A remote communication switching method for a collector line ice melting intelligent switch

Abstract

The present invention relates to the technical field of snow melting on electric wires, and in particular to a remote communication switching method for a collector line ice melting intelligent switch. The method divides different monitoring data sequences into subsequences; analyzes the degree of difference between each subsequence and other subsequences in the stability of temporal changes, obtains the response characteristics of each subsequence, and obtains the response characteristic sequence of each monitoring data sequence based on the distribution of the response characteristics; obtains the effective response situation through the distribution deviation of the response characteristic sequence between the monitoring data sequence and other monitoring data sequences; adjusts the prediction and judgment of the monitoring data through the effective response situation, obtains the communication judgment value for remote communication switching. The present invention takes into account the response state differences of different monitoring data, integrates the analysis and adjusts the judgment process of the switching state of the collector line ice melting intelligent switch, so that the remote communication switching of the collector line ice melting intelligent switch is more efficient and the remote switching is more reliable.

Description

Technical Field

[0001] The invention relates to the technical field of electric wire snow melting, and in particular to a remote communication switching method for an intelligent switch for melting ice on a collector line. Background Art

[0002] With the promotion of Internet of Things technology and automation technology in the power transmission industry, the power grid department has been able to timely discover and grasp various conditions of the transmission lines through online monitoring systems, and has realized the use of ice-melting devices on the collector lines to melt the ice on the lines; among them, wind farms are mostly located in high-altitude micro-topography areas. In winter, wind turbines, transmission lines, overhead collector lines and other equipment are prone to serious icing, which eventually leads to large-scale grid disconnection. The empty collector lines of wind farms have low voltage levels and small conductor cross-sections, and AC ice melting can be used.

[0003] In the process of using the ice melting device to melt ice, it is necessary to disconnect the connection between the low-voltage side of the box transformer and the AC device, and perform a three-phase short circuit at the end of the line to achieve AC ice melting of the collector line. Nowadays, the control of the line ice melting device has been realized by remote control to avoid the obstruction of the ice melting operation due to the complex environment of the collector line. However, the remote control of the integrated circuit ice melting device is affected by the changes in the environment of the collector line. The effectiveness of the monitoring data of the environment where the device is located in the response to the switching of the collector line ice melting intelligent switch will directly affect the efficient control of the ice melting intelligent switch using remote communication. The traditional method of switching and analyzing the monitoring data has poor processing effect when dealing with collector lines with frequent and complex environmental monitoring data, which affects the accuracy of the remote switching analysis of the monitoring data on the ice melting intelligent switch, resulting in generally low efficiency of remote communication switching of the collector line ice melting intelligent switch. Summary of the invention

[0004] In order to solve the technical problem of low efficiency of remote communication switching of the collector line ice melting intelligent switch in the prior art, the object of the present invention is to provide a remote communication switching method for the collector line ice melting intelligent switch, and the technical solution adopted is as follows:

[0005] According to one aspect of an embodiment of the present invention, a remote communication switching method for a collector line ice melting intelligent switch is provided, comprising:

[0006] Obtain different monitoring data sequences of ice melting of collector lines; divide each monitoring data sequence into subsequences;

[0007] In each monitoring data sequence, the response characteristics of each subsequence are obtained according to the degree of stability difference between each subsequence and other subsequences in time series; based on the distribution of the response characteristics of the subsequences in the monitoring data sequence, the response characteristic sequence of each monitoring data sequence is obtained;

[0008] According to the degree of deviation between each monitoring data sequence and other monitoring data sequences in the distribution of response characteristic sequences, the effective response of each monitoring data sequence is obtained;

[0009] The prediction judgment of the monitoring data is adjusted by the effective response of the monitoring data sequence to obtain a communication judgment value; and the remote communication switching of the switch is performed according to the communication judgment value.

[0010] In the above scheme, considering that the environment is complex and causes different monitoring data to change frequently, this will make different monitoring data have different response effects on the switch switching state. Therefore, for different monitoring data, subsequent analysis is performed by dividing subsequences, and the stability of each subsequence and other subsequences in time series is further analyzed. The response characteristics of each subsequence are obtained through the degree of change stability difference, which reflects the influence of the monitoring data on the response in the subsequence time series. The response feature sequence of each monitoring data sequence is obtained based on the distribution of the response feature, so as to reflect the influence of the change of data in each monitoring data sequence with time series through the response feature sequence, thereby analyzing the overall influence of the monitoring data sequence. Through the distribution deviation of the response feature sequence between the monitoring data sequence and other monitoring data sequences, the effective response of each monitoring data sequence is obtained, and the more significant the effective response is, the more accurate it is. Therefore, the prediction judgment of the monitoring data is further adjusted through the effective response situation to obtain the communication judgment value, and the necessary degree of the switching judgment is analyzed more accurately, so as to perform remote communication switching based on the communication judgment value. The present invention considers the difference in response states of different monitoring data, integrates analysis and adjusts the judgment on the switching state of the collector line ice-melting intelligent switch, so that the remote communication switching of the collector line ice-melting intelligent switch is more efficient and the remote switching is more reliable.

