A transformer voltage regulating method and transformer

Abstract

The application discloses a transformer voltage regulation method and a transformer, and relates to the technical field of transformer voltage regulation. The method comprises the following steps: obtaining input end voltage and output end voltage of the transformer respectively according to a set frequency, obtaining an access end voltage curve and an output end voltage curve; setting an input end normal voltage interval and an output end normal voltage interval; obtaining an input end abnormal voltage curve based on the input end normal voltage interval and the input end voltage curve, and obtaining an output end abnormal voltage curve based on the output end normal voltage interval and the output end voltage curve; the transformer voltage regulation method and the transformer. Through analysis of voltage fluctuation characteristics, corresponding time, other electrical parameters and environmental data, and the probability of automatic recovery of voltage within a specified time, and setting of a probability threshold, when the prediction probability is greater than the probability threshold, it is determined that the voltage can be automatically recovered within the specified time without regulation of the transformer voltage, and the prediction accuracy is further improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transformer voltage regulation, and in particular to a transformer voltage regulation method and a transformer. BACKGROUND

[0002] A transformer is a basic device for power transmission and distribution, widely used in industry, agriculture, transportation, urban communities and other fields. It is a device that uses the principle of electromagnetic induction to change the alternating voltage. The main components are the primary coil, secondary coil and core (magnetic core). The main functions include voltage transformation, current transformation, impedance transformation, isolation, voltage stabilization (magnetic saturation transformer), etc. When the voltage of the transformer fluctuates, it is necessary to reasonably regulate the transformer voltage to ensure the stability of the user's electricity.

[0003] The prior art with publication number CN112994040B discloses a transformer voltage regulation method and a transformer. The method is executed by a control module of the transformer, and the method comprises: obtaining the transformer voltage at a preset time interval; detecting whether the grid voltage is abnormally fluctuating based on the transformer voltage; if it is determined that the grid voltage is abnormally fluctuating and the grid voltage and the pre-stored historical voltage of the same period belong to the same kind of fluctuation, then outputting a voltage regulation signal to the load switching of the transformer based on the historical voltage and the transformer voltage of the same period, wherein the voltage regulation signal is used to instruct the load switching to regulate the output voltage of the transformer. It predicts the current abnormal transformer voltage according to the change rule of the historical voltage, and adjusts the output voltage of the transformer in advance according to the prediction result, so as to solve the problem that the output voltage is adjusted laggingly when the current grid voltage is abnormal.

[0004] However, the above-mentioned prior art only monitors and analyzes the voltage fluctuation of the grid to determine whether voltage regulation is needed, but the cause of the transformer voltage fluctuation may come from the input end, i.e. the grid side, and may also come from the output end. At the same time, current, frequency, environmental factors, equipment failure and other external factors can all affect the fluctuation of the voltage. Therefore, only analyzing the voltage fluctuation of the grid side may cause a certain deviation in the prediction, and the prediction accuracy is relatively low. SUMMARY

[0005] The purpose of the present application is to provide a transformer voltage regulation method and a transformer to solve at least one of the above-mentioned deficiencies in the prior art.

[0006] In order to achieve the above-mentioned purpose, the present application provides the following technical solution: a transformer voltage regulation method, comprising the following steps:

[0007] S1, respectively, according to the set frequency, the input end voltage of the transformer is obtained, and the output end voltage is obtained, the access end voltage curve and the output end voltage curve are obtained, wherein the input end refers to one end of the transformer connected with the power grid before voltage transformation, and the output end refers to one end of the transformer connected with the power grid after voltage transformation;

[0008] S2, the input end normal voltage interval and the output end normal voltage interval are set;

[0009] S3, the input end abnormal voltage curve is obtained based on the input end normal voltage interval and the input end voltage curve, and the output end abnormal voltage curve is obtained based on the output end normal voltage interval and the output end voltage curve, wherein the input end abnormal voltage curve is the part of the input end voltage curve exceeding the input end normal voltage interval, and the output end abnormal voltage curve is the part of the output end voltage curve exceeding the output end normal voltage interval;

[0010] S4, the input end abnormal voltage curve and the curve before the first set time interval are obtained, the input end voltage characteristic curve is obtained, and the output end abnormal voltage curve and the curve before the first set time interval are obtained, the output end voltage characteristic curve is obtained;

[0011] S5, the input end voltage deviation threshold and the output end voltage deviation threshold are set for the historical input end voltage characteristic curve and the historical output end voltage characteristic curve respectively;

