Line icing prediction method and related equipment
By preprocessing and analyzing transmission line data, using nuclear limit learning machine and whale optimization algorithm to build a hybrid nuclear limit learning machine model, the problem of inaccurate line ice prediction caused by the diversity of factors in the existing technology is solved, and the problem of inaccurate line ice prediction is achieved. Accurate ice-cover thickness prediction.
Patent Information
- Application Number
- CN202411256780.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-09
- Publication Date
- 2024-12-06
AI Technical Summary
The prior art has too many factors when predicting the ice covering of transmission lines, making it difficult to determine which factor is used to determine the specific thickness of the ice covering of transmission lines, resulting in inaccurate prediction.
By preprocessing the collected data, the data with the highest correlation with the line ice-covered thickness is determined, and a hybrid nuclear limit learning machine model is constructed using the nuclear limit learning machine and whale optimization algorithm to predict the line ice-covered thickness.
Improve the accuracy and reliability of line ice-cover prediction, select high correlation factors as ice-cover prediction input, use the fast learning and strong generalization capabilities of the hybrid core limit learning machine to automatically search for optimal model parameters, and realize Effective prediction of the overflow of transmission lines.
Smart Images

Figure CN119090086A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of electric power technology, and in particular to a line icing prediction method and related equipment. Background Art
[0002] With climate change and global warming, the frequent increase in extreme weather events has brought great challenges to the safe and stable operation of the power grid. Transmission lines are prone to ice in high-altitude areas and mountainous areas with high humidity and low temperature in winter. Icing on the lines can easily cause accidents such as conductor dancing, insulator flashover, broken wires and poles, which can lead to large-scale power outages and serious obstruction of power supply.
[0003] However, there are too many factors that affect line icing in the prior art, and it is difficult to determine which factor should be used to determine the specific thickness of line icing during specific predictions. Summary of the invention
[0004] In view of this, the purpose of the present application is to propose a line icing prediction method and related equipment.
[0005] Based on the above purpose, the present application provides a line icing prediction method, comprising: Preprocessing the collected data to obtain first data; Analyze the first data to determine the data with the highest correlation with the ice thickness of the line, and obtain second data; Presetting a first parameter of the kernel extreme learning machine, and optimizing the first parameter using a whale optimization algorithm to obtain a second parameter; The second parameter is used as a parameter of the kernel extreme learning machine, and based on the second data, the kernel extreme learning machine is used to predict the ice thickness of the line.
[0006] In a possible implementation manner, preprocessing the collected data to obtain the first data includes: The collected data is iteratively interpolated for missing values, monitored for outliers, and normalized to obtain the first data.
[0007] In a possible implementation, the iterative interpolation of missing values on the collected data includes: Performing column-wise randomization processing on the set of collected data to obtain a plurality of first subsets; interpolating any missing value in the first subset to obtain a second subset; In response to at least one missing value in the second subset, interpolating any missing value therein to obtain a third subset; In response to no missing values existing in the third subset, the interpolation process is stopped.
[0008] In a possible implementation, the performing abnormal value monitoring on the collected data includes: Randomly select a first number of data from the collected data as subsamples and place a node; A feature dimension is randomly specified in the sub-sample, and a cutting point is randomly specified within the data range of the sub-sample; the value range of the cutting point is between the maximum value and the minimum value in the data of the sub-sample; Generate a hyperplane based on the cutting point, divide the data space of the subsample into two subspaces, use data less than the cutting point as the left branch of the node, and use data greater than or equal to the cutting point as the right branch of the node; Repeating the designation of feature dimensions and cutting points in the left branch and the right branch, respectively, and continuing to generate hyperplanes until only one data exists in the current node; An abnormality score value of each data in the sub-sample is calculated, and an abnormality monitoring is performed on the collected data based on the abnormality score value.
[0009] In a possible implementation, analyzing the first data to determine the data with the highest correlation with the line ice thickness to obtain the second data includes: The line ice thickness is used as a reference data column; Classifying the first data according to types, and using each of the classified first data as a comparison data sequence; Calculating the correlation coefficient matrix of each comparison data sequence and the reference data column respectively; Calculating the degree of association between each of the comparison data sequences and the reference data sequence according to the association coefficient matrix; The comparison data sequence corresponding to the highest association value is used as the second data.
