of the abnormal parameters of each sensor and the corresponding time record. Based on the update parameters of the abnormal parameters of each sensor and the corresponding time record, the monitoring data set is replaced with data to obtain a corrected monitoring data set.

[0060] The settlement monitoring system for a power tower provided in an embodiment of the present invention can execute the settlement monitoring method for a power tower provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0061] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or the server, and the various units and modules included are only divided according to functional logic, but are not limited to the above divisions, as long as the corresponding functions can be achieved. In addition, the specific names of the functional units are only for the convenience of distinguishing from each other, and are not used to limit the scope of protection of the present invention.

[0062] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for monitoring the settlement of a power tower, characterized in that: The method comprises: Collect monitoring information transmitted by a settlement monitoring sensor group on a power tower, pre-process the monitoring information and complete time correspondence to obtain a monitoring data set; Performing sensor data anomaly analysis on the monitoring data set to obtain an anomaly analysis result; Extracting the sensor category corresponding to the abnormal analysis result, and obtaining the environmental impact factor corresponding to the sensor category; Extracting the environmental parameters of the time corresponding to the abnormal analysis result based on the environmental impact factor; Based on the environmental parameters and the abnormal analysis results, parameter correction is performed through a parameter abnormality correction model to obtain a parameter correction result; Based on the parameter correction result, the abnormal parameters of the monitoring data set are corrected to obtain a corrected monitoring data set; The settlement anomaly of the modified monitoring data set is identified by using a settlement trend model, a settlement anomaly identification result is obtained, and a settlement anomaly early warning is performed based on the settlement anomaly identification result.

2. The method for monitoring the settlement of a power tower according to claim 1, characterized in that: Collect monitoring information transmitted by the settlement monitoring sensor group on the power tower, pre-process the monitoring information and complete time correspondence to obtain a monitoring data set, including: Acquire monitoring information transmitted by a settlement monitoring sensor group, wherein the monitoring information includes monitoring data collected by each sensor according to a preset time interval, and the settlement monitoring sensor group includes multiple sensor categories; Classify the monitoring information according to sensor categories to obtain monitoring data groups of each sensor; The monitoring data groups of the sensors are matched with each other in terms of time, and a monitoring data set is obtained based on the matching results.

3. The settlement monitoring method for a power tower according to claim 1, characterized in that: Performing sensor data anomaly analysis on the monitoring data set to obtain an anomaly analysis result includes: Acquire the monitoring data set, extract monitoring data groups of multiple sensors; Performing data fitting on the monitoring data groups of the multiple sensors to obtain multiple fitting curves; Obtaining deviation ratios of data in the monitoring data groups of the multiple sensors based on the multiple fitting curves; Determine whether the deviation ratio of the data is greater than or equal to a preset threshold, if it is greater than or equal to the preset threshold, mark the corresponding data as abnormal data, if it is less than the preset threshold, mark the corresponding data as normal data, and obtain the analysis results of the monitoring data group of the multiple sensors according to the data marking; The abnormal data extraction is performed on the analysis results of the monitoring data groups of the multiple sensors, the abnormal data includes an abnormal time mark, and it is determined whether multiple abnormal data are included at the same time. If included, the abnormal mark of the abnormal data at that time is cancelled to obtain updated abnormal data, and the abnormal analysis result is obtained according to the updated abnormal data.

4. The method for monitoring the settlement of a power tower according to claim 1, characterized in that: Extracting the sensor category corresponding to the abnormal analysis result and obtaining the environmental impact factor corresponding to the sensor category includes: Obtaining the sensor category corresponding to the abnormal data in the abnormal analysis result; The environmental impact factor corresponding to the sensor category is obtained based on big data analysis, where the environmental impact factor is an environmental parameter category that affects the sensor category.

5. The method for monitoring the settlement of a power tower according to claim 1, characterized in that: Based on the environmental parameters and the abnormal analysis results, parameter correction is performed through a parameter abnormality correction model to obtain parameter correction results, including: The parameter anomaly correction model includes a parameter classification layer and a parameter processing layer; Inputting the environmental parameters and the abnormal analysis results into the parameter classification layer to obtain sensor abnormality data and sensor environmental parameters corresponding to each sensor; The sensor abnormality data and the sensor environmental parameters corresponding to each sensor are input into a parameter processing layer, and parameter correction of the sensor abnormality data corresponding to each sensor is performed to obtain the parameter correction result.

6. The method for monitoring the settlement of a power tower according to claim 5, characterized in that: include: Obtaining historical detection and correction records, wherein the historical detection and correction records include historical sensor abnormality data, environmental parameters at corresponding times, and historical abnormality correction parameters; Acquiring training data according to the historical detection and correction records; The training data is used to perform supervised training on a multi-layer feedforward neural network, a trained model is obtained, and a parameter processing layer is obtained.

7. The method for monitoring the settlement of a power tower according to claim 1, characterized in that: Based on the parameter correction result, the abnormal parameters of the monitoring data set are corrected to obtain a corrected monitoring data set, including: Obtaining the parameter correction result, wherein the parameter correction result includes update parameters of abnormal parameters of each sensor and corresponding time records; Based on the update parameters of the abnormal parameters of each sensor and the corresponding time records, data replacement is performed on the monitoring data set to obtain a corrected monitoring data set.

8. A settlement monitoring system for a power tower, characterized in that: The system comprises: A data acquisition module is used to collect monitoring information transmitted by a settlement monitoring sensor group on a power tower, pre-process the monitoring information and complete time correspondence to obtain a monitoring data set; An anomaly analysis module, used to perform sensor data anomaly analysis on the monitoring data set and obtain an anomaly analysis result; An impact factor acquisition module, used to extract the sensor category corresponding to the abnormal analysis result, and obtain the environmental impact factor corresponding to the sensor category; An environmental parameter acquisition module, used to extract the environmental parameters corresponding to the time of the abnormal analysis result based on the environmental influencing factors; A correction result acquisition module, used to perform parameter correction through a parameter anomaly correction model based on the environmental parameters and the anomaly analysis results, and obtain parameter correction results; A data correction module, used for correcting abnormal parameters of the monitoring data set based on the parameter correction result to obtain a corrected monitoring data set; The abnormal warning module is used to identify the settlement anomaly of the modified monitoring data set through a settlement trend model, obtain the settlement anomaly identification result, and issue a settlement anomaly warning based on the settlement anomaly identification result.

Images