Station automation terminal with efficient power data collection function

Abstract

The application relates to the technical field of data collection, in particular to a station automation terminal with efficient electric energy data collection function; a very abnormal degree is obtained according to the neighborhood fluctuation characteristics of data points in an electric energy monitoring time sequence; a very abnormal point is obtained according to the very abnormal degree; a fluctuation stability value of the very abnormal point is obtained according to the interval characteristics and data difference characteristics of the very abnormal point and other adjacent very abnormal points; a weight coefficient of the very abnormal point in an ARIMA model is adjusted according to the fluctuation stability value, and an adaptive ARIMA model is obtained; the electric energy monitoring time sequence is predicted according to the adaptive ARIMA model, and an electric energy prediction time sequence is obtained. According to the difference characteristics of the very abnormal points between the electric energy prediction time sequence and the electric energy monitoring time sequence, the preset collection frequency is adjusted, the adaptive collection frequency is obtained, and the electric energy data after the adjustment point is collected, so that the efficiency and accuracy of electric energy data collection are improved.

Description

Technical Field

[0001] This invention relates to the field of data acquisition technology, and specifically to a station automation terminal with efficient power data acquisition function. Background Technology

[0002] Station automation terminals are typically installed in switching stations, ring main units, and small substations to collect and calculate data such as position signals, voltage, current, power, power factor, and energy of switching equipment. These terminals monitor system faults, control switching operations, and record and store operational data. However, current station automation terminals use a fixed data collection frequency, which cannot be dynamically adjusted based on load fluctuations or equipment status. Setting a high collection frequency can lead to high redundancy and low data storage space utilization, while setting a low frequency may result in low data acquisition accuracy and loss of information from critical nodes, ultimately affecting the efficiency and accuracy of power data acquisition. Summary of the Invention

[0003] To address the technical problem that a fixed acquisition frequency can easily lead to high redundancy or low data accuracy, affecting the efficiency and accuracy of power data acquisition, the present invention aims to provide a station automation terminal with efficient power data acquisition capabilities. The specific technical solution adopted is as follows:

[0004] The data acquisition module is used to acquire the power monitoring time sequence before the adjustment point;

[0005] The data analysis module is used to obtain the degree of anomalousness of the data points based on the neighborhood fluctuation characteristics of the data points in the power monitoring time series; to obtain anomalous points based on the degree of anomalousness; and to obtain the fluctuation stability value of the anomalous points based on the interval characteristics and data difference characteristics between the anomalous points and other adjacent anomalous points.

[0006] The prediction and adjustment module is used to adjust the weight coefficients of the non-normal points in the ARIMA model according to the fluctuation stability value to obtain an adaptive ARIMA model; and to predict the power monitoring time series according to the adaptive ARIMA model to obtain the power prediction time series.

[0007] An adaptive acquisition module is used to adjust a preset acquisition frequency based on the difference characteristics of abnormal points between the power prediction time series and the power monitoring time series to obtain an adaptive acquisition frequency; and to acquire power data after the adjustment point based on the adaptive acquisition frequency.

[0008] Furthermore, the step of obtaining the degree of anomalousness of the data point based on the neighborhood fluctuation characteristics of the data point in the power monitoring time series includes:

[0009] Construct a Cartesian coordinate system for the power monitoring time series, with time on the horizontal axis and power monitoring values ​​on the vertical axis; calculate and normalize the sum of the Euclidean distances between the data point and the two adjacent data points to obtain discrete feature values; calculate and normalize the sum of the absolute values ​​of the slopes of the lines connecting the data point and the two adjacent data points to obtain variation feature values; calculate the average of the discrete feature values ​​and the variation feature values ​​to obtain the degree of anomalousness of the data point.

[0010] Furthermore, the step of obtaining the abnormal point based on the degree of abnormality includes:

[0011] Data points whose abnormality level exceeds a preset threshold are defined as abnormal points.

[0012] Further, the step of obtaining the fluctuation stability value of the abnormal point based on the interval characteristics and data difference characteristics between the abnormal point and other adjacent abnormal points includes:

[0013] Calculate the reciprocal of the sum of the intervals between the abnormal point and the two adjacent abnormal points to obtain the nearest neighbor feature value; calculate the reciprocal of the absolute value of the difference between the abnormal point and the two adjacent abnormal points to obtain the regularity feature value; calculate the product of the nearest neighbor feature value and the regularity feature value and normalize it to obtain a first value; calculate the absolute value of the difference between the abnormal point and the previous adjacent abnormal point to obtain a first difference value; calculate the absolute value of the difference between the abnormal point and the next adjacent abnormal point to obtain a second difference value; calculate the reciprocal of the sum of the first difference and the second difference and normalize it to obtain a second value; calculate the average of the first value and the second value to obtain the fluctuation stability value of the abnormal point.

