Transformer parameter optimization method and system based on intelligent sensor

Abstract

The present application provides a transformer parameter optimization method and system based on intelligent sensors, which relates to the field of transformer technology, including: collecting monitoring data, identifying data trend characteristics of monitoring data sets respectively, performing adjustment analysis of associated parameter mapping, collecting parameter values ​​of target transformers, performing directional optimization, generating M optimized parameter values, sending them to the parameter control unit of the target transformer, and adjusting the parameters of the target transformer. The present application can solve the technical problem in the prior art that the operating efficiency of transformers is unstable due to the reliance of traditional transformers on fixed sensor arrangements and manual adjustments. By using intelligent sensor arrays for efficient data collection, combined with data analysis and parameter mapping, real-time optimization and automatic adjustment of transformer parameters are achieved, thereby improving the operating efficiency and reliability of transformers.

Description

Technical Field

[0001] The present application relates to the technical field of transformers, and in particular to a transformer parameter optimization method and system based on intelligent sensors. Background Art

[0002] With the rapid development of the power industry and the gradual advancement of smart grids, transformers are an indispensable part of the power system, and their operating efficiency and safety are of great significance to the stable operation of the entire power grid. However, there are some problems in the design and operation of traditional transformers, such as unreasonable parameter settings, low operating efficiency, and untimely fault warnings. These problems not only affect the service life of the transformer, but may also pose a threat to the stable operation of the power grid. Traditional transformers usually rely on fixed sensor arrangements, which cannot cover all the key parameters of the transformer and cannot adapt to the different operating conditions and special needs of the transformer.

[0003] In summary, the prior art has a technical problem in that the operation efficiency of the transformer is unstable because the traditional transformer relies on a fixed sensor arrangement and manual adjustment. Summary of the invention

[0004] The purpose of this application is to provide a transformer parameter optimization method and system based on intelligent sensors to solve the technical problem in the prior art that the operating efficiency of the transformer is unstable due to the traditional transformer's reliance on fixed sensor arrangements and manual adjustments.

[0005] In view of the above problems, the present application provides a transformer parameter optimization method and system based on intelligent sensors.

[0006] In a first aspect, the present application provides a transformer parameter optimization method based on smart sensors, which is implemented by a transformer parameter optimization system based on smart sensors, wherein the transformer parameter optimization method based on smart sensors includes: using a sensor array of an interactive target transformer to collect monitoring data within a preset monitoring window to generate K monitoring data sets, wherein the sensor array includes K monitoring sensors; performing data trend feature identification on the K monitoring data sets respectively to determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data feature values ​​and K trend identifiers; performing associated parameter mapping adjustment analysis based on the K monitoring data feature values ​​and the K trend identifiers to determine M parameter optimization directions and M parameter optimization scales; collecting the parameter value of the target transformer at the current moment to generate M real-time parameter values; performing directional optimization on the M real-time parameter values ​​based on the M parameter optimization directions and the M parameter optimization scales to generate M optimized parameter values; sending the M optimized parameter values ​​to the parameter control unit of the target transformer to adjust the parameters of the target transformer.