[0010] In a possible implementation, optimizing the first parameter by using a whale optimization algorithm to obtain a second parameter includes: constructing a first vector according to the first parameter; generating a preset number of whale populations based on the first vector; Each individual whale in the whale population swims toward the individual whale closest to the optimal solution, and updates the current position of the individual whale according to the distance between the current position of the individual whale and the position of the individual whale corresponding to the optimal solution; The mean square error of the current individual whale position is calculated, and the parameter corresponding to the individual whale with the smallest mean square error is used as the second parameter.
[0011] Based on the same inventive concept, the embodiment of the present application also provides a line icing prediction device, including: A preprocessing module is configured to preprocess the collected data to obtain first data; An analysis module is configured to analyze the first data, determine the data with the highest correlation with the ice thickness of the line, and obtain second data; An optimization module is configured to preset a first parameter of the kernel extreme learning machine and optimize the first parameter using a whale optimization algorithm to obtain a second parameter; The prediction module is configured to use the second parameter as a parameter of the kernel extreme learning machine, and based on the second data, use the kernel extreme learning machine to predict the ice thickness of the line.
[0012] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a line icing prediction method as described in any one of the above items is implemented.
[0013] Based on the same inventive concept, an embodiment of the present application further provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute any of the above-mentioned line icing prediction methods.
[0014] Based on the same inventive concept, an embodiment of the present application further provides a computer program product, which includes computer program instructions, and the computer instructions are used to enable the computer program product to execute any of the above-mentioned line icing prediction methods.
[0015] From the above, it can be seen that the line icing prediction method and related equipment provided by the present application obtain first data by preprocessing the collected data; analyze the first data to determine the data with the highest correlation with the line icing thickness to obtain second data; preset the first parameter of the core extreme learning machine, and use the whale optimization algorithm to optimize the first parameter to obtain the second parameter; use the second parameter as the parameter of the core extreme learning machine, and based on the second data, use the core extreme learning machine to predict the line icing thickness. The embodiment of the present application performs the preprocessing of the collected data by interpolating missing values and monitoring outliers of meteorological and line multi-service collected data such as temperature, humidity, wind speed, and conductor tension, and then normalizing the data. Further determine the degree of correlation between each type of influencing factor in the collected data and the icing thickness, and select factors with higher correlation as the input for icing prediction. Next, a whale optimization-hybrid core extreme learning machine prediction model is constructed, and the advantages of the hybrid core extreme learning machine with fast learning speed and strong generalization ability are utilized, and the whale optimization algorithm is used to automatically search for the optimal hybrid core extreme learning machine model parameters to carry out icing prediction of transmission lines. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0017] Figure 1 This is a schematic diagram of the process flow of the line icing prediction method according to an embodiment of the present application; Figure 2 A schematic diagram of the interpolation process of an embodiment of the present application; Figure 3 This is a schematic diagram of the structure of a line icing prediction device according to an embodiment of the present application; Figure 4 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0018] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.
[0019] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0020] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.
[0021] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.
[0022] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.
[0023] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.
[0024] As mentioned in the background technology section, with climate change and global warming, the frequent increase in extreme weather events has brought great challenges to the safe and stable operation of the power grid. Transmission lines are prone to ice in high-altitude areas and mountainous areas with high humidity and low temperature in winter. Icing on the lines can easily cause accidents such as conductor dancing, insulator flashover, broken wires and poles, resulting in large-scale power outages and serious obstruction of power supply.
[0025] However, there are too many factors that affect line icing in the prior art, and it is difficult to determine which factor should be used to determine the specific thickness of line icing during specific predictions.
[0026] Based on the above considerations, the embodiment of the present application proposes a line icing prediction method, which obtains first data by preprocessing the collected data; analyzes the first data to determine the data with the highest correlation with the line icing thickness to obtain second data; presets the first parameter of the kernel extreme learning machine, and optimizes the first parameter using the whale optimization algorithm to obtain the second parameter; uses the second parameter as the parameter of the kernel extreme learning machine, and based on the second data, uses the kernel extreme learning machine to predict the line icing thickness. The embodiment of the present application completes the preprocessing of the collected data by interpolating missing values and monitoring outliers of meteorological and line multi-service collected data such as temperature, humidity, wind speed, and conductor tension, and then normalizes the data. The degree of correlation between each type of influencing factor in the collected data and the icing thickness is further determined, and the factors with higher correlation are selected as the input for icing prediction. Next, a whale optimization-hybrid kernel extreme learning machine prediction model is constructed, and the advantages of the hybrid kernel extreme learning machine with fast learning speed and strong generalization ability are utilized, and the whale optimization algorithm is used to automatically search for the optimal hybrid kernel extreme learning machine model parameters to carry out icing prediction of transmission lines.