[0014] Further, the step of adjusting the weight coefficients of the non-normal points in the ARIMA model according to the fluctuation stability value to obtain the adaptive ARIMA model includes:

[0015] Calculate the product of the preset autoregressive coefficient of the non-normal point in the ARIMA model and the stable fluctuation value to obtain the adaptive autoregressive coefficient; calculate the product of the preset moving average coefficient of the non-normal point in the ARIMA model and the stable fluctuation value to obtain the adaptive moving average coefficient; obtain the adaptive ARIMA model based on the adaptive autoregressive coefficient and the adaptive moving average coefficient.

[0016] Further, the step of adjusting the preset acquisition frequency based on the difference characteristics of abnormal points between the power prediction time series and the power monitoring time series to obtain an adaptive acquisition frequency includes:

[0017] Calculate the ratio of the number of abnormal points between the power prediction time series and the power monitoring time series to obtain the adjustment coefficient; calculate the product of the adjustment coefficient and the acquisition frequency of the power monitoring time series to obtain the adaptive acquisition frequency.

[0018] Furthermore, the step of collecting electrical energy data after the adjustment point according to the adaptive acquisition frequency includes:

[0019] The power data after the adjustment point is collected according to the adaptive acquisition frequency until the time of the next adjustment point is stopped; a new adaptive acquisition frequency is obtained according to the new power monitoring timing before the next adjustment point, and the power data after the next adjustment point is adaptively collected.

[0020] The present invention has the following beneficial effects:

[0021] In this invention, since monitoring abnormal conditions such as load fluctuations is more important in circuit monitoring, obtaining the degree of abnormality can characterize the fluctuation characteristics of data points in the power monitoring time series; obtaining abnormal points can show the distribution of abnormal fluctuation characteristics in the power monitoring time series. Since abnormal fluctuation characteristics are not only caused by circuit load fluctuations, noise data can also lead to abnormal points, obtaining the fluctuation stability value can characterize the possibility that abnormal points are caused by noise data. Obtaining the adaptive ARIMA model can improve the accuracy of power prediction time series, reduce the impact of noise data on prediction results, and make the calculation results of the adaptive acquisition frequency more accurate. Obtaining the power prediction time series can predict the power data change characteristics in future periods based on the power monitoring time series, and thus determine the adaptive acquisition frequency based on the difference in the number of abnormal points between the power prediction time series and the power monitoring time series. Obtaining the adaptive acquisition frequency can select a suitable acquisition frequency based on the changing trend of load fluctuations, thereby improving the accuracy of acquisition frequency calculation. By acquiring power data after the adjustment point according to the adaptive acquisition frequency, the efficiency and accuracy of power data acquisition are ultimately improved. Attached Figure Description

[0022] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a block diagram of a station automation terminal with efficient power data acquisition function provided in one embodiment of the present invention. Detailed Implementation

[0024] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a station automation terminal with efficient power data acquisition function proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0026] The following description, in conjunction with the accompanying drawings, details a specific solution for a station automation terminal with efficient power data acquisition function provided by the present invention.

[0027] Please see Figure 1 The diagram illustrates a station automation terminal with efficient power data acquisition function according to an embodiment of the present invention. The terminal includes the following modules:

[0028] The data acquisition module S1 is used to acquire the power monitoring time sequence before the adjustment point.

[0029] In this embodiment of the invention, the implementation scenario is to adaptively collect power data through the station automation terminal to improve the efficiency and accuracy of power data collection. First, the power monitoring time sequence before the adjustment point is obtained. In this embodiment of the invention, the preset collection frequency for the first collection is 50Hz, and the collection range is the data within 10 seconds before the adjustment point. The monitoring objects include current, voltage and power. The power monitoring time sequence of each monitoring object before the adjustment point is obtained respectively.

[0030] The data analysis module S2 is used to obtain the degree of anomalousness of data points based on the neighborhood fluctuation characteristics of data points in the power monitoring time series; to obtain anomalous points based on the degree of anomalousness; and to obtain the fluctuation stability value of anomalous points based on the interval characteristics and data difference characteristics between anomalous points and other adjacent anomalous points.