[0007] In a second aspect, the present application further provides a transformer parameter optimization system based on an intelligent sensor, which is used to execute the transformer parameter optimization method based on an intelligent sensor as described in the first aspect, wherein the transformer parameter optimization system based on an intelligent sensor includes: a monitoring data acquisition module, the monitoring data acquisition module is used to interact with the sensor array of the target transformer to collect monitoring data within a preset monitoring window, and generate K monitoring data sets, wherein the sensor array includes K monitoring sensors; a data processing module, the data processing module is used to identify data trend features of the K monitoring data sets respectively, and determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data feature values ​​and K trend identifiers; A parameter optimization analysis module, the parameter optimization analysis module is used to perform associated parameter mapping adjustment analysis based on the K monitoring data characteristic values ​​and the K trend identifiers, and determine M parameter optimization directions and M parameter optimization scales; a real-time parameter acquisition module, the real-time parameter acquisition module is used to collect the parameter value of the target transformer at the current moment, and generate M real-time parameter values; a directional optimization module, the directional optimization module is used to perform directional optimization on the M real-time parameter values ​​based on the M parameter optimization directions and the M parameter optimization scales, and generate M optimized parameter values; a parameter control module, the parameter control module is used to send the M optimized parameter values ​​to the parameter control unit of the target transformer, and perform parameter adjustment on the target transformer.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The sensor array of the interactive target transformer is used to collect monitoring data within a preset monitoring window to generate K monitoring data sets, wherein the sensor array includes K monitoring sensors; the data trend characteristics of the K monitoring data sets are respectively identified to determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data characteristic values ​​and K trend identifications; the associated parameter mapping adjustment analysis is performed based on the K monitoring data characteristic values ​​and the K trend identifications to determine M parameter optimization directions and M parameter optimization scales; the parameter value of the target transformer at the current moment is collected to generate M real-time parameter values; the M real-time parameter values ​​are directional optimized based on the M parameter optimization directions and the M parameter optimization scales to generate M optimized parameter values; the M optimized parameter values ​​are sent to the parameter control unit of the target transformer to adjust the parameters of the target transformer. That is to say, by using the intelligent sensor array for efficient data collection, combined with data analysis and parameter mapping, the real-time optimization and automatic adjustment of transformer parameters are realized, and the operation efficiency and reliability of the transformer are improved.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 This is a flow chart of the transformer parameter optimization method based on smart sensors in this application.

[0013] Figure 2 This is a schematic diagram of the structure of the transformer parameter optimization system based on intelligent sensors in this application.

[0014] Explanation of the accompanying drawings: monitoring data acquisition module 11, data processing module 12, parameter optimization and analysis module 13, real-time parameter acquisition module 14, directional optimization module 15, parameter control module 16. DETAILED DESCRIPTION

[0015] This application provides a transformer parameter optimization method and system based on intelligent sensors, which solves the technical problem in the prior art that the operating efficiency of the transformer is unstable due to the traditional transformer's reliance on fixed sensor arrangements and manual adjustments. By using an intelligent sensor array for efficient data collection, combined with data analysis and parameter mapping, real-time optimization and automatic adjustment of transformer parameters are achieved, thereby improving the operating efficiency and reliability of the transformer.

[0016] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0017] For example, please refer to the attached Figure 1 The present application provides a transformer parameter optimization method based on an intelligent sensor, wherein the transformer parameter optimization method based on an intelligent sensor is applied to a transformer parameter optimization system based on an intelligent sensor, and the transformer parameter optimization method based on an intelligent sensor specifically comprises the following steps:

[0018] Step 1: The sensor array of the interactive target transformer collects monitoring data within a preset monitoring window to generate K monitoring data sets, wherein the sensor array includes K monitoring sensors.

[0019] Specifically, an array of K monitoring sensors is deployed on the target transformer to capture the key operating parameters of the transformer, including but not limited to temperature, voltage, and current. According to the operating characteristics or maintenance requirements of the transformer, a monitoring window is preset, which can be a few minutes, hours, or longer. At the beginning of the preset monitoring window, all sensors are started to ensure that they are in normal working condition. Each sensor starts recording data, including continuous analog signals or discrete digital signals. The data collected by each sensor within the monitoring window is organized into a data set, with a total of K such sets, each corresponding to a sensor. Comprehensive monitoring of the transformer status is achieved, providing a reliable data basis for subsequent analysis.

[0020] Step 2: Perform data trend feature identification on the K monitoring data sets respectively to determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data feature values ​​and K trend identifiers.

[0021] Specifically, each monitoring data set is analyzed to identify its data trend characteristics. Data trend characteristics refer to the changing trend of the data in the data set over time or other variables, such as rising, falling or remaining stable. Extract numerical values ​​representing its characteristics from the data, such as mean, median, variance, kurtosis, skewness, etc. Construct a change curve for each monitoring data set in chronological order, generate K data change curves, perform trend identification, and assign one or more trend identifiers to each monitoring data set. Based on the results of data trend feature identification, determine a monitoring data feature set for each monitoring data set. This feature set includes two main parts: monitoring data characteristic values ​​and trend identifiers. Through data trend feature identification and determination of monitoring data feature sets, the main characteristics and changing trends of each monitoring data set can be better understood.

[0022] Step 3: Based on the K monitoring data characteristic values ​​and the K trend identifiers, an associated parameter mapping adjustment analysis is performed to determine M parameter optimization directions and M parameter optimization scales.