[0011] Furthermore, the method for acquiring the response characteristics includes:

[0012] For each subsequence, the characteristic vector of each moment is obtained through the time series decomposition algorithm; the characteristic vector is composed of the trend component, period component and residual component corresponding to the moment;

[0013] In each subsequence, the relative stability of each moment is obtained according to the similarity of the feature vector between each moment and other moments;

[0014] Fit the relative stability curves of all moments in each subsequence to obtain the stability characteristic curve of each subsequence;

[0015] According to the difference between each subsequence and other subsequences in the monitoring data sequence on the stable characteristic curve, the response characteristics of each subsequence are obtained.

[0016] In the above scheme, the values ​​of the three components at each moment are obtained through time series decomposition of the subsequences to form a feature vector, thereby characterizing the changing trend characteristics of the data at each moment, and then the degree of difference in the feature vectors between the moments in the subsequences is used to reflect the stability of each moment. Combined with the difference in relative stability between the subsequences, the response characteristics of each subsequence are obtained, thereby reflecting the significance of each subsequence in the change of the monitored data. The more significant the subsequence performance is, the greater the impact on the response situation.

[0017] Furthermore, the relative stability of each moment is obtained according to the similarity between the feature vectors of each moment and other moments, including:

[0018] In each subsequence, the cosine similarity of the feature vector between each moment and every other moment is taken as the proximity between each moment and every other moment;

[0019] The average of the similarities between each moment and all other moments is taken as the relative stability of each moment.

[0020] In the above scheme, each moment is combined with the degree of approximation of the feature vector of each other moment, and then comprehensively represents the relative consistency of the change trend of each moment in the subsequence. The higher the relative stability of the moment, the less significant the monitoring change frequency of the moment in the subsequence, which facilitates the subsequent analysis of the entire subsequence.

[0021] Furthermore, obtaining the response characteristics of each subsequence according to the difference between each subsequence and other subsequences in the monitoring data sequence on the stable characteristic curve includes:

[0022] In each monitoring data sequence, according to the difference between the stable characteristic curves of each subsequence and each other subsequence, the difference characteristics of each subsequence and each other subsequence are obtained;

[0023] The mean of the difference features between each subsequence and all other subsequences is taken as the response feature of each subsequence.

[0024] In the above scheme, in the monitoring data sequence, each subsequence is analyzed separately for differences in stable characteristics with other subsequences, and then the response characteristics of the subsequence are obtained by combining them, reflecting the significant differences between the subsequence and other subsequences in the monitoring data sequence. The higher the overall difference, the greater the influence of the subsequence on the response change, so as to facilitate the subsequent analysis of the entire monitoring data sequence.

[0025] Furthermore, the method for acquiring the response characteristic sequence includes:

[0026] In all monitoring data sequences, the total number of subsequences with the largest number of subsequences is taken as the number of sequences; the total amount of data in the response feature sequence is set as the number of sequences;

[0027] For any monitoring data sequence, the response features in the monitoring data sequence are placed into the response feature sequence in sequence according to the time sequence of the subsequence. If the response features in the monitoring data sequence cannot fill the response feature sequence, the remaining vacancies in the response feature sequence are filled with zeros to obtain the response feature sequence of the monitoring data sequence.

[0028] In the above scheme, taking into account the complex characteristics of changes in different monitoring data, different monitoring data sequences may not be able to guarantee consistent division results when dividing subsequences for analysis. Therefore, the case with the most divided subsequences is used as a reference for constructing the response feature sequence of all monitoring data sequences, ensuring that the length of the response feature sequence can meet all division situations and that the length of the response feature sequence among all monitoring data sequences is consistent, so as to facilitate the subsequent different situations of different monitoring data between the changing time intervals.

[0029] Furthermore, the method for obtaining the effective response situation includes:

[0030] According to the difference between each monitoring data sequence and each other monitoring data sequence in the response feature sequence, the feature difference significance between each monitoring data sequence and each other monitoring data sequence is obtained;

[0031] The mean of the significance of the characteristic differences between each monitoring data series and all other monitoring data series is taken as the effective response of each monitoring data series.

[0032] In the above scheme, since the response feature sequence is set to correspond to the largest number of subsequences, the differences of the response feature sequences are obtained respectively, and the difference analysis is performed not only on the numerical value of the response feature, but also on the division of the subsequences. This avoids the error caused by the fact that when the subsequence response feature values ​​are similar but the gap between the divided subsequence intervals is large when the data distribution characteristics in the monitoring data sequence are divided, which is conducive to more accurate subsequent judgments on multiple monitoring data sequences.

[0033] Furthermore, the method of adjusting the prediction judgment of the monitoring data by the effective response of the monitoring data sequence to obtain the communication judgment value includes:

[0034] For any monitoring data sequence, normalize the effective response of the monitoring data sequence to obtain the judgment confidence of the monitoring data sequence;

[0035] Obtaining a predicted value of the monitoring data sequence; using the difference between the predicted value of the monitoring data sequence and the corresponding preset monitoring threshold as a judgment feature value of the monitoring data sequence;

[0036] The product of the judgment characteristic value and the judgment confidence of the monitoring data sequence is used as the judgment coefficient of the monitoring data sequence; and the sum of the judgment coefficients of all monitoring data sequences is normalized and used as the communication judgment value.

[0037] In the above scheme, when judging the impact of icing through prediction, the response impact of each monitoring data is taken into account and the judgment value is adjusted. The higher the effective response, the higher the impact on the corresponding monitoring data in the response judgment process. Therefore, when all monitoring data are integrated for switching judgment, the adjustment degree is higher. Finally, the communication judgment value can more accurately reflect the difference between the monitoring situation and the threshold for subsequent control judgment.