[0031] Under normal circumstances, electrical energy monitoring data exhibits slight fluctuations during production and transmission. These slight fluctuations are considered normal fluctuations, characterized by small amplitude and high repetition. Using a higher acquisition frequency would lead to high redundancy, affecting storage space utilization. When the load in a circuit changes frequently and significantly, the monitoring data will show abnormal fluctuations. These abnormal fluctuations usually reflect key information such as load anomalies and changes in the circuit. Using a lower acquisition frequency would result in less data, making it difficult to fully characterize the key circuit information during monitoring. Therefore, during monitoring, normal fluctuation data that does not reflect key information can be collected at a lower acquisition frequency to improve storage space utilization and reduce redundancy. Abnormal fluctuation data that reflects key information requires a higher acquisition frequency to improve the completeness of the key information representation and enhance the accuracy of circuit analysis. Electrical energy data exhibits temporal correlation and regularity. Future data can be predicted using historical data from similar time periods, and the acquisition frequency of monitoring data can be adaptively adjusted based on the predicted future data. To improve the accuracy of the adaptive acquisition frequency, the accuracy of the prediction results needs to be improved.

[0032] Furthermore, to improve prediction accuracy and the accuracy of adaptive acquisition frequency, it is first necessary to determine whether there are abnormal fluctuations in the power monitoring sequence. Compared with normal slight fluctuations, abnormal fluctuation data have larger amplitudes and steeper waveforms. Therefore, the degree of abnormality of a data point is obtained based on the neighborhood fluctuation characteristics of the data point in the power monitoring time series. Preferably, in this embodiment of the invention, the step of obtaining the degree of abnormality includes: constructing a rectangular coordinate system for the power monitoring time series, with the horizontal axis representing time and the vertical axis representing the power monitoring value; calculating and normalizing the sum of the Euclidean distances between the data point and two adjacent data points to obtain discrete feature values; the larger the discrete feature value, the larger the amplitude at the data point, and the more likely it is to be an abnormal point; it should be noted that the normalization method in this embodiment of the invention is linear normalization. Calculating and normalizing the sum of the absolute values ​​of the slopes of the lines connecting the data point and two adjacent data points to obtain variation feature values; the larger the variation feature value, the steeper the waveform at the data point, and the more likely it is to be an abnormal point. Calculate the average of discrete eigenvalues ​​and varying eigenvalues ​​to obtain the degree of anomalousness for a data point. A higher degree of anomalousness indicates that the data point is more likely to represent anomalous fluctuations and is therefore more likely to be an anomalous point. Formulas for obtaining the degree of anomalousness include:

[0033]

[0034] In the formula, R represents the degree of non-normality, norm() represents normalization, D1 represents the Euclidean distance between the data point and the previous data point, D2 represents the Euclidean distance between the data point and the next data point, norm(D1+D2) represents the discrete eigenvalue, K1 represents the slope of the line connecting the data point and the previous data point, K2 represents the slope of the line connecting the data point and the next data point, and norm(|K1|+|K2|) represents the variable eigenvalue.

[0035] Furthermore, abnormal points can be obtained based on the degree of abnormality. Specifically, this includes: data points whose degree of abnormality exceeds a preset threshold are considered abnormal points; abnormal points characterize abnormal fluctuations in the power monitoring time series at that moment. In this embodiment of the invention, since the degree of abnormality of data points is obtained through linearly normalized discrete feature values ​​and changing feature values, the degree of abnormality of data points with normal fluctuations tends to 0, and the degree of abnormality of data points with abnormal fluctuations tends to 1. Since these two are located at opposite ends of the value range, the preset threshold is 0.5, which can be determined by the implementer according to the implementation scenario.

[0036] For power monitoring time series, the existing ARIMA autoregressive differential moving average model can be used for prediction. This model mainly fits and predicts based on the distribution of abnormal fluctuations to obtain the possible fluctuations in the power monitoring time series. Therefore, the authenticity of the fluctuations has a significant impact on the accuracy of the prediction results. In the actual monitoring process of power data, monitoring sensors will produce noisy data, which will reduce the accuracy of model predictions. Therefore, further analysis is needed at abnormal points to reduce the impact of noise data on predictions. Normal circuit load fluctuations will last for a period of time, showing a step-like or sloping data change trend over the duration. Therefore, abnormal points caused by real fluctuations will have other abnormal points in the nearby time. However, noise data usually appears instantaneously, only at one or two independent monitoring points, presenting as spikes or isolated jumps. On the other hand, load fluctuation changes are continuous, and the differences between adjacent abnormal points are small. However, noise data has a high degree of randomness, and the numerical differences between adjacent abnormal points are large. Therefore, the stable value of the fluctuation at an abnormal point can be obtained based on the interval characteristics and data difference characteristics between the abnormal point and other adjacent abnormal points.