[0023] Specifically, K monitoring data characteristic values ​​and trend marks are associated with the operating parameters of the transformer to find the intrinsic connection between the data characteristics and the transformer parameters. The impact of data characteristics on the operation of the transformer is analyzed to determine which parameters may need to be adjusted and find parameter optimization paths that can improve the performance of the transformer. Based on the analysis results, M parameter optimization directions are determined. The parameter optimization direction refers to the direction in which the transformer parameters should be adjusted, such as increasing or decreasing the value of a certain parameter. The choice of optimization direction depends on the monitoring data characteristic values ​​and trend marks, as well as the problems to be solved. M parameter optimization directions mean that there are M different parameters or parameter combinations that need to be considered. At the same time, M parameter optimization scales need to be determined. The parameter optimization scale refers to the size or degree of a single adjustment of the parameter. The determination of the optimization scale also depends on the monitoring data characteristic values ​​and trend marks, as well as the goals to be achieved. The M parameter optimization scales correspond to the M parameter optimization directions, and each direction has a corresponding adjustment scale. Conduct correlation analysis on the characteristic values ​​and trend indicators of monitoring data to determine the parameters that need to be optimized and their optimization directions. Based on the analysis results of the characteristic values ​​of monitoring data, determine the appropriate optimization scale, form a set of parameter optimization plans, determine how to adjust the parameters of the transformer to optimize its performance, and more effectively improve the performance and reliability of the transformer.

[0024] Step 4: Collect the parameter values ​​of the target transformer at the current moment and generate M real-time parameter values.

[0025] Specifically, the operating status of the target transformer is monitored in real time, and the parameter values ​​at the current moment are collected, including voltage, current, temperature, power factor, etc. According to the number of parameters in the monitoring data and the analysis requirements, M key parameter values ​​are extracted from the collected real-time monitoring data, and each real-time parameter value represents a key parameter of the transformer at the current moment. The real-time parameter value can be used for real-time monitoring and control, as well as comparative analysis with historical data, representing the operating status of the transformer at the current moment. By collecting and generating the parameter values ​​of the transformer in real time and monitoring the operating status of the transformer in real time, it is helpful to promptly discover and solve potential problems, thereby improving the performance and reliability of the transformer.

[0026] Step 5: Directionally optimize the M real-time parameter values ​​based on the M parameter optimization directions and the M parameter optimization scales to generate M optimized parameter values.

[0027] Specifically, M real-time parameter values ​​are adjusted according to M parameter optimization directions and M parameter optimization scales, and optimization targets are set to ensure that the parameter values ​​are within a preset operating range, such as a preset operating temperature range, a preset current and voltage range, and a preset noise range. The adjusted parameter values ​​are evaluated for fitness and the effect of the adjusted parameter values. When the fitness and performance of the adjusted parameter values ​​within the preset operating range meet the preset standards, the adjusted parameter values ​​at this time are used as the current optimal parameter values, i.e., the M optimized parameter values. Through directional optimization, parameter settings that can maximize transformer performance or minimize cost can be found, which helps to improve the operating efficiency of the transformer and reduce energy consumption.

[0028] Step six: Send the M optimized parameter values ​​to the parameter control unit of the target transformer to adjust the parameters of the target transformer.

[0029] Specifically, the M optimized parameter values ​​represent the best parameter settings under certain conditions, and the obtained M optimized parameter values ​​are sent to the parameter control unit of the target transformer through an appropriate communication interface. The parameter control unit is a part of the transformer control system, which is responsible for receiving the parameter settings and performing corresponding adjustments. The parameter control unit receives the optimized parameter values ​​and converts them into executable instructions for adjusting the operating parameters of the transformer, such as voltage, current, temperature, etc. The parameter control unit adjusts the operating parameters of the transformer according to the received optimized parameter values, such as changing the current setting, voltage regulation, temperature control, etc. Through parameter adjustment, the target transformer will operate according to the optimized parameter values ​​to achieve the preset optimization goals, such as improving efficiency, reducing energy consumption, and enhancing stability. By applying the optimized parameter values ​​to the parameter control unit of the transformer, the preset optimization goals can be achieved, which helps to improve the performance and reliability of the transformer while reducing operating costs and maintenance requirements.