[0038] Furthermore, the remote communication switching of the switch according to the communication judgment value includes:

[0039] When the communication judgment value is greater than the preset judgment threshold, a remote switching control signal is sent to the collector line ice-melting intelligent switch; otherwise, no remote switching control signal is sent.

[0040] Further, the obtaining of the predicted value of the monitoring data sequence includes:

[0041] The EMA algorithm is used on the monitoring data sequence to obtain the monitoring data at the next moment as the predicted value.

[0042] Furthermore, each monitoring data sequence adopts a sequence segmentation algorithm to obtain a subsequence.

[0043] According to another aspect of an embodiment of the present invention, a remote communication switching system for an intelligent switch for melting ice on a collector line is provided.

[0044] A remote communication switching system for an intelligent switch for melting ice on a collector line, comprising:

[0045] A data acquisition module is used to acquire different monitoring data sequences of ice melting of collector lines; and divide each monitoring data sequence into subsequences;

[0046] A response analysis module is used to obtain the response characteristics of each subsequence in each monitoring data sequence according to the degree of stability difference between each subsequence and other subsequences in time series; based on the distribution of the response characteristics of the subsequences in the monitoring data sequence, obtain the response characteristic sequence of each monitoring data sequence;

[0047] An effective response analysis module is used to obtain the effective response of each monitoring data sequence according to the degree of deviation between each monitoring data sequence and other monitoring data sequences in the distribution of response characteristic sequences;

[0048] The remote communication switching module is used to adjust the prediction judgment of the monitoring data by monitoring the effective response of the data sequence to obtain the communication judgment value; and perform remote communication switching of the switch according to the communication judgment value.

[0049] According to yet another aspect of an embodiment of the present invention, a computer device is provided.

[0050] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of a remote communication switching method for a collector line ice-melting intelligent switch as described above are implemented.

[0051] According to yet another aspect of an embodiment of the present invention, a computer-readable storage medium is provided.

[0052] A computer-readable storage medium stores a computer program, which can be executed by at least one processor. When the computer program code runs on the processor of a computer, the computer executes the steps of a remote communication switching method for a collector line ice-melting intelligent switch as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 A flow chart of a remote communication switching method for a collector line ice melting intelligent switch provided by one embodiment of the present invention;

[0055] Figure 2 A flow chart of a method for obtaining response characteristics provided by an embodiment of the present invention;

[0056] Figure 3 A flow chart of a method for obtaining a communication judgment value provided by an embodiment of the present invention;

[0057] Figure 4 A structural diagram of a remote communication switching system for a collector line ice-melting intelligent switch provided by one embodiment of the present invention;

[0058] Figure 5 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a remote communication switching method for a collector line ice melting intelligent switch proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0060] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0061] The specific scheme of the remote communication switching method for the collector line ice-melting intelligent switch provided by the present invention is described in detail below with reference to the accompanying drawings.

[0062] Embodiment 1:

[0063] The embodiment of the present invention provides a remote communication switching method for a collector line ice melting intelligent switch, see Figure 1 , which shows a flow chart of a remote communication switching method for a collector line ice melting intelligent switch provided by an embodiment of the present invention, the method comprising the following steps:

[0064] S1: Obtain different monitoring data sequences of ice melting of collector lines; divide each monitoring data sequence into subsequences.

[0065] The switch of the collector line ice melting device is a 35kV collector line ice melting intelligent switch. The collector line ice melting intelligent switch is deployed at the end of the collector line. The three-phase short circuit of the end line is quickly realized through remote control of the collector line ice melting intelligent switch to ensure efficient line ice melting. Among them, the collector line ice melting switch includes a primary part and a secondary part. The primary part includes a switch box and a 35kV circuit breaker; the secondary part includes a control terminal design, a switch lightning protection design, a filter design, a switch protection design, and a switch remote communication design. The switch lightning protection design and filter design are equipped with a 35kV large current arrester on the primary side, and a secondary lightning protection module and a two-layer EMI filter circuit are designed in the control terminal. EMI (Electromagnetic Interference) represents electromagnetic interference, which can discharge the lightning current and overvoltage conducted by the collector line, and can also protect the back-end circuit from overvoltage. The control terminal in the secondary part of the collector line ice melting switch uses a high-performance single-chip microcomputer as the core to ensure the normal communication and functional control of multiple functional modules of the ice melting controller. The operating temperature of the main control chip is -40℃ to +85℃, which can ensure the normal operation of the ice-melting short-circuit switch control terminal in harsh environments.

[0066] Each ice-melting short-circuit switch is controlled by a local control terminal as the main control to perform local control and protection of the switch, as well as control and parameter detection of peripheral equipment. In addition, each ice-melting short-circuit switch is designed around the control terminal with a voltage transformer, lightning protection system, backup power supply and management system, circuit breaker mechanism, current transformer, environment and cabinet temperature detection and heating system, etc. The control system is deployed in the host computer.

[0067] The switching of the collector line ice-melting intelligent switch is mainly through remote communication. The design of the switch remote communication adopts two methods: 4G network communication and optical fiber communication. In extremely harsh conditions, Beidou short messages can also be used to achieve remote control and status monitoring, which can realize background monitoring of switch status, battery power, and line power status, and realize remote control of switches. The start and stop of the ice-melting device is judged by the collector line monitoring data obtained by the ice-melting device in the current environment, that is, the switching of the collector line ice-melting intelligent switch is remotely controlled. Therefore, it is necessary to judge and analyze the ice coverage of the collector line through monitoring data, and then perform remote communication switching.