[0037] Preferably, in this embodiment of the invention, the step of obtaining the fluctuation stability value includes: calculating the reciprocal of the sum of the number of intervals between an abnormal point and two adjacent abnormal points to obtain a nearest neighbor feature value; the number of intervals is the number of data points in the interval. The smaller the number of intervals, the closer the time interval between the abnormal point and the other adjacent abnormal points, the larger the nearest neighbor feature value, the less likely the abnormal point is caused by noise data, and the more likely it is to reflect the true load fluctuation. Calculating the reciprocal of the absolute value of the difference between the number of intervals between an abnormal point and two adjacent abnormal points to obtain a regularity feature value; the more similar the number of intervals between the abnormal point and the other abnormal points on both sides, the larger the regularity feature value, meaning the more obvious the continuous change characteristics of the load fluctuation, the more likely the abnormal point is caused by the true load fluctuation, and the less likely it is caused by noise data. Calculating the product of the nearest neighbor feature value and the regularity feature value and normalizing it to obtain a first value; the larger the first value, the more likely the abnormal point is caused by the true load fluctuation, and the more it can reflect the true load fluctuation characteristics. Calculate the absolute value of the difference between the abnormal point and its preceding adjacent abnormal point to obtain the first difference; calculate the absolute value of the difference between the abnormal point and its following adjacent abnormal point to obtain the second difference; the larger the first and second differences, the more likely the abnormal point is caused by random noise data. Calculate the reciprocal of the sum of the first and second differences and normalize it to obtain the second value; the larger the second value, the smaller the difference between adjacent abnormal points, and the more likely the abnormal point is caused by continuous load fluctuations. Calculate the average of the first and second values ​​to obtain the fluctuation stability value of the abnormal point; the larger the fluctuation stability value, the more likely the abnormal point is caused by real load fluctuations; the smaller the fluctuation stability value, the more likely the abnormal point is caused by noise data. The formula for obtaining the fluctuation stability value includes:

[0038]

[0039] In the formula, W represents the stable fluctuation value, norm() represents normalization, t1 represents the interval between the non-normal point and its preceding adjacent non-normal point, t2 represents the interval between the non-normal point and its following adjacent non-normal point, and a represents a preset minimum positive number that is included in the calculation when the denominator is 0. In this embodiment of the invention, it is 0.01. Represents the nearest neighbor eigenvalues. Represents the characteristic value of the pattern. F1 represents the first value, F2 represents the first difference, and F2 represents the second difference. This indicates the second numerical value.

[0040] The prediction and adjustment module S3 is used to adjust the weight coefficients of non-normal points in the ARIMA model according to the fluctuation stability value to obtain an adaptive ARIMA model; and to predict the power monitoring time series according to the adaptive ARIMA model to obtain the power prediction time series.

[0041] In the process of using the ARIMA model to predict power monitoring data, it is necessary to reduce the impact of noise data on the prediction results and avoid reflecting the fluctuation characteristics of noise data in the prediction results, thereby improving the accuracy of calculating non-normal points in the prediction results. The fluctuation stability value reflects the probability that the non-normal point is noise. The smaller the fluctuation stability value, the more likely the non-normal point is caused by noise data. In the prediction process, the weight of the non-normal point should be smaller, thereby weakening the fluctuation characteristics of noise data. Therefore, the weight coefficients of non-normal points in the ARIMA model are adjusted according to the fluctuation stability value to obtain an adaptive ARIMA model. Preferably, in this embodiment of the invention, the steps of obtaining the adaptive ARIMA model include: calculating the product of the preset autoregressive coefficient of the non-normal point in the ARIMA model and the fluctuation stability value to obtain the adaptive autoregressive coefficient; calculating the product of the preset moving average coefficient of the non-normal point in the ARIMA model and the fluctuation stability value to obtain the adaptive moving average coefficient. It should be noted that the preset autoregressive coefficient and the preset moving average coefficient are the weight coefficients of the non-normal point obtained in the original prediction process. The specific acquisition steps are not described in detail here. The smaller the stable value of the fluctuation at the non-normal point, the smaller the adaptive autoregressive coefficient and the adaptive moving average coefficient, thus reducing the data contribution of the non-normal point in the prediction process. An adaptive ARIMA model is obtained based on the adaptive autoregressive coefficient and the adaptive moving average coefficient. The original coefficients are replaced by the adaptive regression coefficients and adaptive moving average coefficients of all non-normal points to obtain the adaptive ARIMA model. The adjusted model can improve the accuracy of power monitoring data.