[0030] Further, step 2 of this application includes:

[0031] The K monitoring data sets are traversed to perform fluctuation variance calculations to generate K fluctuation variances; based on the K fluctuation variances and a preset variance threshold, K fitting scales are determined; the K fitting scales are distributed to the K data fitting channels for the K monitoring data sets to generate the K monitoring data feature values; and the K monitoring data feature values ​​are added to the K monitoring data feature sets.

[0032] Specifically, K monitoring data sets are traversed, and for each monitoring data set (K in total), its fluctuation variance is calculated. Fluctuation variance is a statistic that measures the degree of data fluctuation. It is obtained by calculating the average of the square of the difference between each data point in the data set and the average value, and is usually used to describe the degree of deviation of the data point from the average value. For each data set, after calculating its fluctuation variance, K fluctuation variance values ​​will be obtained, and each value corresponds to a monitoring data set. According to factors such as the type, operating conditions, and historical data of the transformer, one or more variance thresholds are set to determine whether the fluctuation variance is within an acceptable range. The fluctuation variance of each monitoring data set is compared with the preset variance threshold to determine whether the fluctuation of the data set is too large or too small, that is, whether the data is stable. Based on the comparison results of the fluctuation variance and the threshold, K fitting scales are determined. The fitting scale can be a numerical value used to quantify the degree of matching between the data set and the preset standard. If the fluctuation variance is lower than the threshold, it means that the data is relatively stable; if the fluctuation variance is higher than the threshold, it means that the data fluctuates greatly.

[0033] According to the determined fitting scale, each monitoring data set is distributed to the corresponding data fitting channel, and each channel will use a specific fitting method (such as linear fitting, polynomial fitting, etc.) to process the data. By using statistical methods, machine learning algorithms or other data processing techniques, find the value that best represents the characteristics of the data set, that is, the eigenvalue. Through the data fitting process, eigenvalues ​​are extracted from each data set to represent the main characteristics or trends of the data set. A monitoring data feature set is created for each monitoring data set, and the generated monitoring data eigenvalues ​​are added to the corresponding monitoring data feature set. Each feature set now contains data eigenvalues ​​and other relevant information (such as trend identification). By extracting meaningful eigenvalues ​​from the original monitoring data and organizing them into feature sets, the operating status of the transformer can be monitored and analyzed.

[0034] Furthermore, the present application also includes the following steps:

[0035] Construct change curves for the K monitoring data sets in chronological order to generate K data change curves; identify trends for the K data change curves respectively to generate K trend identifiers; and add the K trend identifiers into the K monitoring data feature sets.

[0036] Specifically, the data points in each data set are sorted in chronological order and plotted in the coordinate system to form K data change curves, each of which shows the change of the corresponding monitoring parameter over time. The curve can be a line graph in the time domain or a spectrogram in the frequency domain, depending on the analysis needs and the characteristics of the data. Each data change curve is analyzed to identify its trend or pattern. Trend identification refers to analyzing the trend of the curve to determine whether the data is rising, falling or stable, which is achieved by calculating the slope of the curve, using statistical methods or applying machine learning algorithms. According to the identification results, a trend identification is generated for each change curve, which can be a label describing the data change trend, such as rising, falling, stable, and fluctuating. The generated trend identification is added to the corresponding monitoring data feature set, and the feature set now contains data feature values ​​and trend identification. By constructing data change curves and identifying trends, the trend of monitoring parameters over time can be better understood.

[0037] Furthermore, the present application also includes the following steps:

[0038] The K fitting spaces of the K data fitting channels are respectively called; the K monitoring data sets are input into the K fitting spaces to obtain K fitting scatter point sets; m fitting scatter points are randomly selected from the K fitting scatter point sets as K fitting center sets; the K fitting center sets are searched and iterated in the K fitting spaces according to the K fitting scales until a preset number of iterations or a preset search stop constraint is met to obtain K iterative fitting center sets; the K iterative fitting center sets are screened to generate K target fitting centers; and the K monitoring data corresponding to the K target fitting centers are used as the K monitoring data feature values.

[0039] Furthermore, the present application also includes the following steps:

[0040] The preset search stop constraint is that the difference in clustering density between two adjacent search iterations is less than or equal to a preset difference threshold.