[0068] In a specific implementation method of an embodiment of the present invention, monitoring data including temperature data, sag data, tension data, wind speed data and humidity data can be obtained through the temperature sensor, inclination sensor, stress sensor, speed sensor and humidity sensor in the monitoring system of the collector line ice melting device, wherein the process of obtaining sag data according to the inclination data obtained by the inclination sensor is a well-known technology for those skilled in the art and will not be described in detail. When monitoring in time series, each type of monitoring data is arranged in the order of distribution in time series to obtain each monitoring data sequence. It should be noted that the data collection frequency and length of the monitoring data sequence in time series can be set by the implementer according to the specific implementation situation, and are not limited here.

[0069] Since the environment of the collector line may change frequently, it is more difficult to analyze the ice coverage on the surface of the collector line, and the change characteristics of the collected monitoring data of the collector line melting ice are also more complex, resulting in large differences in the monitoring data changes of the collector line melting ice line obtained at different stages. Therefore, in the process of analyzing the obtained monitoring data, the response conditions of the monitoring data of the collector line melting ice at different stages need to be switched to different degrees. Response analysis is required to cope with the frequent changes in the environment of the collector line. Among them, in different time periods, the monitoring data obtained by the collector line end ice melting device changes in complex situations. For example, in a period of low temperature and relatively stable time, the change in environmental wind speed causes the sag data and tension data obtained by monitoring to have corresponding data change characteristics. It can be understood that even if there is no correlation change feature between the changes in different monitoring data in the same time period, the time difference between the data changes and the degree of data changes in different monitoring data is small. Therefore, the collector line ice melting operation is judged based on the ice coverage of the collector line analyzed by the monitoring data, and the accuracy of the analysis of the monitoring data will directly affect the efficient remote communication switching of the intelligent switch for collector line ice melting.

[0070] In order to more accurately analyze the monitoring data under the frequently changing environment, each monitoring data sequence is divided and processed. In some possible implementation methods of the embodiments of the present invention, each monitoring data sequence adopts a sequence segmentation algorithm to obtain subsequences. In the embodiments of the present invention, the Variable-Length Subsequence Clustering in Time Series algorithm is selected for division. The Variable-Length Subsequence Clustering in Time Series algorithm extracts multiple subsequences of variable length through a sliding window for feature analysis, and optimizes the segmentation results through clustering iteration, and adaptively obtains subsequences with locally identical trend characteristics. It should be noted that the Variable-Length Subsequence Clustering in Time Series algorithm is a well-known technical means well known to those skilled in the art and will not be elaborated here.

[0071] S2: In each monitoring data sequence, the response characteristics of each subsequence are obtained according to the degree of stable difference in time series between each subsequence and other subsequences; based on the distribution of response characteristics of subsequences in the monitoring data sequence, the response characteristic sequence of each monitoring data sequence is obtained.

[0072] The changes of each monitoring data sequence are analyzed. Therefore, the stability of the subsequences in time series reflects the response influence of each monitoring data sequence itself. First, the influence of the changes of each subsequence is analyzed to obtain the response characteristics of each subsequence.

[0073] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the response feature can be found in Figure 2 , which shows a flow chart of a method for obtaining a response feature provided by an embodiment of the present invention, the method comprising the following steps:

[0074] S201: Obtain the characteristic vector of each moment for each subsequence through a time series decomposition algorithm; the characteristic vector is composed of a trend component, a period component and a residual component corresponding to the moment.

[0075] The values ​​of the three components at each moment are obtained through the time series decomposition of the subsequence to form a feature vector, thereby characterizing the change trend characteristics of the data at each moment. In the specific implementation of the embodiment of the present invention, the STL (Seasonal and Trend decomposition using Loess) time series decomposition algorithm is used for each subsequence to obtain the trend component, period component and residual component, and the trend component, period component and residual component corresponding to each moment in the subsequence are composed of the feature vector of each moment, reflecting the state change characteristics at different moments in the subsequence. It should be noted that the STL algorithm is a well-known technical means well known to those skilled in the art and will not be described in detail here.

[0076] S202: In each subsequence, the relative stability of each moment is obtained according to the similarity between the feature vectors of each moment and other moments.

[0077] The degree of difference in the feature vectors between moments in the subsequence is then used to reflect the stability of each moment. Each moment is combined with the degree of approximation of the feature vectors of each other moment to comprehensively represent the relative consistency of the change trend of each moment in the subsequence. The higher the relative stability of a moment, the less significant the monitoring change frequency of that moment in the subsequence is, which facilitates the subsequent analysis of the entire subsequence.

[0078] Preferably, in some possible implementations of the embodiments of the present invention, the method for acquiring relative stability includes:

[0079] First, in each subsequence, the cosine similarity of the feature vector between each moment and every other moment is used as the approximation between each moment and every other moment. The larger the cosine similarity, the more similar the feature vectors are and the higher the approximation between the moments.