[0027] The technical solutions of the embodiments of the present application are described in detail below through specific examples.
[0028] refer to Figure 1 The line icing prediction method of the embodiment of the present application comprises the following steps: Step S101, preprocessing the collected data to obtain first data; Step S102, analyzing the first data to determine the data with the highest correlation with the line ice thickness to obtain second data; Step S103, presetting a first parameter of the kernel extreme learning machine, and optimizing the first parameter using a whale optimization algorithm to obtain a second parameter; Step S104: using the second parameter as a parameter of the kernel extreme learning machine, and using the kernel extreme learning machine to predict the ice thickness of the line based on the second data.
[0029] For step S101, line icing is affected by a variety of factors, such as line attributes, micro-meteorological and terrain factors. The acquired data includes monitoring data from the power grid company, transmission line attribute data, line tension data, and meteorological department monitoring data. The above data come from different industries and have different data structures. Different data collection cycles range from seconds to minutes or even days, and are affected by the collection terminals. There are different degrees of data anomalies and missing data. In order to accurately predict the icing of distribution lines, it is necessary to preprocess various data features. The specific preprocessing process is as follows.
[0030] In some embodiments, the preprocessing of the collected data to obtain the first data includes: performing iterative interpolation of missing values, outlier monitoring and normalization processing on the collected data to obtain the first data.
[0031] In some embodiments, the iterative interpolation of missing values of the collected data includes: performing column-wise randomization processing on the set of collected data to obtain multiple first subsets; interpolating any missing value in the first subset to obtain a second subset; in response to the presence of at least one missing value in the second subset, interpolating any missing value therein to obtain a third subset; in response to the absence of missing values in the third subset, stopping the interpolation processing.
[0032] In the embodiment of the present application, the core idea is to calculate the similarity between the missing samples and the non-missing samples, and then estimate the missing values by the weighted values of the non-missing samples.
[0033] refer to Figure 2 , which is a schematic diagram of the interpolation process of an embodiment of the present application.
[0034] like Figure 2 As shown, for samples with missing data , where an arbitrary variable , express The observed value of express Observations other than express The remaining observations except the missing values. To interpolate missing values, we first build a decision tree forest and then , Training y ~ x The model will Take as input prediction and fill missing values , that is, completed However, in the data set, there are usually missing values in multiple variables, so it is necessary to complete the interpolation of the entire sample in an iterative way.
[0035] In some embodiments, the outlier monitoring of the collected data includes: randomly selecting a first number of data from the collected data as a subsample and placing a node; randomly specifying a feature dimension in the subsample, and randomly specifying a cutting point within the data range of the subsample; the value range of the cutting point is between the maximum value and the minimum value in the data of the subsample; generating a hyperplane based on the cutting point, dividing the data space of the subsample into two subspaces, taking data less than the cutting point as the left branch of the node, and taking data greater than or equal to the cutting point as the right branch of the node; repeatedly specifying feature dimensions and cutting points in the left branch and the right branch, respectively, and continuing to generate hyperplanes until there is only one data in the current node; calculating the outlier score value for each data in the subsample, and monitoring the collected data for outliers based on the outlier score value.
[0036] In this embodiment, the data may contain abnormal values due to factors such as low stability or failure of the acquisition equipment and sensors during the acquisition process. In order to ensure the accuracy of the prediction, these abnormal values need to be processed. The specific process is as follows: Step 1: Randomly select from the training data The data are taken as subsamples and put into the root node of an isolated tree; Step 2: Randomly assign a feature dimension in the subsample q , within the current node data range, randomly generate a cutting point p —— Cutting point p The value range is the specified dimension generated in the current sub-sample data q between the maximum and minimum values of ; Step 3: Generate a hyperplane with the cutting point to divide the current node data space into two subspaces: q Upper less than p The data is regarded as the left branch of the current node, which is greater than or equal to p The data is used as the right branch of the current node; Step 4: Recursively call steps 2 and 3 in each branch, and continuously construct new child nodes until there is only one data in the child node or the height limit has been reached. ; Step 5: Repeat steps 1 to 4 until the generated t An isolated tree iTree is used to build an isolated forest; Step 6: In the isolation forest, for each data Perform anomaly assessment and calculate anomaly score using the following formula S :
[0037] in:
[0038]
[0039]
[0040] in, For each data The average path length in the forest, For adjustment items, is the harmonic number, is the Euler constant 0.5772156649, represents the number of samples in the data set, represents a single sample in the dataset, Representation sample The path length.