[0041] Specifically, a fitting space is assigned to each data fitting channel, which defines the parameter range and constraints of data fitting. Each fitting space is a mathematical model or algorithm used to fit a data set. The choice of fitting space depends on the characteristics of the data and the problem to be solved. For example, a linear regression model, a neural network or other machine learning algorithm is used as a fitting space. Each monitoring data set is input into the corresponding fitting space to find the mathematical model or parameter that best represents the data set. Through fitting, each data set will be converted into a set of fitting scatter points, and the positions of these scatter points in the fitting space represent the characteristics of the original data. From these fitting scatter point sets, m scatter points are randomly selected as K fitting center sets. These fitting centers are representatives of the fitting scatter points. The number of fitting centers m selected depends on the characteristics of the data and the accuracy to be achieved.

[0042] In each fitting space, the fitting center set is iteratively searched according to the fitting scale to find better fitting centers so that these centers can more accurately represent the original data. During the iteration process, each fitting center set is randomly moved in the direction of its corresponding fitting scale to generate a new set, called the stage fitting center set. This process will continue until the preset number of iterations or the search stop constraint is met. The preset search stop constraint is a specified condition used to determine whether the iteration process should stop. When the difference in cluster density between two adjacent iterations is less than or equal to the preset difference threshold, it indicates that a relatively dense area has been reached, the search process has converged sufficiently, and there is no need to continue iterating, and the iteration stops. Cluster density refers to the ratio of the number of fitting scatter points distributed in the area constructed with the fitting center as the center and the fitting scale as the radius to the area of ​​the upper area, which is used to measure the representativeness of the fitting center. The preset difference threshold is a pre-set value used to evaluate the significance of the change in cluster density. The current iterative fitting center set is regarded as the optimal or sufficiently ideal fitting center set.

[0043] According to certain screening criteria (such as scatter point density, fitting error, etc.), the optimal fitting center is selected from the iterative fitting center set to form the target fitting center set. The monitoring data points corresponding to each target fitting center are selected as monitoring data feature values, and these feature values ​​are added to the corresponding monitoring data feature set. Through fitting and iteration, meaningful feature points, namely fitting centers, can be extracted from the original monitoring data to more accurately understand and optimize the operating parameters of the transformer. In addition, by presetting the search stop constraint, the efficiency and accuracy of the iterative process can be ensured, and excessive iterations and waste of computing resources can be avoided.

[0044] Furthermore, the present application also includes the following steps:

[0045] In the K fitting spaces, the K fitting center sets are respectively moved in random directions according to the K fitting scales to obtain K stage fitting center sets; it is determined whether the K stage clustering density sets of the K stage fitting center sets are greater than or equal to the K clustering density sets of the K fitting center sets; if so, the K stage fitting center set iterations are used as the starting point, and a search iteration is performed in the K fitting spaces according to the K fitting scales until a preset number of iterations or a preset search stop constraint is satisfied to obtain the K iterative fitting center sets.

[0046] Specifically, in the K fitting spaces, each fitting center is moved in a random direction according to its corresponding fitting scale to explore the fitting space and find a potentially better solution. The distance or amplitude of the movement is determined by the corresponding fitting scale, which may reflect the volatility or uncertainty of the data set. After the movement, each fitting center set is updated to a stage fitting center set, representing the intermediate state after a random move. It is determined whether the clustering density of each stage fitting center set is greater than or equal to the clustering density of the original fitting center set. If it is greater, the random move is considered to be valid, and the stage fitting center set is used as a new starting point to perform search iterations in the fitting space. This process will be repeated until the preset number of iterations or the search stop constraint is met. The preset search stop constraint is intended to control the iterative process and avoid the iterative process from being endless. After the stop condition is met, the final obtained fitting center set is the iterative fitting center set, which represents the better or best fitting center position found in the fitting space. By iteratively optimizing the fitting center set in the fitting space, a better solution, that is, a fitting center that better represents the characteristics of the data set, can be found, which helps to improve the accuracy of data analysis and parameter optimization.