[0080] Furthermore, the relative stability of each moment is obtained by combining the approximation of each moment with all other moments. In the embodiment of the present invention, the average of the approximation of each moment with all other moments is taken as the relative stability of each moment, and all the approximations are combined in the form of the average to reflect the overall approximation of the moment with other moments. The higher the overall approximation, the more stable the change of the moment in the subsequence is.

[0081] S203: fitting the relative stability curves at all moments in each subsequence to obtain the stability characteristic curve of each subsequence; and obtaining the response characteristics of each subsequence according to the difference between each subsequence and other subsequences in the monitoring data sequence on the stability characteristic curve.

[0082] The change distribution characteristics between subsequences are further analyzed through the overall change distribution at continuous moments. Combined with the differences in relative stability between subsequences, the response characteristics of each subsequence are obtained, thereby reflecting the significance of each subsequence in the change of monitoring data. The more significant the subsequence performance, the greater the impact on the response.

[0083] In a specific embodiment of the present invention, the least squares method can be used to perform curve fitting on the relative stability of the moments in the time series to obtain a stable characteristic curve. The ordinate of the stable characteristic curve is the relative stability, and the abscissa is time. It should be noted that the method of performing curve fitting using the least squares method is a technical means well known to those skilled in the art and will not be elaborated here.

[0084] The stability characteristic curve reflects the change in the degree of stability. Further, in the monitoring data sequence, each subsequence is analyzed for differences in stability characteristics with other subsequences, and then the response characteristics of the subsequence are obtained by combining them to reflect the significant difference between the subsequence and other subsequences in the monitoring data sequence. The higher the overall difference, the greater the influence of the subsequence on the response change, which facilitates the subsequent analysis of the entire monitoring data sequence.

[0085] Preferably, in some possible implementations of the embodiments of the present invention, the method for acquiring the response feature includes:

[0086] First, in each monitoring data sequence, the difference characteristics between each subsequence and each other subsequence are obtained according to the difference between the stable characteristic curves of each subsequence and each other subsequence. In the embodiment of the present invention, the KL divergence is used to calculate the difference between the stable characteristic curves of each subsequence and each other subsequence to obtain the difference characteristics. The larger the KL divergence, the higher the deviation of the distribution difference between the stable characteristic curve representations, and therefore the larger the difference characteristics.

[0087] In other embodiments of the present invention, the difference in curve distribution can also be calculated using Euclidean distance or DTW value. It should be noted that KL (Kullback-Leibler) divergence is also relative entropy, and DTW value is the similarity value between sequences calculated by dynamic time warping (DTW) algorithm. KL divergence, Euclidean distance or DTW value are technical means well known to those skilled in the art and will not be elaborated here.

[0088] The difference features of each subsequence and all other subsequences are further combined to obtain the response feature of each subsequence. In the embodiment of the present invention, the mean of the difference features of each subsequence and all other subsequences is used as the response feature of each subsequence, and combined in the form of the mean to reflect the overall difference between each subsequence and other subsequences.

[0089] At this point, the analysis of the change stability of each subsequence is completed, and the response characteristics of each subsequence are obtained.

[0090] Further considering the complex characteristics of changes in different monitoring data, different monitoring data sequences may not be able to guarantee consistent division results when dividing subsequences for analysis. The case with the most divided subsequences is used as a reference for constructing the response feature sequence of all monitoring data sequences to ensure that the length of the response feature sequence can meet all division situations and that the length of the response feature sequence among all monitoring data sequences is consistent, so as to facilitate the subsequent treatment of different situations of different monitoring data between changing time intervals.

[0091] Preferably, in some possible implementations of the embodiments of the present invention, among all monitoring data sequences, the total number of subsequences when the number of subsequences is the largest is taken as the sequence number, and the total amount of data in the response feature sequence is set as the sequence number, that is, the total amount of numbers in the response feature sequence corresponding to each monitoring data sequence is the sequence number.

[0092] Furthermore, for any monitoring data sequence, the response features in the monitoring data sequence are sequentially placed into the response feature sequence according to the time sequence of the subsequences. If the response features in the monitoring data sequence cannot fill the response feature sequence, the remaining vacancies in the response feature sequence are filled with zeros to obtain the response feature sequence of the monitoring data sequence. That is, only the monitoring data sequence with the most subsequences can fill the response feature sequence with the corresponding response features. The difference between the number of subsequences and the number of sequences in the remaining monitoring data sequences is the number of zeros that need to be filled at the end of the response feature sequence to make the data amount in all the response feature sequences equal.

[0093] It can be understood that when the response conditions of different monitoring data during the collector line ice melting monitoring process are relatively consistent, the time interval of the monitoring data feature changes should be close, that is, the difference in the subsequence division results is small, and the difference between the obtained response feature sequences is also small. On the contrary, if the response conditions between different monitoring data are very different, the difference between the obtained response feature sequences is also large. By limiting the response feature sequence for zero-filling analysis, different monitoring data can be further compared and analyzed.

[0094] S3: According to the degree of deviation between each monitoring data sequence and other monitoring data sequences in the distribution of response feature sequences, the effective response of each monitoring data sequence is obtained.

[0095] Since the response feature sequence is set to correspond to the most subsequences, the differences of the response feature sequences are obtained respectively, and the difference analysis is performed not only on the value of the response feature, but also on the division of the subsequences. This avoids the error caused by the fact that when the subsequence response feature values ​​are similar but the gap between the divided subsequence intervals is large when the data distribution characteristics in the monitoring data sequence are divided. The effective response of each monitoring data sequence is obtained through the distribution deviation of the response feature sequence between the monitoring data sequence and other monitoring data sequences. The effective response is more accurate in the more significant monitoring data sequence.