[0012] S6, regression analysis is carried out based on the corresponding data of the historical input end voltage characteristic curve and the historical output end voltage characteristic curve respectively, a plurality of input end voltage abnormal category curves with error less than the input end voltage deviation threshold and a plurality of output end voltage abnormal category curves with error less than the output end voltage deviation threshold are obtained; that is, the error of each input end voltage abnormal category curve and the corresponding input end voltage characteristic curve is within the set deviation threshold, and the error of each output end voltage abnormal category curve and the corresponding output end voltage characteristic curve is within the set deviation threshold;

[0013] S7, based on the need of monitoring, the corresponding sensor is set, the transformer and the input end side and the output end side thereof are monitored, the corresponding electrical data and environmental data are obtained, and the external influencing factor is obtained, wherein the sensor monitoring electrical data includes but is not limited to monitoring high-frequency pulse current through high-frequency current sensor or coupling capacitor; high-frequency electromagnetic wave is monitored through radio frequency antenna or receiver; cable or equipment surface accumulated charge is monitored through charge coupled sensor; wherein the environmental data includes temperature, humidity, air dust content and the like;

[0014] S8, analyze the newly acquired input voltage curve and output voltage curve, the corresponding input voltage abnormal category curve and output voltage abnormal category curve, and analyze the input voltage abnormal category curve and output voltage abnormal category curve under the corresponding external factor condition, the probability of the input voltage or output voltage returning to normal;

[0015] S9, set a judgment threshold, if the probability of the input voltage or output voltage returning to normal within the second setting time is greater than the judgment threshold, no voltage regulation of the transformer is needed, otherwise, voltage regulation of the transformer is performed.

[0016] Further, the S8 comprises the following steps:

[0017] Based on the error, the input voltage abnormal category curve and the output voltage abnormal category curve are matched with the input voltage curve and the output voltage curve;

[0018] The input voltage feature curve of the input voltage curve and the output voltage feature curve of the output voltage curve are extracted;

[0019] The average error of the input voltage feature curve and each input voltage abnormal category curve, and the average error of the output voltage feature curve and each output voltage abnormal category curve are calculated;

[0020] The input voltage abnormal category curve with the minimum average error is matched with the input voltage curve, and the output voltage abnormal category curve with the minimum average error is matched with the output voltage curve;

[0021] After extracting the input voltage feature curve from the input voltage curve, the input voltage judgment curve is obtained, and after extracting the output voltage feature curve from the output voltage curve, the output voltage judgment curve is obtained;

[0022] According to the difference of the input voltage abnormal category curve, the input voltage automatic recovery probability prediction sub-model is obtained by training the Bayesian time series analysis model based on the input voltage judgment curve, the corresponding external influencing factor and the recovery condition of the input voltage within the second setting time, which is used to predict the probability of the input voltage recovery within the second setting time based on the input input voltage judgment curve and external influencing factor;

[0023] According to the different input voltage abnormality category curves, the output voltage judgment curve and the corresponding external influence factor are matched based on the recovery of the output voltage in the second setting time, the Bayesian time series analysis model is trained to obtain the output voltage automatic recovery probability prediction sub-model, which is used to predict the probability of the output voltage recovery in the second setting time based on the input output voltage judgment curve and the external influence factor;

[0024] All input voltage automatic recovery probability prediction sub-models are combined to obtain an input voltage automatic recovery probability prediction model, and all output voltage automatic recovery probability prediction sub-models are combined to obtain an output voltage automatic recovery probability prediction model.

[0025] Further, the S8 further includes the following steps:

[0026] The newly obtained input voltage curve and output voltage curve are respectively input into the input voltage automatic recovery probability prediction model and the output voltage automatic recovery probability prediction model;

[0027] The input voltage automatic recovery probability prediction model matches the corresponding input voltage abnormality category curve and the input voltage automatic recovery probability prediction sub-model, and obtains the corresponding input voltage judgment curve;

[0028] The obtained input voltage judgment curve is input into the matched input voltage automatic recovery probability prediction sub-model to obtain the probability of the input voltage recovery in the second setting time;

[0029] The output voltage automatic recovery probability prediction model matches the corresponding output voltage abnormality category curve and the output voltage automatic recovery probability prediction sub-model, and obtains the corresponding output voltage judgment curve;

[0030] The obtained output voltage judgment curve is input into the matched output voltage automatic recovery probability prediction sub-model to obtain the probability of the output voltage recovery in the second setting time.

[0031] Further, the method further includes the following steps:

[0032] The input current and the output current of the transformer are respectively obtained according to the set frequency to obtain the input current curve and the output current curve;

[0033] The input current curve corresponding to the input voltage feature curve time is obtained to obtain the input current feature curve, and the output current curve corresponding to the output voltage feature curve time is obtained to obtain the output current feature curve, that is, the feature change curve of the input current at the same time of each input voltage feature curve and the feature change curve of the output current at the same time of each output voltage feature curve are obtained.