[0047] Further, step five of this application includes:

[0048] Based on the M parameter optimization directions and the M parameter optimization scales, the M real-time parameter values ​​are adjusted to obtain M adjusted parameter values; taking a preset operating temperature range, a preset current and voltage range, and a preset noise range as optimization targets, and combining the K monitoring data characteristic values, the M adjusted parameter values ​​are identified for fitness to obtain M adjustment fitnesses; when the M adjustment fitnesses meet the preset fitnesses, the M adjustment parameter values ​​are used as the M optimization parameter values.

[0049] Specifically, according to the determined M parameter optimization directions and M parameter optimization scales, the M real-time parameter values ​​are adjusted, the parameter values ​​are increased or decreased, the performance of the transformer is optimized, and the adjusted parameter values ​​are used as the M adjustment parameter values, which represent the best operating state that the transformer can achieve under the current optimization direction and scale. The preset operating temperature range, the preset current and voltage range, and the preset noise range are used as optimization targets, and the basic conditions that need to be met during the operation of the transformer are set. The temperature cannot be too high or too low, the current and voltage must be within a safe range, and the noise level is controlled within a certain limit. The K monitoring data feature values ​​are combined with the M adjustment parameter values, and the effect of the adjustment parameter values ​​is evaluated through fitness identification. Fitness identification is an evaluation method in the optimization algorithm, which is used to evaluate the quality of the solution or parameter setting. In transformer parameter optimization, the purpose of fitness identification is to determine whether the adjusted parameter value meets the preset operating conditions, such as temperature, current and voltage, and noise range. For each adjusted parameter value, its corresponding fitness value is calculated. The preset fitness is set according to the optimization target and actual application requirements. When the M adjustment fitnesses meet the preset fitnesses, these adjustment parameter values ​​are considered to be optimized parameter values. At this time, the parameter values ​​can meet the preset operating conditions, and the M adjustment parameter values ​​are taken as the final optimized parameter values. Through real-time monitoring and data-based optimization strategies, the transformer can achieve more efficient operation, thereby improving its performance and reliability.

[0050] In summary, the transformer parameter optimization method based on intelligent sensors provided in this application has the following technical effects:

[0051] The sensor array of the interactive target transformer is used to collect monitoring data within a preset monitoring window to generate K monitoring data sets, wherein the sensor array includes K monitoring sensors; the data trend characteristics of the K monitoring data sets are respectively identified to determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data characteristic values ​​and K trend identifications; the associated parameter mapping adjustment analysis is performed based on the K monitoring data characteristic values ​​and the K trend identifications to determine M parameter optimization directions and M parameter optimization scales; the parameter value of the target transformer at the current moment is collected to generate M real-time parameter values; the M real-time parameter values ​​are directional optimized based on the M parameter optimization directions and the M parameter optimization scales to generate M optimized parameter values; the M optimized parameter values ​​are sent to the parameter control unit of the target transformer to adjust the parameters of the target transformer. That is to say, by using the intelligent sensor array for efficient data collection, combined with data analysis and parameter mapping, the real-time optimization and automatic adjustment of transformer parameters are realized, and the operation efficiency and reliability of the transformer are improved.

[0052] Embodiment 2, based on the same inventive concept as the transformer parameter optimization method based on intelligent sensors in the above embodiment, the present application also provides a transformer parameter optimization system based on intelligent sensors, please refer to the attached Figure 2 , the transformer parameter optimization system based on intelligent sensor includes:

[0053] The monitoring data acquisition module 11 is used to interact with the sensor array of the target transformer to collect monitoring data within a preset monitoring window to generate K monitoring data sets, wherein the sensor array includes K monitoring sensors.

[0054] The data processing module 12 is used to identify data trend characteristics of the K monitoring data sets respectively, and determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data characteristic values ​​and K trend identifiers.

[0055] The parameter optimization analysis module 13 is used to perform correlation parameter mapping adjustment analysis based on the K monitoring data characteristic values ​​and the K trend identifiers, and determine M parameter optimization directions and M parameter optimization scales.

[0056] The real-time parameter acquisition module 14 is used to acquire the parameter value of the target transformer at the current moment and generate M real-time parameter values.

[0057] The directional optimization module 15 is used to perform directional optimization on the M real-time parameter values ​​based on the M parameter optimization directions and the M parameter optimization scales to generate M optimized parameter values.

[0058] The parameter control module 16 is used to send the M optimized parameter values ​​to the parameter control unit of the target transformer to adjust the parameters of the target transformer.