[0096] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the effective response situation includes:

[0097] First, according to the difference between each monitoring data sequence and each other monitoring data sequence in the response feature sequence, the feature difference significance of each monitoring data sequence and each other monitoring data sequence is obtained. In an embodiment of the present invention, the Manhattan distance can be used to calculate the difference between the response feature sequences to obtain the feature difference significance. The larger the Manhattan distance, the greater the deviation between the response feature sequences, the more significant the representation of the monitoring data sequence in data change, and the more significant the impact on the response feature.

[0098] The effective response of each monitoring data sequence is further obtained by combining the significance of the characteristic difference between each monitoring data sequence and all other monitoring data sequences. In an embodiment of the present invention, the mean of the significance of the characteristic difference between each monitoring data sequence and all other monitoring data sequences is used as the effective response of each monitoring data sequence, and is combined in the form of a mean to comprehensively characterize the significance of the difference between the monitoring data sequence and each other monitoring data sequence. When the monitoring data sequence reflects more significant data change characteristics, the greater the impact on the effective response for remote communication switching.

[0099] S4: adjusting the prediction judgment of the monitoring data by the effective response of the monitoring data sequence to obtain a communication judgment value; performing remote communication switching of the switch according to the communication judgment value.

[0100] In order to improve the efficiency of ice melting, the ice cover situation is analyzed in advance through the prediction of multiple monitoring data to determine whether to perform ice melting. However, considering the different effective responses of the monitoring data sequences, the judgment of each monitoring data sequence needs to be adjusted in combination with the effective response situation during the analysis. The higher the effective response situation, the higher the influence on the corresponding monitoring data in the response judgment process. Therefore, when switching judgment is made based on all monitoring data, the adjustment degree is higher. Finally, the communication judgment value can more accurately reflect the difference between the monitoring situation and the threshold for subsequent control judgment.

[0101] Preferably, in some possible implementation methods of the embodiments of the present invention, the method for obtaining the communication judgment value can be found in Figure 3 , which shows a flow chart of a method for obtaining a communication judgment value provided by an embodiment of the present invention, the method comprising the following steps:

[0102] S401: For any monitoring data sequence, normalize the effective response of the monitoring data sequence to obtain the judgment confidence of the monitoring data sequence.

[0103] In an embodiment of the present invention, the judgment confidence of the monitoring data sequence is obtained by calculating the proportion of the effective response of the monitoring data sequence in the effective response and value of all monitoring data sequences, and the judgment confidence is obtained by normalizing the data through the sum normalization method. The higher the judgment confidence, the higher the response impact of the monitoring data sequence, and the more reliable the result of the judgment analysis. It should be noted that normalization is a technical means well known to those skilled in the art, and other normalization processing options may be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0104] S402: Obtain a predicted value of the monitoring data sequence; and use the difference between the predicted value of the monitoring data sequence and the corresponding preset monitoring threshold as a judgment feature value of the monitoring data sequence.

[0105] In the embodiment of the present invention, the EMA algorithm is used to obtain the monitoring data of the next moment as the predicted value for the monitoring data sequence. Through the predictive analysis of the monitoring data, the remote communication control process is made more efficient and the adverse effects caused by severe icing are reduced. It should be noted that the EMA (Exponential Moving Average) algorithm is a weighted moving average algorithm. The EMA algorithm is a well-known technical means for prediction by those skilled in the art and will not be described in detail here.

[0106] The severity of the abnormal situation is further reflected by the gap between the predicted situation and the threshold. The difference between the predicted value of the monitoring data sequence and the corresponding preset monitoring threshold is used as the judgment characteristic value. The larger the judgment characteristic value is, the more serious the impact of icing on the collector line may be. Among them, the preset monitoring threshold corresponds to one for each monitoring data sequence. For example, when the monitoring data sequence corresponds to humidity, the preset monitoring threshold can be set to 95%; when the monitoring data sequence corresponds to wind speed, the preset monitoring threshold can be set to 10m / s. The specific value of the preset monitoring threshold is set according to the specific monitoring data of the specific implementation situation, and is not limited here.

[0107] S403: combining the judgment characteristic value and the judgment confidence of the monitoring data sequence to obtain the judgment coefficient of the monitoring data sequence; fusing the judgment coefficients of all monitoring data sequences to obtain the communication judgment value.

[0108] Finally, after adjusting the judgment of each monitoring data sequence, all judgment results are combined to obtain the communication judgment value. In an embodiment of the present invention, the product of the judgment characteristic value and the judgment confidence of the monitoring data sequence is obtained to obtain the judgment coefficient of the monitoring data sequence, and the two are combined in the form of multiplication. The larger the judgment characteristic value, the more serious the abnormal situation reflected by the monitoring data sequence. The larger the judgment confidence, the more significant the response of the monitoring data sequence to the judgment, and the higher the credibility of the judgment result. Therefore, the larger the judgment coefficient, the more significant the ice cover abnormality reflected by the monitoring data sequence, and the higher the degree of communication switching required.

[0109] The judgment coefficients of all monitoring data sequences are further integrated to obtain the communication judgment value. In an embodiment of the present invention, the sum of the judgment coefficients of all monitoring data sequences is normalized as the communication judgment value and integrated in the form of accumulation. The higher the overall judgment coefficient of all monitoring data sequences, the higher the need for communication switching.