[0034] The historical input end current characteristic curve and the historical output end current characteristic curve are respectively set with input end current deviation threshold and output end current deviation threshold;

[0035] Based on the corresponding data of the historical input end current characteristic curve and the historical output end current characteristic curve, regression analysis is performed to obtain multiple input end current abnormal category curves with errors less than the input end current deviation threshold and multiple output end current abnormal category curves with errors less than the output end current deviation threshold.

[0036] Further, the S8 further includes the following steps:

[0037] Based on the errors of the input end current curve and the output end current curve, the input end current abnormal category curve and the output end current abnormal category curve are matched;

[0038] The input end current characteristic curve of the input end current curve and the output end current characteristic curve of the output end current curve are extracted;

[0039] The average errors of the input end current characteristic curve and each input end current abnormal category curve and the average errors of the output end current characteristic curve and each output end current abnormal category curve are calculated;

[0040] The input end current abnormal category curve with the minimum average error is matched with the input end current curve, and the output end current abnormal category curve with the minimum average error is matched with the output end current curve;

[0041] The input end current curve after the input end current characteristic curve is extracted from the input end current curve to obtain an input end current judgment curve, and the output end current curve after the output end current characteristic curve is extracted from the output end current curve to obtain an output end current judgment curve.

[0042] Further, the S8 further includes the following steps:

[0043] According to the differences between the input end voltage abnormal category curve and the input end current abnormal category curve, the Bayesian time series analysis model is trained based on the input end voltage judgment curve, the corresponding external influencing factor, the input end current judgment curve, and the recovery situation of the input end voltage in the second set time period to obtain an input end voltage automatic recovery probability prediction sub-model, which is used to predict the probability of the recovery of the input end voltage in the second set time period based on the input input end voltage judgment curve, the external influencing factor, and the output input end current judgment curve.

[0044] According to the difference between the input end voltage abnormality category curve and the output end current abnormality category curve, the output end voltage judgment curve and the corresponding external influence factor, the output end current judgment curve, and the recovery condition of the output end voltage in the second setting time period are used to train the Bayesian time series analysis model to obtain an output end voltage automatic recovery probability prediction sub-model, which is used to output the probability of the output end voltage recovery in the second setting time period based on the input output end voltage judgment curve, the external influence factor, and the output end current judgment curve.

[0045] All input end voltage automatic recovery probability prediction sub-models are combined to obtain an input end voltage automatic recovery probability prediction model, and all output end voltage automatic recovery probability prediction sub-models are combined to obtain an output end voltage automatic recovery probability prediction model.

[0046] Further, the S8 further includes the following steps:

[0047] The newly obtained input end voltage curve and input end current curve are input into the input end voltage automatic recovery probability prediction model.

[0048] The input end voltage automatic recovery probability prediction model matches the corresponding input end voltage abnormality category curve, current abnormality category curve, and input end voltage automatic recovery probability prediction sub-model, and obtains the corresponding input end voltage judgment curve and input end current judgment curve.

[0049] The obtained input end voltage judgment curve and input end current judgment curve are input into the matched input end voltage automatic recovery probability prediction sub-model to obtain the probability of the input end voltage recovery in the second setting time period.

[0050] The newly obtained output end voltage curve and output end current curve are input into the output end voltage automatic recovery probability prediction model.

[0051] The output end voltage automatic recovery probability prediction model matches the corresponding output end voltage abnormality category curve, current abnormality category curve, and output end voltage automatic recovery probability prediction sub-model, and obtains the corresponding output end voltage judgment curve and output end current judgment curve.

[0052] The obtained output end voltage judgment curve and output end current judgment curve are output to the matched output end voltage automatic recovery probability prediction sub-model to obtain the probability of the output end voltage recovery in the second setting time period.

[0053] A transformer includes a control module and a load regulating switch, wherein the control module is connected to the load regulating switch.

[0054] The control module is used to execute a transformer voltage regulation method.

[0055] The load regulating switch is used for regulating the output voltage of the transformer in response to the voltage regulating signal output by the control module.

[0056] 1. Compared with the prior art, the transformer voltage regulating method and transformer provided by the application analyze the fluctuation characteristics of the voltage, judge the similarity of the newly obtained voltage fluctuation characteristics and the historical voltage fluctuation characteristics through error, thereby facilitating the judgment of whether the voltage can automatically recover within a specified time through the historical voltage fluctuation characteristics, and the judgment of whether the voltage needs to be regulated, and improving the analysis accuracy through the analysis of the voltage fluctuation of the input end and the output end of the transformer respectively.