[0059] Furthermore, the data processing module 12 in the transformer parameter optimization system based on intelligent sensors is also used for:

[0060] The K monitoring data sets are traversed to perform fluctuation variance calculations to generate K fluctuation variances; based on the K fluctuation variances and a preset variance threshold, K fitting scales are determined; the K fitting scales are distributed to the K data fitting channels for the K monitoring data sets to generate the K monitoring data feature values; and the K monitoring data feature values ​​are added to the K monitoring data feature sets.

[0061] Furthermore, the data processing module 12 in the transformer parameter optimization system based on intelligent sensors is also used for:

[0062] Construct change curves for the K monitoring data sets in chronological order to generate K data change curves; identify trends for the K data change curves respectively to generate K trend identifiers; and add the K trend identifiers into the K monitoring data feature sets.

[0063] Furthermore, the transformer parameter optimization system based on intelligent sensors also includes a monitoring data characteristic value module, which is used to:

[0064] The K fitting spaces of the K data fitting channels are respectively called; the K monitoring data sets are input into the K fitting spaces to obtain K fitting scatter point sets; m fitting scatter points are randomly selected from the K fitting scatter point sets as K fitting center sets; the K fitting center sets are searched and iterated in the K fitting spaces according to the K fitting scales until a preset number of iterations or a preset search stop constraint is met to obtain K iterative fitting center sets; the K iterative fitting center sets are screened to generate K target fitting centers; and the K monitoring data corresponding to the K target fitting centers are used as the K monitoring data feature values.

[0065] Furthermore, the transformer parameter optimization system based on intelligent sensors also includes an iterative fitting center set module, which is used to:

[0066] In the K fitting spaces, the K fitting center sets are respectively moved in random directions according to the K fitting scales to obtain K stage fitting center sets; it is determined whether the K stage clustering density sets of the K stage fitting center sets are greater than or equal to the K clustering density sets of the K fitting center sets; if so, the K stage fitting center set iterations are used as the starting point, and a search iteration is performed in the K fitting spaces according to the K fitting scales until a preset number of iterations or a preset search stop constraint is satisfied to obtain the K iterative fitting center sets.

[0067] Furthermore, the transformer parameter optimization system based on intelligent sensors also includes a search stop constraint module, which is used to:

[0068] The preset search stop constraint is that the difference in clustering density between two adjacent search iterations is less than or equal to a preset difference threshold.

[0069] Furthermore, the directional optimization module 15 in the transformer parameter optimization system based on intelligent sensors is also used for:

[0070] Based on the M parameter optimization directions and the M parameter optimization scales, the M real-time parameter values ​​are adjusted to obtain M adjusted parameter values; taking a preset operating temperature range, a preset current and voltage range, and a preset noise range as optimization targets, and combining the K monitoring data characteristic values, the M adjusted parameter values ​​are identified for fitness to obtain M adjustment fitnesses; when the M adjustment fitnesses meet the preset fitnesses, the M adjustment parameter values ​​are used as the M optimization parameter values.

[0071] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The transformer parameter optimization method based on intelligent sensors and the specific examples in the first embodiment are also applicable to the transformer parameter optimization system based on intelligent sensors in the present embodiment. Through the above detailed description of the transformer parameter optimization method based on intelligent sensors, those skilled in the art can clearly know the transformer parameter optimization system based on intelligent sensors in the present embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0072] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0073] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A transformer parameter optimization method based on intelligent sensors, characterized in that: include: The sensor array of the interactive target transformer collects monitoring data within a preset monitoring window to generate K monitoring data sets, wherein the sensor array includes K monitoring sensors; Respectively performing data trend feature identification on the K monitoring data sets to determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data feature values ​​and K trend identifiers; Based on the K monitoring data characteristic values ​​and the K trend identifiers, an associated parameter mapping adjustment analysis is performed to determine M parameter optimization directions and M parameter optimization scales; Collecting the parameter value of the target transformer at the current moment and generating M real-time parameter values; Directionally optimizing the M real-time parameter values ​​based on the M parameter optimization directions and the M parameter optimization scales to generate M optimized parameter values; The M optimized parameter values ​​are sent to the parameter control unit of the target transformer to adjust the parameters of the target transformer.

