CN-121995106-A - Electric energy power estimation method, device, equipment and medium

CN121995106ACN 121995106 ACN121995106 ACN 121995106ACN-121995106-A

Abstract

The present invention relates to the field of electric energy meters, and in particular, to a method, an apparatus, a device, and a medium for estimating electric energy power. The method comprises the steps of determining front sampling point time and rear sampling point time corresponding to abnormal sampling point time, obtaining freezing type of an electric energy power data sequence, determining a target estimation model matched with the freezing type, inputting the abnormal sampling point time, the front electric energy power data, the rear sampling point time and the rear electric energy power data into the target estimation model, and outputting estimated electric energy power data corresponding to the abnormal sampling point time by combining the target correction value. By detecting the abnormality of the electric energy power data sequence, the corresponding time point of the abnormality data can be accurately positioned, and by setting the time of the front sampling point and the time of the rear sampling point with different time lengths from the time of the abnormality sampling point, the more stable historical data trend before the abnormality occurs and the more approximate instant data information after the abnormality occurs can be more fully utilized, so that the estimation precision is improved.

Inventors

LI KEGUANG
CHEN ZUOZHI
LIU JIANJUN

Assignees

深圳市思达仪表有限公司

Dates

Publication Date: 20260508
Application Date: 20260106

Claims (10)

1. A method of estimating electrical power, the method comprising: Acquiring an electric energy power data sequence of a plurality of acquired sampling point times, carrying out anomaly detection on the electric energy power data of each sampling point time, and determining anomaly electric power data, anomaly type of the anomaly electric power data and anomaly sampling point time corresponding to the anomaly electric power data; The method comprises the steps of obtaining a target abnormal type to be estimated, and determining a front sampling point time and a rear sampling point time corresponding to the abnormal sampling point time according to any abnormal sampling point time in the target abnormal type, wherein the distance between the front sampling point time and the abnormal sampling point time is a first duration, the distance between the rear sampling point time and the abnormal sampling point time is a second duration, and the first duration is longer than the second duration; Acquiring a freezing type of the electric energy power data sequence, and determining a target estimation model matched with the freezing type, wherein the target estimation model comprises parameters corresponding to the abnormal sampling point time, the front electric energy power data of the front sampling point time, the rear sampling point time and the rear electric energy power data of the rear sampling point time, and a target correction term; And calculating a target correction value of the target correction term, inputting the abnormal sampling point time, the front electric energy power data, the rear sampling point time and the rear electric energy power data into the target estimation model, and outputting estimated electric energy power data corresponding to the abnormal sampling point time by combining the target correction value.
2. The estimation method of claim 1, wherein the anomaly types include a missing anomaly type, a spike anomaly type, and a persistent anomaly type; The abnormality detection for each piece of electric energy power data, determining abnormal electric power data, an abnormal type of the abnormal electric power data and an abnormal sampling point time corresponding to the abnormal electric power data, includes: Detecting missing abnormality of each piece of electric energy power data, and determining abnormal electric power data corresponding to the missing abnormality type and abnormal sampling point time corresponding to the abnormal electric power data; Performing peak value anomaly detection on each piece of electric energy power data, and determining anomaly sampling point time corresponding to the anomaly electric power data corresponding to the peak value anomaly type and the anomaly electric power data; And carrying out continuous anomaly detection on each piece of electric energy power data, and determining the anomaly sampling point time of the anomaly electric power data corresponding to the continuous anomaly type and the anomaly electric power data.
3. The estimation method as set forth in claim 1, wherein the determining the front sampling point time and the rear sampling point time corresponding to the abnormal sampling point time includes: based on the time sequence, calculating the first time length forwards from the abnormal sampling point time to obtain the time of the previous sampling point; and based on the time sequence, calculating the second time length backwards from the abnormal sampling point time to obtain the time of the rear sampling point.
4. The estimation method of claim 1, wherein the freeze type includes a month freeze type; Before the determining the target estimation model matched with the freezing type, the method further comprises: acquiring the days of the month corresponding to the abnormal sampling point time and a first correction item matched with the month freezing type; And constructing a first estimation model matched with the month freezing type according to the days, the abnormal sampling point time, the front electric energy power data, the rear sampling point time, the rear electric energy power data and the first correction term.
5. The estimation method of claim 1, wherein the freeze type comprises a daily freeze type; Before the determining the target estimation model matched with the freezing type, the method further comprises: Acquiring a second correction term matched with the daily freezing type; And constructing a second estimation model matched with the daily freezing type according to the abnormal sampling point time, the front electric energy power data, the rear sampling point time, the rear electric energy power data and the second correction term.
6. The estimation method of claim 1, wherein the freeze type comprises a continuous freeze type; Before the determining the target estimation model matched with the freezing type, the method further comprises: acquiring an acquisition period of the continuous freezing type and a third correction item matched with the continuous freezing type; And constructing a third estimation model matched with the continuous freezing type according to the acquisition period, the abnormal sampling point time, the front electric energy power data, the rear sampling point time, the rear electric energy power data and the third correction term.
7. An estimation device of electric energy power, characterized in that it comprises: the detection module is used for acquiring an electric energy power data sequence of a plurality of acquired sampling point times, carrying out anomaly detection on the electric energy power data of each sampling point time, and determining anomaly electric power data, anomaly type of the anomaly electric power data and anomaly sampling point time corresponding to the anomaly electric power data; The device comprises a first determining module, a first judging module and a second judging module, wherein the first determining module is used for obtaining a target abnormal type to be estimated, and determining a front sampling point time and a rear sampling point time corresponding to the abnormal sampling point time aiming at any abnormal sampling point time in the target abnormal type, wherein the distance between the front sampling point time and the abnormal sampling point time is a first duration, the distance between the rear sampling point time and the abnormal sampling point time is a second duration, and the first duration is longer than the second duration; The second determining module is used for obtaining the freezing type of the electric energy power data sequence, determining a target estimation model matched with the freezing type, wherein the target estimation model comprises the abnormal sampling point time, the front electric energy power data of the front sampling point time, parameters corresponding to the rear sampling point time and the rear electric energy power data of the rear sampling point time, and a target correction term; And the output module is used for calculating and obtaining a target correction value of the target correction term, inputting the abnormal sampling point time, the front electric energy power data, the rear sampling point time and the rear electric energy power data into the target estimation model, and outputting estimated electric energy power data corresponding to the abnormal sampling point time by combining the target correction value.
8. The estimation device of claim 7, wherein the detection module comprises: the first detection unit is used for carrying out missing abnormality detection on each piece of electric energy power data and determining abnormal sampling point time corresponding to the abnormal electric power data corresponding to the missing abnormality type and the abnormal electric power data; The second detection unit is used for carrying out peak value abnormality detection on each piece of electric energy power data and determining abnormal sampling point time corresponding to abnormal electric power data corresponding to the peak value abnormality type and the abnormal electric power data; and the third detection unit is used for carrying out continuous anomaly detection on each electric energy power data and determining the anomaly sampling point time corresponding to the anomaly electric power data corresponding to the continuous anomaly type and the anomaly electric power data.
9. A computer device, characterized in that it comprises a processor, a memory and a computer program stored in the memory and executable on the processor, which processor implements the evaluation method according to any one of claims 1 to 6 when executing the computer program.
10. A computer readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the estimation method according to any one of claims 1 to 6.