[0057] 2. Compared with the prior art, the transformer voltage regulating method and transformer provided by the application analyze the probability of the automatic recovery of the voltage within a specified time under the condition of each voltage fluctuation characteristic, the corresponding time, other electrical parameters and environmental data, and set a probability threshold, so that the voltage is judged to be able to automatically recover within a specified time without regulating the voltage of the transformer only when the prediction probability is greater than the probability threshold, thereby further improving the prediction accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art according to these drawings.

[0059] Fig. 1 The method step diagram provided for the embodiment of the present application;

[0060] Fig. 2 The S8 step diagram provided for one embodiment of the present application;

[0061] Fig. 3 The S8 step diagram provided for another embodiment of the present application. DETAILED DESCRIPTION

[0062] In order to make those skilled in the art better understand the technical solutions of the present application, the present application will be further described in detail with reference to the drawings.

[0063] In the description of the application, it should be understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise" and the like indicate the orientation or positional relationship shown in the drawings based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the application.

[0064] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined as "first", "second" can explicitly or implicitly include one or more of the features. In the description of the application, the meaning of "multiple" is two or more, unless otherwise specifically limited. In addition, the terms "mounting", "connecting", "connecting" should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication between two elements. For those skilled in the art, the specific meaning of the above terms in the application can be understood according to the specific circumstances.

[0065] In the following, example embodiments will be described more fully with reference to the accompanying drawings, but the example embodiments can be embodied in different forms and should not be interpreted as being limited to the embodiments set forth herein. Rather, the purpose of providing these embodiments is to make the disclosure thorough and complete, and to enable those skilled in the art to fully understand the scope of the disclosure.

[0066] The embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0067] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0068] The terms used herein are only used to describe specific embodiments, and are not intended to limit the disclosure. As used herein, the singular forms "a" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. It will also be understood that when the terms "comprise" and / or "consist of" are used in the specification, the specified features, integers, steps, operations, elements, and / or components are present, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0069] The embodiments described herein can be described with reference to plan views and / or cross-sectional views by virtue of the idealized illustrative representations of the underlying components of the present disclosure. Accordingly, the example illustrations are merely meant to be illustrative and other example illustrations can be utilized based on manufacturing techniques and / or tolerances. Thus, the embodiments are not limited to the embodiments illustrated in the figures, but include modifications that can be made to the configurations based on manufacturing processes and / or tolerances. Therefore, the regions illustrated in the figures are schematic only and the shapes of the regions illustrated in the figures do not necessarily illustrate the precise shape of the regions in a device made by the manufacturing process and can be simplified. The same reference numerals can be used in different illustrations to indicate the same or similar elements.

[0070] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and the present disclosure and will not be interpreted in an overly literal or overly formal sense unless expressly so defined herein.

[0071] See Figs. 1-3 A transformer voltage regulation method, comprising the following steps:

[0072] S1, respectively, according to the set frequency, the input end voltage and the output end voltage of the transformer are obtained, and the access end voltage curve and the output end voltage curve are obtained, wherein the input end refers to one end of the transformer connected with the power grid before voltage transformation, and the output end refers to one end of the transformer connected with the power grid after voltage transformation;

[0073] S2, set the input end normal voltage interval and the output end normal voltage interval;

[0074] S3, based on the input end normal voltage interval and the input end voltage curve, the input end abnormal voltage curve is obtained, and based on the output end normal voltage interval and the output end voltage curve, the output end abnormal voltage curve is obtained, wherein the input end abnormal voltage curve is the part of the input end voltage curve exceeding the input end normal voltage interval, and the output end abnormal voltage curve is the part of the output end voltage curve exceeding the output end normal voltage interval;

[0075] S4, the input end abnormal voltage curve and the curve before the first set time interval are obtained, and the input end voltage characteristic curve is obtained, and the output end abnormal voltage curve and the curve before the first set time interval are obtained, and the output end voltage characteristic curve is obtained;

[0076] S5, respectively, for the historical input end voltage characteristic curve and the historical output end voltage characteristic curve, set the input end voltage deviation threshold and the output end voltage deviation threshold;

[0077] S6, based on the historical input voltage characteristic curve and the historical output voltage characteristic curve corresponding data, regression analysis is carried out, respectively, to obtain a plurality of input voltage abnormal category curves with error less than the input voltage deviation threshold, and a plurality of output voltage abnormal category curves with error less than the output voltage deviation threshold; That is, the error of each input voltage abnormal category curve and the corresponding input voltage characteristic curve is within the set deviation threshold, and the error of each output voltage abnormal category curve and the corresponding output voltage characteristic curve is within the set deviation threshold;