[0110] Therefore, the remote communication switching of the switch is finally performed based on the communication judgment value. In some possible implementation methods of the embodiment of the present invention, when the communication judgment value is greater than the preset judgment threshold, it means that the monitored data changes and develops more seriously, and a remote switching control signal is sent to the collector line ice-melting intelligent switch. The remote switching control signal is transmitted to the ice-melting intelligent switch at the end of the collector line through 4G network communication or optical fiber communication. The ice-melting intelligent switch implements the three-phase short circuit of the end line according to the received remote switching control signal, and starts the heating system through the three-phase short circuit to melt the ice on the collector line, so as to ensure efficient ice melting of the line. Otherwise, it means that the ice covering of the collector line is relatively normal, and the remote switching control signal is not sent. Among them, in the embodiment of the present invention, the preset judgment threshold can be set to 0.7, and the specific value can be set according to the specific implementation scenario. The specific value is not limited here.

[0111] It should be noted that, in order to facilitate calculation, the data involved in the calculation in the embodiment of the present invention are all pre-processed to eliminate the dimension effect. The specific means of eliminating the dimension effect are technical means well known to those skilled in the art and are not limited here.

[0112] In summary, the present invention takes into account that the environment is complex and causes different monitoring data to change frequently, which will cause different monitoring data to have different response effects on the switch switching state. Therefore, for different monitoring data, subsequent analysis is performed by dividing subsequences, and the stability of each subsequence and other subsequences in time series is further analyzed. The response characteristics of each subsequence are obtained through the degree of change stability difference, which is reflected in the characterization of the influence of the monitoring data on the response in the subsequence time series. The response characteristic sequence of each monitoring data sequence is obtained based on the distribution of the response characteristics, so as to reflect the influence of the change of data in each monitoring data sequence with time series through the response characteristic sequence, thereby analyzing the overall influence of the monitoring data sequence. Through the response characteristic sequence distribution deviation between the monitoring data sequence and other monitoring data sequences, the effective response of each monitoring data sequence is obtained, and the more significant the effective response is, the more accurate it is. Therefore, the prediction judgment of the monitoring data is further adjusted by the effective response situation, and the communication judgment value is obtained, and the necessary degree of the switching judgment is analyzed more accurately, so as to perform remote communication switching based on the communication judgment value. The present invention considers the difference in response states of different monitoring data, integrates analysis and adjusts the judgment on the switching state of the collector line ice-melting intelligent switch, so that the remote communication switching of the collector line ice-melting intelligent switch is more efficient and the remote switching is more reliable.

[0113] Embodiment 2:

[0114] The embodiment of the present invention provides a remote communication switching system for a collector line ice melting intelligent switch, see Figure 4 , which shows a structural diagram of a remote communication switching system for a collector line ice-melting intelligent switch provided by an embodiment of the present invention. The system 500 includes: a data acquisition module 501, a response analysis module 502, an effective response analysis module 503 and a remote communication switching module 504.

[0115] The data acquisition module 501 is used to acquire different monitoring data sequences of ice melting of collector lines; and divide each monitoring data sequence into subsequences;

[0116] The response analysis module 502 is used to obtain the response characteristics of each subsequence in each monitoring data sequence according to the degree of stability difference between each subsequence and other subsequences in time series; based on the distribution of the response characteristics of the subsequences in the monitoring data sequence, obtain the response characteristic sequence of each monitoring data sequence;

[0117] The effective response analysis module 503 is used to obtain the effective response of each monitoring data sequence according to the deviation degree of each monitoring data sequence from other monitoring data sequences in the response feature sequence distribution;

[0118] The remote communication switching module 504 is used to adjust the prediction judgment of the monitoring data by monitoring the effective response of the data sequence to obtain a communication judgment value; and perform remote communication switching of the switch according to the communication judgment value.

[0119] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. Since the specific implementation process of a remote communication switching system for a collector line ice melting intelligent switch in this embodiment is the same as the specific implementation process of a remote communication switching method for a collector line ice melting intelligent switch in the above embodiment, it will not be repeated here.

[0120] Embodiment 3:

[0121] The present invention also provides a computer device, see Figure 5 , which shows a schematic diagram of the structure of a computer device provided by an embodiment of the present invention, the computer device 600 at least includes, but is not limited to, a memory 601, a processor 602 and a communication interface 603 that can be interconnected and communicated via a bus 610.

[0122] The memory 601 may include a large-capacity memory for data or instructions, including at least one type of computer-readable storage medium, the readable storage medium including a flash memory, a hard disk, a read-only memory, a magnetic disk or an optical disk, etc. The memory 801 may be inside or outside the computer device. In an embodiment of the present invention, the memory 601 may be a read-only memory for storing computer program instructions.

[0123] The processor 602 may include one or more integrated circuits for controlling the overall operation of the computer device 600. In the embodiment of the present invention, the processor 602 reads and executes computer program instructions stored in the memory 601 to implement Figure 1 The steps of a remote communication switching method for a collector line ice-melting intelligent switch in embodiment 1 are shown.

[0124] The communication interface 603 is used to establish a communication connection between the computer device 600 and other electronic devices. In an embodiment of the present invention, the communication interface 603 is used to connect the computer device 600 to an external terminal through a network, establish a data transmission channel and a communication connection between the computer device 600 and the external terminal, etc.