[0041] Anomaly score value S Between [0,1]. S If it is close to 1, it must be an outlier. S If the anomaly score is much less than 0.5, it is definitely not an anomaly. If the scores of all points are around 0.5, there is probably no anomaly in the sample. S When it is 0.6, it is considered a potential outlier.
[0042] Furthermore, all outliers are removed and the data is normalized.
[0043] Normalizing the data can limit the data to a certain range, eliminate the impact of different dimensions and dimensional units on the data analysis results, improve the comparability between data indicators, reduce the negative impact caused by singular sample data, and improve machine learning accuracy. After data normalization, the data is finally converted into a standard normal distribution with a mean of 0 and a variance of 1, as shown in the following formula:
[0044] In the formula is the sample mean and is the sample mean difference, represents sample data, z represents normalized data, and the calculation formula is as follows:
[0045] in, represents the number of data points in the sample, Indicates the first data.
[0046] Further, with respect to step S102, the first data is analyzed to determine the data with the highest correlation with the ice thickness of the line, thereby obtaining the second data.
[0047] In some embodiments, the first data is analyzed to determine the data with the highest correlation with the line ice thickness to obtain the second data, including: taking the line ice thickness as a reference data column; classifying the first data according to type, and taking each classified first data as a comparison data sequence; respectively calculating the correlation coefficient matrix between each comparison data sequence and the reference data column; calculating the correlation degree between each comparison data sequence and the reference data column according to the correlation coefficient matrix; and taking the comparison data sequence corresponding to the highest correlation value as the second data.
[0048] In this embodiment, icing of the transmission line is the result of the combined effects of multiple factors such as temperature, humidity and wind speed, and there is a nonlinear mapping relationship between the ice thickness of the line and its influencing factors. Therefore, it is necessary to select the main icing influencing factors from many influencing factors. This application selects the main icing influencing factors by quantifying the degree of correlation between the ice thickness and each influencing factor.
[0049] The embodiment of the present application determines whether the reference data column and several comparison data columns are closely related by determining the geometric similarity between them. The higher the similarity, the closer the influencing factor is to the reference data in trend, that is, the stronger the correlation is with the reference data.
[0050] The specific steps are as follows: 1) Determine the reference data column: Select the line ice thickness as the reference data column X 0, as shown in the following formula:
[0051] Various factors affecting ice thickness as comparative data series X i , as shown below:
[0052] in m Represents the number of influencing factors related to ice thickness.
[0053] 2) Calculate the correlation coefficient: Calculate each comparison data sequence separately X i With reference data column X Correlation coefficient matrix of elements corresponding to 0 :
[0054] in is the correlation coefficient, which indicates the correlation between the reference data column and the i The correlation between subsequences is determined by the value of [0.321, 0.588]. The best resolution is obtained when the value is in the range of [0.321, 0.588]. , Indicates the first samples.
[0055] 3) Calculate the correlation: Calculate the correlation between the reference sequence and each comparison data sequence , usually using the mean method, the calculation formula is as follows:
[0056] in, Indicates the number of samples in the data column.
[0057] 4) Association ranking: The larger the value between the reference data column and the comparison data sequence, the closer the relationship between the two; conversely, the smaller the value of the result, the looser the relationship.
[0058] Further, for step S103 and step S104, the first parameter is optimized using the whale optimization algorithm to obtain the second parameter, including: constructing a first vector according to the first parameter; generating a preset number of whale populations based on the first vector; each whale individual in the whale population swims toward the whale individual closest to the optimal solution, and updates the current position of the whale individual according to the distance between the current position of the whale individual and the position of the whale individual corresponding to the optimal solution; calculating the mean square error of the current position of the whale individual, and taking the parameter corresponding to the whale individual with the smallest mean square error as the second parameter.
[0059] In this embodiment, compared with the traditional feedforward neural network, the model structure of the extreme learning machine (ELM) is simple and generalizable, and the output weight can be solved directly without iterative optimization, thereby greatly accelerating the learning process. However, the standard ELM may encounter difficulties in dealing with certain complex nonlinear problems. On this basis, the kernel function is introduced to improve the ELM, namely the kernel extreme learning machine (KELM). The nonlinear mapping ability of the kernel function is used to make the nonlinear samples linearly separable, making the data easier to be processed by the linear model in the new feature space.
[0060] The feedforward neural network expression can be written as:
[0061] In the formula, is the input vector of the neural network, is the hidden layer input function, is the weight vector, is the mapping matrix.