Description

Electric energy power estimation method, device, equipment and medium Technical Field The present invention relates to the field of electric energy meters, and in particular, to a method, an apparatus, a device, and a medium for estimating electric energy power. Background In the prior art, when the electric energy power of the electric energy meter is estimated, a traditional sampling calculation mode is generally adopted, for example, the current data is estimated by taking the average value of the data in a certain time window, the current data is estimated by taking the adjacent data, and the like, and although the method is simple to realize, the method has obvious limitations in practical application. When the electric power environment where the electric energy meter is located has rapid fluctuation, sudden load change or high-frequency noise interference, the electric energy meter is estimated only by relying on the average value of a single time window or the time of adjacent sampling points, so that larger deviation is generated between an estimation result and a real power value, and the dynamic change characteristic of instantaneous power cannot be accurately reflected. In the environment with high-frequency electromagnetic interference, the adjacent sampling point time may be polluted by noise, and the noise is introduced into the result by directly estimating the nearest sampling point time, so that the estimation accuracy is further reduced. Therefore, how to improve the estimation accuracy in the electric power estimation process is a problem to be solved. Disclosure of Invention In view of this, the embodiments of the present application provide a method, an apparatus, a device, and a medium for estimating electric power, so as to solve the problem of low estimation accuracy in the electric power estimation process. In a first aspect, an embodiment of the present application provides a method for estimating electric power, where the method includes: Acquiring an electric energy power data sequence of a plurality of acquired sampling point times, carrying out anomaly detection on the electric energy power data of each sampling point time, and determining anomaly electric power data, anomaly type of the anomaly electric power data and anomaly sampling point time corresponding to the anomaly electric power data; The method comprises the steps of obtaining a target abnormal type to be estimated, and determining a front sampling point time and a rear sampling point time corresponding to the abnormal sampling point time according to any abnormal sampling point time in the target abnormal type, wherein the distance between the front sampling point time and the abnormal sampling point time is a first duration, the distance between the rear sampling point time and the abnormal sampling point time is a second duration, and the first duration is longer than the second duration; Acquiring a freezing type of the electric energy power data sequence, and determining a target estimation model matched with the freezing type, wherein the target estimation model comprises parameters corresponding to the abnormal sampling point time, the front electric energy power data of the front sampling point time, the rear sampling point time and the rear electric energy power data of the rear sampling point time, and a target correction term; And calculating a target correction value of the target correction term, inputting the abnormal sampling point time, the front electric energy power data, the rear sampling point time and the rear electric energy power data into the target estimation model, and outputting estimated electric energy power data corresponding to the abnormal sampling point time by combining the target correction value. In a second aspect, an embodiment of the present application provides an estimation apparatus of electric power, the estimation apparatus including: the detection module is used for acquiring an electric energy power data sequence of a plurality of acquired sampling point times, carrying out anomaly detection on the electric energy power data of each sampling point time, and determining anomaly electric power data, anomaly type of the anomaly electric power data and anomaly sampling point time corresponding to the anomaly electric power data; The device comprises a first determining module, a first judging module and a second judging module, wherein the first determining module is used for obtaining a target abnormal type to be estimated, and determining a front sampling point time and a rear sampling point time corresponding to the abnormal sampling point time aiming at any abnormal sampling point time in the target abnormal type, wherein the distance between the front sampling point time and the abnormal sampling point time is a first duration, the distance between the rear sampling point time and the abnormal sampling point time is a second duration, and the first duration is longer t