[0078] S7, based on the need to set the corresponding sensor, the transformer and its input side and output side are monitored to obtain corresponding electrical data, environmental data, and external influencing factors, wherein the sensor monitoring electrical data includes but is not limited to monitoring high-frequency pulse current through high-frequency current sensor or coupling capacitor; Monitor high-frequency electromagnetic waves through radio frequency antenna or receiver; Monitor the accumulated charge on the surface of the cable or equipment through the charge coupled sensor, etc. To determine whether the power grid line is faulty; Wherein, the environmental data includes temperature, humidity, air dust content, etc.;

[0079] S8, analyze the newly obtained input voltage curve and output voltage curve, the corresponding input voltage abnormal category curve and output voltage abnormal category curve, and analyze the probability of input voltage or output voltage returning to normal under the corresponding external factor condition of input voltage abnormal category curve and output voltage abnormal category curve, including the following steps:

[0080] S8.1a, based on the error of the input voltage curve and the output voltage curve, match the input voltage abnormal category curve and the output voltage abnormal category curve, which includes the following steps:

[0081] S8.1a.1, extract the input voltage characteristic curve of the input voltage curve, and the output voltage characteristic curve of the output voltage curve;

[0082] S8.1a.2, calculate the average error of the input voltage characteristic curve and each input voltage abnormal category curve, and the average error of the output voltage characteristic curve and each output voltage abnormal category curve;

[0083] S8.1a.3, respectively, the input voltage abnormal category curve with the minimum average error is matched with the input voltage curve, and the output voltage abnormal category curve with the minimum average error is matched with the output voltage curve;

[0084] S8.2a, after extracting the input voltage feature curve from the input voltage curve, the input voltage judgment curve is obtained, and after extracting the output voltage feature curve from the output voltage curve, the output voltage judgment curve is obtained;

[0085] S8.3a, according to the different input voltage abnormality category curves, based on the input voltage judgment curve and the corresponding external influence factor, the recovery of the input voltage in the second setting time is trained to obtain the input voltage automatic recovery probability prediction sub-model, which is used to predict the probability of the input voltage recovery in the second setting time based on the input voltage judgment curve and the external influence factor; wherein the Bayesian time series analysis model needs to be based on historical data and newly acquired data during the training process, and the prior probability, observation probability, transition probability and posterior probability are used for analysis and calculation updating to improve the accuracy of probability prediction.

[0086] S8.4a, according to the different input voltage abnormality category curves, based on the output voltage judgment curve and the corresponding external influence factor, the recovery of the output voltage in the second setting time is trained to obtain the output voltage automatic recovery probability prediction sub-model, which is used to predict the probability of the output voltage recovery in the second setting time based on the input voltage judgment curve and the external influence factor.

[0087] S8.5a, all input voltage automatic recovery probability prediction sub-models are combined to obtain the input voltage automatic recovery probability prediction model, and all output voltage automatic recovery probability prediction sub-models are combined to obtain the output voltage automatic recovery probability prediction model.

[0088] S8.6a, the newly acquired input voltage curve and the output voltage curve are respectively input into the input voltage automatic recovery probability prediction model and the output voltage automatic recovery probability prediction model;

[0089] S8.7a, the input voltage automatic recovery probability prediction model matches the corresponding input voltage abnormality category curve and the input voltage automatic recovery probability prediction sub-model through S8.1a, and obtains the corresponding input voltage judgment curve;

[0090] S8.8a, the obtained input voltage judgment curve is input into the matched input voltage automatic recovery probability prediction sub-model to obtain the probability of the input voltage recovery in the second setting time;

[0091] S8.9a, the output voltage automatic recovery probability prediction model matches the corresponding output voltage abnormality category curve and the output voltage automatic recovery probability prediction sub-model, and obtains the corresponding output voltage judgment curve;

[0092] S8.10a. Input the obtained output terminal voltage judgment curve into the matching output terminal voltage automatic recovery probability prediction sub-model to obtain the probability of output terminal voltage recovery within the second set time period.

[0093] S9. Set a judgment threshold. If the probability that the input voltage or the output voltage returns to normal within the second set time period is greater than the judgment threshold, there is no need to adjust the voltage of the transformer. Otherwise, adjust the voltage of the transformer.

[0094] In one embodiment, before step S8, the method further includes the following steps:

[0095] A1. Obtain the input current and output current of the transformer according to the set frequency, and obtain the input current curve and the output current curve.

[0096] A2. Obtain an input-end current curve corresponding to the time of the input-end voltage characteristic curve to obtain an input-end current characteristic curve, obtain an output-end current curve corresponding to the time of the output-end voltage characteristic curve to obtain an output-end current characteristic curve, that is, obtain characteristic change curves of the input-end current at the same time for each input-end voltage characteristic curve, and characteristic change curves of the output-end current at the same time for each output-end voltage characteristic curve.