[0062] The kernel matrix of the kernel function in KELM is:
[0063] in is the kernel matrix, , is the element in the mapping matrix, is the kernel function.
[0064] The expressions of KELM output layer and output layer weight vector are shown as follows:
[0065] In the formula, is the penalty factor, is the unit diagonal matrix, Output matrix for the target In order to give full play to the advantages of local kernel function and global kernel function, the hybrid kernel function composed of the two is used to improve the regression performance of KELM. The expression of the hybrid kernel function is:
[0066]
[0067]
[0068] in, represents the local kernel function, represents the global kernel function, represents the mixed kernel function, is any positive real number, is a parameter item, is any non-negative real number, Input sample data for training, Input sample data for training or testing, is the weight coefficient. In order to avoid large errors caused by manually setting the initial parameters, the whale optimization algorithm is used to , , , , Parameters are optimized.
[0069] The whale optimization algorithm (WOA) is a heuristic optimization algorithm that simulates the special bubble net foraging and hunting behavior of humpback whales. It consists of three stages: the hunting process, the hunting process, and the target search.
[0070] (1) Hunting process After the whale group finds prey, the entire group constantly adjusts its position and swims towards the position of the individual whale closest to the optimal solution, so that the whales gradually approach the optimal solution. At this stage, the mathematical model of the algorithm is:
[0071] In the above formula, is the number of iterations, is the current optimal whale individual position, is the current position of the individual whale, and is the coefficient vector, and are random numbers, To control the parameters, it decreases linearly from 2 to 0 as the number of iterations increases, that is:
[0072] in, is the maximum number of iterations.
[0073] (2) Hunting process At this stage, the whale population uses spiral bubble nets for prey hunting, and the formula for updating the spiral position is as follows:
[0074] In the formula, is the distance between the current whale individual position and the best whale individual, b is the helical constant coefficient, l is a random number between [-1,1].
[0075] (3) Target Search The mathematical model of the target prey search phase is:
[0076] In the formula represents the position vector of a randomly selected whale from the population, and represents the coefficient vector, is the current iteration number. Conduct in-office searches; Move away from prey to find new prey and conduct a global search.
[0077] The WOA algorithm is used to optimize the HKELM model parameters. The mean square error is used as the fitness function, and the HKELM model parameters to be optimized are selected as vectors. , the optimized parameter set with the smallest mean square error is selected as the optimal parameters of the HKELM model to improve the prediction performance of the HKELM model.
[0078] It can be seen from the above embodiments that the line icing prediction method described in the embodiment of the present application obtains the first data by preprocessing the collected data; analyzes the first data to determine the data with the highest correlation with the line icing thickness to obtain the second data; presets the first parameter of the kernel extreme learning machine, and optimizes the first parameter using the whale optimization algorithm to obtain the second parameter; uses the second parameter as the parameter of the kernel extreme learning machine, and based on the second data, uses the kernel extreme learning machine to predict the line icing thickness. The embodiment of the present application performs the preprocessing of the collected data by interpolating missing values and monitoring outliers of meteorological and line multi-service collected data such as temperature, humidity, wind speed, and conductor tension, and then normalizes the data. The degree of correlation between each type of influencing factor in the collected data and the icing thickness is further determined, and the factors with higher correlation are selected as the input for icing prediction. Next, a whale optimization-hybrid kernel extreme learning machine prediction model is constructed, and the advantages of the hybrid kernel extreme learning machine with fast learning speed and strong generalization ability are utilized, and the whale optimization algorithm is used to automatically search for the optimal hybrid kernel extreme learning machine model parameters to carry out icing prediction of transmission lines.
[0079] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.
[0080] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, 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.
[0081] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a line icing prediction device.
[0082] refer to Figure 3 , the line icing prediction device comprises: The preprocessing module 31 is configured to preprocess the collected data to obtain first data; An analysis module 32 is configured to analyze the first data, determine the data with the highest correlation with the ice thickness of the line, and obtain second data; The optimization module 33 is configured to preset a first parameter of the kernel extreme learning machine and optimize the first parameter using a whale optimization algorithm to obtain a second parameter; The prediction module 34 is configured to use the second parameter as a parameter of the kernel extreme learning machine, and predict the ice thickness of the line by using the kernel extreme learning machine based on the second data.
[0083] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.
[0084] The device of the above embodiment is used to implement the corresponding line icing prediction method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.
[0085] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the line icing prediction method described in any of the above embodiments is implemented.
[0086] Figure 4 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.
[0087] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.