[0042] Furthermore, the power monitoring time series can be predicted based on the adaptive ARIMA model to obtain the power prediction time series. It should be noted that the prediction process of this model is existing technology, and the specific steps will not be elaborated here. The power prediction time series can be obtained based on the actual operating data in the power monitoring time series, which improves the prediction accuracy and makes the setting of subsequent acquisition frequency more reasonable.

[0043] The adaptive acquisition module S4 is used to adjust the preset acquisition frequency based on the difference characteristics of abnormal points between the power prediction time series and the power monitoring time series to obtain an adaptive acquisition frequency; and to acquire power data after the adjustment point based on the adaptive acquisition frequency.

[0044] After obtaining the power prediction time series, the changing trend of load fluctuation characteristics can be judged based on the ratio of the number of abnormal points between the power prediction time series and the power monitoring time series. Therefore, the preset acquisition frequency is adjusted according to the difference characteristics of the abnormal points between the power prediction time series and the power monitoring time series to obtain an adaptive acquisition frequency. Preferably, in this embodiment of the invention, the step of obtaining the adaptive acquisition frequency includes: calculating the ratio of the number of abnormal points between the power prediction time series and the power monitoring time series to obtain an adjustment coefficient. When the number of abnormal points in the power prediction time series is greater than that in the power monitoring time series, the larger the adjustment coefficient is, it means that the load fluctuation characteristics have a more serious trend, and the acquisition frequency needs to be increased to ensure the completeness and accuracy of monitoring abnormal characteristics such as load fluctuations. Conversely, when the adjustment coefficient is smaller, it means that the load fluctuation characteristics are weakened, and the acquisition frequency can be reduced to reduce the redundancy of monitoring data and improve the utilization of storage space. The product of the adjustment coefficient and the acquisition frequency of the power monitoring time series is calculated to obtain the adaptive acquisition frequency. It should be noted that the acquisition frequency of the power monitoring time series is the previous adaptive acquisition frequency obtained at the previous adjustment point.

[0045] Furthermore, energy data after the adjustment point can be collected according to an adaptive acquisition frequency. Specifically, this includes: collecting energy data after the adjustment point according to the adaptive acquisition frequency until the next adjustment point; obtaining a new adaptive acquisition frequency based on the new energy monitoring time sequence before the next adjustment point and adaptively acquiring energy data after the next adjustment point. For example, obtaining the adaptive acquisition frequency for the next 10 seconds based on the energy monitoring time sequence 10 seconds before the current adjustment point, acquiring energy data for the next 10 seconds according to the adaptive acquisition frequency, predicting energy data after the next adjustment point based on the acquisition results and obtaining a new adaptive acquisition frequency, and acquiring energy data for the next 10 seconds according to the new adaptive acquisition frequency; then updating the acquisition frequency every 10 seconds based on the changing characteristics of the number of abnormal points, so that the acquisition frequency can be adaptively adjusted based on the changing trend of load fluctuations. When the load fluctuation is relatively stable, the acquisition frequency is reduced to improve storage space utilization; when the load fluctuation is more obvious, the acquisition frequency is increased to improve the accuracy of load fluctuation monitoring. In this embodiment of the invention, the minimum sampling frequency is 20Hz and the maximum sampling frequency is 200Hz to prevent the adaptive sampling frequency from exceeding this limit, which would result in the sampling frequency being too high or too low. The implementer can determine the frequency according to the implementation scenario.

[0046] In summary, this invention provides a station automation terminal with efficient power data acquisition capabilities. It obtains the degree of anomalousness based on the neighborhood fluctuation characteristics of data points in the power monitoring time series; identifies anomalous points based on the degree of anomalousness; obtains the fluctuation stability value of the anomalous points based on the interval characteristics and data difference characteristics between the anomalous points and their neighbors; adjusts the weight coefficients of the anomalous points in the ARIMA model based on the fluctuation stability value to obtain an adaptive ARIMA model; and predicts the power monitoring time series based on the adaptive ARIMA model to obtain a predicted power time series. This invention adjusts the preset acquisition frequency based on the difference characteristics of anomalous points between the predicted power time series and the power monitoring time series to obtain an adaptive acquisition frequency and acquires power data after the adjustment point, thereby improving the efficiency and accuracy of power data acquisition.