[0097] A3. Set an input-end current deviation threshold and an output-end current deviation threshold for the historical input-end current characteristic curve and the historical output-end current characteristic curve, respectively.

[0098] A4. Based on the corresponding data of the historical input-end current characteristic curve and the historical output-end current characteristic curve, regression analysis is performed to obtain multiple input-end current anomaly category curves whose errors are less than the input-end current deviation threshold, and multiple output-end current anomaly category curves whose errors are less than the output-end current deviation threshold.

[0099] In this embodiment, S8 further includes the following steps:

[0100] S8.1b. Matching the input-end voltage abnormality category curve and the output-end voltage abnormality category curve based on the error between the input-end voltage curve and the output-end voltage curve, and matching the input-end current abnormality category curve and the output-end current abnormality category curve based on the error between the input-end current curve and the output-end current curve, specifically comprising the following steps:

[0101] S8.1b.1, extract the input voltage feature curve of the input voltage curve and the output voltage feature curve of the output voltage curve; calculate the average error of the input voltage feature curve and each input voltage abnormality category curve, and the average error of the output voltage feature curve and each output voltage abnormality category curve; extract the input current feature curve of the input current curve and the output current feature curve of the output current curve.

[0102] S8.1b.2, calculate the average error of the input current feature curve and each input current abnormality category curve, and the average error of the output current feature curve and each output current abnormality category curve;

[0103] S8.1b.3, respectively, the input voltage abnormality category curve with the minimum average error is matched with the input voltage curve, and the output voltage abnormality category curve with the minimum average error is matched with the output voltage curve; respectively, the input current abnormality category curve with the minimum average error is matched with the input current curve, and the output current abnormality category curve with the minimum average error is matched with the output current curve.

[0104] S8.2b, after extracting the input voltage feature curve from the input voltage curve, the input voltage judgment curve is obtained, and after extracting the output voltage feature curve from the output voltage curve, the output voltage judgment curve is obtained; after extracting the input current feature curve from the input current curve, the input current judgment curve is obtained, and after extracting the output current feature curve from the output current curve, the output current judgment curve is obtained.

[0105] S8.3b, according to the difference between the input voltage abnormality category curve and the input current abnormality category curve, respectively based on the input voltage judgment curve and its corresponding external influence factor, the input current judgment curve, the recovery of the input voltage in the second setting time, the Bayesian time series analysis model is trained, the input voltage automatic recovery probability prediction sub-model is obtained, which is used for predicting the probability of the input voltage recovery in the second setting time based on the input input voltage judgment curve, external influence factor and input current judgment curve;

[0106] S8.4b, according to the difference between the input end voltage abnormality category curve and the output end current abnormality category curve, respectively based on the output end voltage judgment curve, the corresponding external influence factor, the recovery condition of the output end voltage in the second setting time, training the Bayesian time series analysis model, obtaining the output end voltage automatic recovery probability prediction sub-model, for outputting the probability of the output end voltage recovery in the second setting time based on the input output end voltage judgment curve, external influence factor and output end current judgment curve;

[0107] S8.5b, combining all the input end voltage automatic recovery probability prediction sub-models to obtain the input end voltage automatic recovery probability prediction model, and combining all the output end voltage automatic recovery probability prediction sub-models to obtain the output end voltage automatic recovery probability prediction model.

[0108] S8.6b, inputting the newly acquired input end voltage curve and input end current curve into the input end voltage automatic recovery probability prediction model.

[0109] S8.7b, the input end voltage automatic recovery probability prediction model matches the corresponding input end voltage abnormality category curve, current abnormality category curve and input end voltage automatic recovery probability prediction sub-model through S8.1b, and obtains the corresponding input end voltage judgment curve and input end current judgment curve.

[0110] S8.8b, inputting the obtained input end voltage judgment curve and input end current judgment curve into the matched input end voltage automatic recovery probability prediction sub-model to obtain the probability of the input end voltage recovery in the second setting time.

[0111] S8.9b, inputting the newly acquired output end voltage curve and output end current curve into the output end voltage automatic recovery probability prediction model.

[0112] S8.10b, the output end voltage automatic recovery probability prediction model matches the corresponding output end voltage abnormality category curve, current abnormality category curve and output end voltage automatic recovery probability prediction sub-model, and obtains the corresponding output end voltage judgment curve and output end current judgment curve.

[0113] S8.11b, inputting the obtained output end voltage judgment curve and output end current judgment curve into the matched output end voltage automatic recovery probability prediction sub-model to obtain the probability of the output end voltage recovery in the second setting time.