2. The transformer parameter optimization method based on intelligent sensor according to claim 1, characterized in that: include: Traversing the K monitoring data sets to perform fluctuation variance calculations and generate K fluctuation variances; Analyze the K fluctuation variances and the preset variance threshold to determine K fitting scales; Distributing the K fitting scales and the K monitoring data sets to K data fitting channels to generate the K monitoring data feature values; The K monitoring data feature values ​​are added into the K monitoring data feature sets.

3. The transformer parameter optimization method based on intelligent sensor according to claim 1, characterized in that: include: Constructing change curves for the K monitoring data sets in chronological order to generate K data change curves; Performing trend identification on the K data change curves respectively to generate K trend identifiers; The K trend identifiers are added to the K monitoring data feature sets.

4. The transformer parameter optimization method based on intelligent sensor according to claim 2, characterized in that: include: respectively calling the K fitting spaces of the K data fitting channels; Inputting the K monitoring data sets into the K fitting spaces to obtain K fitting scattered point sets; Randomly select m fitting scattered points from the K fitting scattered point sets as K fitting center sets; Iteratively searching the K fitting center sets in the K fitting spaces according to the K fitting scales until a preset number of iterations or a preset search stop constraint is satisfied, thereby obtaining K iterative fitting center sets; Screening the K iterative fitting center sets to generate K target fitting centers; The K monitoring data corresponding to the K target fitting centers are used as the K monitoring data feature values.

5. The transformer parameter optimization method based on intelligent sensor according to claim 4, characterized in that: include: In the K fitting spaces, the K fitting center sets are respectively moved in random directions according to the K fitting scales to obtain K stage fitting center sets; Determine whether the K stage clustering density sets of the K stage fitting center sets are greater than or equal to the K clustering density sets of the K fitting center sets. If so, take the K stage fitting center set iteration as the starting point, and perform search iterations in the K fitting spaces according to the K fitting scales until a preset number of iterations or a preset search stop constraint is met, so as to obtain the K iterative fitting center sets.

6. The transformer parameter optimization method based on intelligent sensor according to claim 4, characterized in that: The preset search stop constraint is that the difference in clustering density between two adjacent search iterations is less than or equal to a preset difference threshold.

7. The transformer parameter optimization method based on intelligent sensor according to claim 1, characterized in that: include: Performing parameter adjustment on the M real-time parameter values ​​based on the M parameter optimization directions and the M parameter optimization scales to obtain M adjusted parameter values; Taking the preset operating temperature range, the preset current and voltage range, and the preset noise range as optimization targets, combining the K monitoring data characteristic values, performing fitness identification on the M adjustment parameter values, and obtaining M adjustment fitnesses; When the M adjustment fitnesses satisfy the preset fitnesses, the M adjustment parameter values ​​are used as the M optimization parameter values.

8. Transformer parameter optimization system based on intelligent sensor, characterized in that: The steps for implementing the transformer parameter optimization method based on the intelligent sensor according to any one of claims 1 to 7, the transformer parameter optimization system based on the intelligent sensor comprises: A monitoring data acquisition module, wherein the monitoring data acquisition module is used to interact with the sensor array of the target transformer to collect monitoring data within a preset monitoring window to generate K monitoring data sets, wherein the sensor array includes K monitoring sensors; A data processing module, the data processing module is used to perform data trend feature identification on the K monitoring data sets respectively, and determine K monitoring data feature sets, wherein the K monitoring data feature sets include K monitoring data feature values ​​and K trend identifiers; A parameter optimization analysis module, the parameter optimization analysis module is used to perform an associated parameter mapping adjustment analysis based on the K monitoring data characteristic values ​​and the K trend identifiers, and determine M parameter optimization directions and M parameter optimization scales; A real-time parameter acquisition module, wherein the real-time parameter acquisition module is used to acquire the parameter value of the target transformer at the current moment and generate M real-time parameter values; A directional optimization module, the directional optimization module is used to perform directional optimization on the M real-time parameter values ​​based on the M parameter optimization directions and the M parameter optimization scales to generate M optimized parameter values; A parameter control module, wherein the parameter control module is used to send the M optimized parameter values ​​to a parameter control unit of the target transformer to adjust the parameters of the target transformer.

Images