[0125] Embodiment 4:

[0126] The embodiment of the present invention also provides a computer-readable storage medium, such as a flash memory, a hard disk, an optical disk or a server, etc., on which computer program instructions are stored, and the program instructions are executed by a processor to implement corresponding functions. The computer-readable storage medium in the embodiment of the present invention is used to store computer program instructions for a remote communication switching method for a collector line ice melting intelligent switch, and the processor executes the steps of the above-mentioned remote communication switching method for a collector line ice melting intelligent switch.

[0127] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0128] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A remote communication switching method for a collector line ice melting intelligent switch, characterized in that: The method comprises: Obtain different monitoring data sequences of ice melting of collector lines; divide each monitoring data sequence into subsequences; In each monitoring data sequence, the response characteristics of each subsequence are obtained according to the degree of stability difference between each subsequence and other subsequences in time series; based on the distribution of the response characteristics of the subsequences in the monitoring data sequence, the response characteristic sequence of each monitoring data sequence is obtained; According to the degree of deviation between each monitoring data sequence and other monitoring data sequences in the distribution of response characteristic sequences, the effective response of each monitoring data sequence is obtained; Adjust the prediction judgment of the monitoring data by monitoring the effective response of the data sequence to obtain the communication judgment value; perform remote communication switching of the switch according to the communication judgment value; The method for obtaining the effective response situation includes: According to the difference between each monitoring data sequence and each other monitoring data sequence in the response feature sequence, the feature difference significance between each monitoring data sequence and each other monitoring data sequence is obtained; The mean of the characteristic difference significance between each monitoring data series and all other monitoring data series is taken as the effective response of each monitoring data series; The method of adjusting the prediction judgment of the monitoring data by the effective response of the monitoring data sequence to obtain the communication judgment value includes: For any monitoring data sequence, normalize the effective response of the monitoring data sequence to obtain the judgment confidence of the monitoring data sequence; Obtaining a predicted value of the monitoring data sequence; using the difference between the predicted value of the monitoring data sequence and the corresponding preset monitoring threshold as a judgment feature value of the monitoring data sequence; The product of the judgment characteristic value and the judgment confidence of the monitoring data sequence is used as the judgment coefficient of the monitoring data sequence; and the sum of the judgment coefficients of all monitoring data sequences is normalized and used as the communication judgment value.

2. According to claim 1, a remote communication switching method for a collector line ice melting intelligent switch is characterized in that: The method for obtaining the response characteristics includes: For each subsequence, the characteristic vector of each moment is obtained through the time series decomposition algorithm; the characteristic vector is composed of the trend component, period component and residual component corresponding to the moment; In each subsequence, the relative stability of each moment is obtained according to the similarity of the feature vector between each moment and other moments; Fit the relative stability curves of all moments in each subsequence to obtain the stability characteristic curve of each subsequence; According to the difference between each subsequence and other subsequences in the monitoring data sequence on the stable characteristic curve, the response characteristics of each subsequence are obtained.

3. A remote communication switching method for a collector line ice melting intelligent switch according to claim 2, characterized in that: The relative stability of each moment is obtained according to the similarity between the feature vectors of each moment and other moments, including: In each subsequence, the cosine similarity of the feature vector between each moment and every other moment is taken as the proximity between each moment and every other moment; The average of the similarities between each moment and all other moments is taken as the relative stability of each moment.

4. A remote communication switching method for a collector line ice melting intelligent switch according to claim 2, characterized in that: The step of obtaining the response characteristics of each subsequence according to the difference between each subsequence and other subsequences in the monitoring data sequence on the stable characteristic curve includes: In each monitoring data sequence, according to the difference between the stable characteristic curves of each subsequence and each other subsequence, the difference characteristics between each subsequence and each other subsequence are obtained; The mean of the difference features between each subsequence and all other subsequences is taken as the response feature of each subsequence.

5. A remote communication switching method for a collector line ice melting intelligent switch according to claim 1, characterized in that: The method for acquiring the response characteristic sequence includes: In all monitoring data sequences, the total number of subsequences with the largest number of subsequences is taken as the number of sequences; the total amount of data in the response feature sequence is set as the number of sequences; For any monitoring data sequence, the response features in the monitoring data sequence are placed into the response feature sequence in sequence according to the time sequence of the subsequence. If the response features in the monitoring data sequence cannot fill the response feature sequence, the remaining vacancies in the response feature sequence are filled with zeros to obtain the response feature sequence of the monitoring data sequence.

6. A remote communication switching method for a collector line ice melting intelligent switch according to claim 1, characterized in that: The remote communication switching of the switch according to the communication judgment value includes: When the communication judgment value is greater than the preset judgment threshold, a remote switching control signal is sent to the collector line ice-melting intelligent switch; otherwise, no remote switching control signal is sent.

7. A remote communication switching method for a collector line ice melting intelligent switch according to claim 1, characterized in that: The obtaining of the predicted value of the monitoring data sequence includes: The EMA algorithm is used on the monitoring data sequence to obtain the monitoring data at the next moment as the predicted value.

8. A remote communication switching method for a collector line ice melting intelligent switch according to claim 1, characterized in that: Each monitoring data sequence adopts a sequence segmentation algorithm to obtain a subsequence.

Images