[0114] A transformer comprising a control module and a load regulating switch, the control module being connected to the load regulating switch.

[0115] The control module is used for the transformer voltage regulation method provided by the application.

[0116] The load regulating switch is used to regulate the output voltage of the transformer in response to a voltage regulating signal output by the control module.

[0117] Certain exemplary embodiments of the application have been described herein in an illustrative manner, for purposes of comprehension of the inventive faculty herein. No limitation on the scope of the application is intended by any of the details of the description, unless specifically indicated otherwise.

Claims

1. A transformer voltage regulation method, characterized in that: The following steps are involved: S1. Obtain the input voltage and output voltage of the transformer according to the set frequency, and obtain the input voltage curve and the output voltage curve; S2. Setting the normal voltage range of the input end and the normal voltage range of the output end; S3. Obtain an abnormal input voltage curve based on the normal input voltage range and the input voltage curve, and obtain an abnormal output voltage curve based on the normal output voltage range and the output voltage curve; S4. Obtain an abnormal input voltage curve and a curve thereof before a first set time length to obtain an input voltage characteristic curve; obtain an abnormal output voltage curve and a curve thereof before a first set time length to obtain an output voltage characteristic curve; S5. Setting an input-end voltage deviation threshold and an output-end voltage deviation threshold based on a historical input-end voltage characteristic curve and a historical output-end voltage characteristic curve, respectively; S6. Perform regression analysis based on the corresponding data of the historical input-end voltage characteristic curve and the historical output-end voltage characteristic curve, respectively, to obtain a plurality of input-end voltage anomaly category curves whose errors are less than the input-end voltage deviation threshold and a plurality of output-end voltage anomaly category curves whose errors are less than the output-end voltage deviation threshold; S7. Based on monitoring needs, corresponding sensors are set to monitor the transformer and its input and output ends to obtain corresponding electrical data and environmental data, and to obtain external influencing factors; S8. Analyze the newly acquired input-end voltage curve and output-end voltage curve, and the corresponding input-end voltage abnormality category curve and output-end voltage abnormality category curve, and analyze the probability that the input-end voltage or the output-end voltage returns to normal under the corresponding external factors of the input-end voltage abnormality category curve and the output-end voltage abnormality category curve; The S8 comprises the following steps: Matching the input-end voltage abnormality category curve and the output-end voltage abnormality category curve based on the error; Extracting an input-end voltage curve after the input-end voltage characteristic curve from the input-end voltage curve to obtain an input-end voltage judgment curve, and extracting an output-end voltage curve after the output-end voltage characteristic curve from the output-end voltage curve to obtain an output-end voltage judgment curve; According to different input terminal voltage anomaly category curves, based on the input terminal voltage judgment curve and its corresponding external influencing factors, and the recovery of the input terminal voltage within a second set time period, a Bayesian time series analysis model is trained to obtain an input terminal voltage automatic recovery probability prediction sub-model, which is used to predict the probability of input terminal voltage recovery within the second set time period based on the input terminal voltage judgment curve and the external influencing factors; According to different output terminal voltage anomaly category curves, based on the output terminal voltage judgment curve and its corresponding external influencing factors, and the recovery of the output terminal voltage within a second set time period, a Bayesian time series analysis model is trained to obtain an output terminal voltage automatic recovery probability prediction sub-model, which is used to predict the probability of output terminal voltage recovery within the second set time period based on the input output terminal voltage judgment curve and the external influencing factors; Combining all input-end voltage automatic recovery probability prediction sub-models to obtain an input-end voltage automatic recovery probability prediction model, and combining all output-end voltage automatic recovery probability prediction sub-models to obtain an output-end voltage automatic recovery probability prediction model; S9. Set a judgment threshold. If the probability that the input voltage or the output voltage returns to normal within the second set time period is greater than the judgment threshold, there is no need to adjust the voltage of the transformer. Otherwise, adjust the voltage of the transformer.

2. A transformer voltage regulation method according to claim 1, characterized in that: The S8 further comprises the following steps: Inputting the newly acquired input-end voltage curve and output-end voltage curve into the input-end voltage automatic recovery probability prediction model and the output-end voltage automatic recovery probability prediction model respectively; The input terminal voltage automatic recovery probability prediction model matches the corresponding input terminal voltage abnormality category curve and the input terminal voltage automatic recovery probability prediction sub-model, and obtains the corresponding input terminal voltage judgment curve; The obtained input terminal voltage judgment curve is input into the matching input terminal voltage automatic recovery probability prediction sub-model to obtain the probability of input terminal voltage recovery within a second set time period; The output terminal voltage automatic recovery probability prediction model matches the corresponding output terminal voltage abnormality category curve and the output terminal voltage automatic recovery probability prediction sub-model, and obtains the corresponding output terminal voltage judgment curve; The obtained output terminal voltage judgment curve is input into the matching output terminal voltage automatic recovery probability prediction sub-model to obtain the probability of output terminal voltage recovery within the second set time length.