[0047] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A station automation terminal with efficient power data acquisition function, characterized in that, The terminal includes the following modules: The data acquisition module is used to acquire the power monitoring time sequence before the adjustment point; The data analysis module is used to obtain the degree of anomalousness of the data points based on the neighborhood fluctuation characteristics of the data points in the power monitoring time series; to obtain anomalous points based on the degree of anomalousness; and to obtain the fluctuation stability value of the anomalous points based on the interval characteristics and data difference characteristics between the anomalous points and other adjacent anomalous points. The prediction and adjustment module is used to adjust the weight coefficients of the non-normal points in the ARIMA model according to the fluctuation stability value to obtain an adaptive ARIMA model; and to predict the power monitoring time series according to the adaptive ARIMA model to obtain the power prediction time series. An adaptive acquisition module is used to adjust a preset acquisition frequency based on the difference characteristics of abnormal points between the power prediction time series and the power monitoring time series to obtain an adaptive acquisition frequency; and to acquire power data after the adjustment point based on the adaptive acquisition frequency. The step of obtaining the degree of anomalousness of the data point based on the neighborhood fluctuation characteristics of the data point in the power monitoring time series includes: Construct a rectangular coordinate system for the time series of power monitoring, with the horizontal axis representing time and the vertical axis representing the power monitoring values, where R represents the degree of abnormality. Indicates normalization, This represents the Euclidean distance between the data point and the previous data point. This represents the Euclidean distance between the data point and the next data point. Represents discrete eigenvalues. This represents the slope of the line connecting the data point to the previous data point. This represents the slope of the line connecting the given data point to the next data point. Indicates the characteristic value of change; The step of obtaining the abnormal point based on the degree of abnormality includes: Data points whose abnormality level exceeds a preset threshold are defined as abnormal points. The step of adjusting the weight coefficients of the non-normal points in the ARIMA model according to the fluctuation stability value to obtain the adaptive ARIMA model includes: Calculate the product of the preset autoregressive coefficient of the non-normal point in the ARIMA model and the stable fluctuation value to obtain the adaptive autoregressive coefficient; calculate the product of the preset moving average coefficient of the non-normal point in the ARIMA model and the stable fluctuation value to obtain the adaptive moving average coefficient; obtain the adaptive ARIMA model based on the adaptive autoregressive coefficient and the adaptive moving average coefficient.

2. The station automation terminal with efficient power data acquisition function according to claim 1, characterized in that, The step of obtaining the fluctuation stability value of the abnormal point based on the interval characteristics and data difference characteristics between the abnormal point and other adjacent abnormal points includes: Calculate the reciprocal of the sum of the intervals between the abnormal point and the two adjacent abnormal points to obtain the nearest neighbor feature value; calculate the reciprocal of the absolute value of the difference between the abnormal point and the two adjacent abnormal points to obtain the regularity feature value; calculate the product of the nearest neighbor feature value and the regularity feature value and normalize it to obtain a first value; calculate the absolute value of the difference between the abnormal point and the previous adjacent abnormal point to obtain a first difference value; calculate the absolute value of the difference between the abnormal point and the next adjacent abnormal point to obtain a second difference value; calculate the reciprocal of the sum of the first difference and the second difference and normalize it to obtain a second value; calculate the average of the first value and the second value to obtain the fluctuation stability value of the abnormal point.

3. A station automation terminal with efficient power data acquisition function according to claim 1, characterized in that, The step of adjusting the preset acquisition frequency based on the difference characteristics of abnormal points between the power prediction time series and the power monitoring time series to obtain an adaptive acquisition frequency includes: Calculate the ratio of the number of abnormal points between the power prediction time series and the power monitoring time series to obtain the adjustment coefficient; calculate the product of the adjustment coefficient and the acquisition frequency of the power monitoring time series to obtain the adaptive acquisition frequency.

4. A station automation terminal with efficient power data acquisition function according to claim 1, characterized in that, The step of collecting electrical energy data after the adjustment point according to the adaptive acquisition frequency includes: The power data after the adjustment point is collected according to the adaptive acquisition frequency until the time of the next adjustment point is stopped; a new adaptive acquisition frequency is obtained according to the new power monitoring timing before the next adjustment point, and the power data after the next adjustment point is adaptively collected.

Images