[0088] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.
[0089] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.
[0090] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).
[0091] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 ).
[0092] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.
[0093] The electronic device of the above embodiment is used to implement the corresponding line icing prediction method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.
[0094] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the line icing prediction method described in any of the above embodiments.
[0095] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.
[0096] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the line icing prediction method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.
[0097] Based on the same inventive concept, corresponding to the line icing prediction method described in any of the above embodiments, the present disclosure also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer so that the computer and / or the processor execute the line icing prediction method. Corresponding to the execution subject corresponding to each step in each embodiment of the line icing prediction method, the processor that executes the corresponding step may belong to the corresponding execution subject.
[0098] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the line icing prediction method as described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.
[0099] A person skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.
[0100] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power / ground connections to the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device may be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present application are to be implemented (that is, these details should be fully within the scope of understanding of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or with changes in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.
[0101] Although the present application has been described in conjunction with specific embodiments of the present application, many alternatives, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the discussed embodiments.
[0102] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.
Claims
1. A line icing prediction method, characterized in that: include: Preprocessing the collected data to obtain first data; Analyze the first data to determine the data with the highest correlation with the ice thickness of the line, and obtain second data; Presetting a first parameter of the kernel extreme learning machine, and optimizing the first parameter using a whale optimization algorithm to obtain a second parameter; The second parameter is used as a parameter of the kernel extreme learning machine, and based on the second data, the kernel extreme learning machine is used to predict the ice thickness of the line.
2. The method according to claim 1, characterized in that The preprocessing of the collected data to obtain the first data includes: The collected data is iteratively interpolated for missing values, monitored for outliers, and normalized to obtain the first data.
3. The method according to claim 2, characterized in that The iterative interpolation of missing values for the collected data comprises: Performing column-wise randomization processing on the set of collected data to obtain a plurality of first subsets; interpolating any missing value in the first subset to obtain a second subset; In response to at least one missing value in the second subset, interpolating any missing value therein to obtain a third subset; In response to no missing values existing in the third subset, the interpolation process is stopped.
4. The method according to claim 2, characterized in that: The performing abnormal value monitoring on the collected data includes: Randomly select a first number of data from the collected data as subsamples and place a node; A feature dimension is randomly specified in the sub-sample, and a cutting point is randomly specified within the data range of the sub-sample; the value range of the cutting point is between the maximum value and the minimum value in the data of the sub-sample; Generate a hyperplane based on the cutting point, divide the data space of the subsample into two subspaces, use data less than the cutting point as the left branch of the node, and use data greater than or equal to the cutting point as the right branch of the node; Repeating the designation of feature dimensions and cutting points in the left branch and the right branch, respectively, and continuing to generate hyperplanes until only one data exists in the current node; An abnormality score value of each data in the sub-sample is calculated, and abnormal value monitoring is performed on the collected data based on the abnormality score value.
5. The method according to claim 1, characterized in that The analyzing the first data to determine the data with the highest correlation with the line ice thickness to obtain the second data includes: The line ice thickness is used as a reference data column; Classifying the first data according to types, and using each of the classified first data as a comparison data sequence; Calculating the correlation coefficient matrix of each comparison data sequence and the reference data column respectively; Calculating the degree of association between each of the comparison data sequences and the reference data sequence according to the association coefficient matrix; The comparison data sequence corresponding to the highest association value is used as the second data.
6. The method according to claim 1, characterized in that The step of optimizing the first parameter by using the whale optimization algorithm to obtain the second parameter includes: constructing a first vector according to the first parameter; generating a preset number of whale populations based on the first vector; Each individual whale in the whale population swims toward the individual whale closest to the optimal solution, and updates the current position of the individual whale according to the distance between the current position of the individual whale and the position of the individual whale corresponding to the optimal solution; The mean square error of the current individual whale position is calculated, and the parameter corresponding to the individual whale with the smallest mean square error is used as the second parameter.
7. A line icing prediction device, characterized in that: include: A preprocessing module is configured to preprocess the collected data to obtain first data; An analysis module is configured to analyze the first data, determine the data with the highest correlation with the ice thickness of the line, and obtain second data; An optimization module is configured to preset a first parameter of the kernel extreme learning machine and optimize the first parameter using a whale optimization algorithm to obtain a second parameter; The prediction module is configured to use the second parameter as a parameter of the kernel extreme learning machine, and based on the second data, use the kernel extreme learning machine to predict the ice thickness of the line.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.
9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 6.
10. A computer program product, comprising computer program instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 6.