3. A transformer voltage regulation method according to claim 2, characterized in that: Before step S8, the following steps are also included: The input current and output current of the transformer are obtained according to the set frequency respectively, and the input current curve and the output current curve are obtained; Obtaining an input-end current curve corresponding to the input-end voltage characteristic curve time to obtain an input-end current characteristic curve, obtaining an output-end current curve corresponding to the output-end voltage characteristic curve time to obtain an output-end current characteristic curve; Set the input-end current deviation threshold and the output-end current deviation threshold for the historical input-end current characteristic curve and the historical output-end current characteristic curve respectively; Regression analysis is performed based on the corresponding data of historical input current characteristic curves and historical output current characteristic curves, respectively, to obtain multiple input current anomaly category curves with errors less than the input current deviation threshold and multiple output current anomaly category curves with errors less than the output current deviation threshold.

4. A transformer voltage regulation method according to claim 3, characterized in that: The S8 further comprises the following steps: Matching the input-end current abnormality category curve and the output-end current abnormality category curve based on the error; The input current curve after the input current characteristic curve is extracted from the input current curve to obtain the input current judgment curve, and the output current curve after the output current characteristic curve is extracted from the output current curve to obtain the output current judgment curve.

5. A transformer voltage regulation method according to claim 4, characterized in that: The S8 further comprises the following steps: According to the difference between the input terminal voltage abnormality classification curve and the input terminal current abnormality classification curve, the Bayesian time series analysis model is trained based on the input terminal voltage judgment curve and its corresponding external influencing factors, the input terminal current judgment curve, and the recovery of the input terminal voltage within the second set time period, to obtain an input terminal voltage automatic recovery probability prediction sub-model, which is used to output the probability of input terminal voltage recovery within the second set time period based on the input input terminal voltage judgment curve, the external influencing factor prediction, and the input terminal current judgment curve; According to the difference between the input-end voltage abnormality classification curve and the output-end current abnormality classification curve, the Bayesian time series analysis model is trained based on the output-end voltage judgment curve and its corresponding external influencing factors, the output-end current judgment curve, and the recovery of the output-end voltage within the second set time period, to obtain an output-end voltage automatic recovery probability prediction sub-model, which is used to output the probability of output-end voltage recovery within the second set time period based on the input output-end voltage judgment curve, the external influencing factors, and the output-end current judgment curve; All input-end voltage automatic recovery probability prediction sub-models are combined to obtain an input-end voltage automatic recovery probability prediction model, and all output-end voltage automatic recovery probability prediction sub-models are combined to obtain an output-end voltage automatic recovery probability prediction model.

6. A transformer voltage regulation method according to claim 5, characterized in that: The S8 further comprises the following steps: Inputting the newly acquired input terminal voltage curve and input terminal current curve into the input terminal voltage automatic recovery probability prediction model; The input terminal voltage automatic recovery probability prediction model matches the corresponding input terminal voltage abnormality category curve, current abnormality category curve and input terminal voltage automatic recovery probability prediction sub-model, and obtains the corresponding input terminal voltage judgment curve and input terminal current judgment curve; The obtained input terminal voltage judgment curve and input terminal current judgment curve are input into the matching input terminal voltage automatic recovery probability prediction sub-model to obtain the probability of input terminal voltage recovery within a second set time length; Input the newly acquired output terminal voltage curve and output terminal current curve into the output terminal voltage automatic recovery probability prediction model; The output terminal voltage automatic recovery probability prediction model matches the corresponding output terminal voltage abnormality category curve, current abnormality category curve and output terminal voltage automatic recovery probability prediction sub-model, and obtains the corresponding output terminal voltage judgment curve and output terminal current judgment curve; The obtained output terminal voltage judgment curve and the output terminal current judgment curve are output to the output terminal voltage automatic recovery probability prediction sub-model that matches the output terminal voltage judgment curve to obtain the probability of output terminal voltage recovery within the second set time length.

7. A transformer, characterized in that: It includes a control module and a load adjustment switch, wherein the control module is connected to the load adjustment switch; The control module is used to execute a transformer voltage regulation method according to any one of claims 1 to 6; The load adjustment switch is used to adjust the output voltage of the transformer in response to the voltage adjustment signal output by the control module.

Images