Risk early warning method for electrical equipment of mountain transformer substation

Abstract

The invention discloses a risk early warning method for electrical equipment of a mountain transformer substation, and belongs to the technical field of electric power. According to the invention, abnormal data is corrected; meanwhile, normalization is carried out to realize dimensionless processing, and the processed data is processed by adopting an LSTM algorithm so that prediction precision canbe effectively enhanced, a manager is enabled to acquire more intuitive and clear equipment temperature change conditions, prediction can be realized, state switching can be carried out through a finite state machine, and thus risk early warning and generator tripping processing can be realized.

Description

technical field

[0001] The invention relates to a risk early warning method for electrical equipment of a mountain substation, and belongs to the technical field of electric power. Background technique

[0002] At present, many substations in our country are forced to be built in mountainous areas. There are many hidden safety risks in mountainous substations, such as forest fires, storms, mountain torrents, thunderstorms, heavy snow, earthquakes and other geological disasters. The harsh geological environment directly threatens the operation safety of electrical equipment. Electrical equipment is disturbed by various external factors during operation, and the insulating materials are gradually degraded. If a fault occurs, it will cause a power outage in some or even the entire area. Therefore, it is necessary to manage real-time online monitoring of mountain substations; use various early warning and monitoring technologies to monitor various equipment in mountain substatio...

Examples

Embodiment 1

[0014] Embodiment 1: as figure 1 As shown, a risk early warning method for electrical equipment in a mountainous substation, the steps of the method are as follows:

[0015] S1. Collect the temperature of key points of electrical equipment, the ambient temperature of the transformer or the load value of the equipment (each type of data is processed according to the following steps);

[0016] One sampling point every 30 minutes, the collected data is a matrix of 48 units for n days;

[0017] S2. Judging the collected data as an abnormal number: if it satisfies then use Perform data correction; otherwise, do not process; get If it is normal data, the data in the matrix is ​​X n,i ,otherwise,

[0018] Among them, X n,i is the data of the i-th sampling point on the nth day, is the mean value of the i-th sampling point on n days, and the variance of the i-th sampling point on n days is ε is the threshold value, usually 1 to 1.5; is the corrected data of the i-th ...

technical field

[0001] The invention relates to a risk early warning method for electrical equipment of a mountain substation, and belongs to the technical field of electric power. Background technique

[0002] At present, many substations in our country are forced to be built in mountainous areas. There are many hidden safety risks in mountainous substations, such as forest fires, storms, mountain torrents, thunderstorms, heavy snow, earthquakes and other geological disasters. The harsh geological environment directly threatens the operation safety of electrical equipment. Electrical equipment is disturbed by various external factors during operation, and the insulating materials are gradually degraded. If a fault occurs, it will cause a power outage in some or even the entire area. Therefore, it is necessary to manage real-time online monitoring of mountain substations; use various early warning and monitoring technologies to monitor various equipment in mountain substatio...

Examples

Embodiment 1

[0014] Embodiment 1: as figure 1 As shown, a risk early warning method for electrical equipment in a mountainous substation, the steps of the method are as follows:

[0015] S1. Collect the temperature of key points of electrical equipment, the ambient temperature of the transformer or the load value of the equipment (each type of data is processed according to the following steps);

[0016] One sampling point every 30 minutes, the collected data is a matrix of 48 units for n days;

[0017] S2. Judging the collected data as an abnormal number: if it satisfies then use Perform data correction; otherwise, do not process; get If it is normal data, the data in the matrix is ​​X n,i ,otherwise,

[0018] Among them, X n,i is the data of the i-th sampling point on the nth day, is the mean value of the i-th sampling point on n days, and the variance of the i-th sampling point on n days is ε is the threshold value, usually 1 to 1.5; is the corrected data of the i-th ...

Images