CN-122022216-A - Energy system processing method and device, electronic equipment and storage medium

CN122022216ACN 122022216 ACN122022216 ACN 122022216ACN-122022216-A

Abstract

The method comprises the steps of predicting a future environment parameter, combining a current energy production parameter and a current load parameter of a target energy system, obtaining the corresponding future energy production parameter and future load parameter of the target energy system in a future time period by prediction, and further determining the corresponding energy management strategy of the target energy system in the future time period according to the current energy production parameter, the current load parameter, the energy price parameter, the future energy production parameter and the future load parameter. So as to be able to effectively cope with the randomness and intermittence of the renewable energy sources, as well as the uncertainty of the user's needs, in order to improve the stability and economy of the target energy system.

Inventors

HAN XIAOHAN
WANG HAO
JI XIAOFAN
SUN XUN
FAN YUJIAN
ZHANG WENQI
WANG JIAKUN

Assignees

国家能源投资集团有限责任公司
北京低碳清洁能源研究院
国华(栖霞)风力发电有限公司

Dates

Publication Date: 20260512
Application Date: 20241106

Claims (10)

1. An energy system processing method, comprising: acquiring a historical environment parameter corresponding to a target energy system in a historical time period; determining a future environmental parameter corresponding to the target energy system in a future time period according to the historical environmental parameter; Processing the future environmental parameters, the current energy production parameters and the current load parameters of the target energy system through a target management model to obtain the future energy production parameters and the future load parameters corresponding to the target energy system in the future time period, wherein the target management model is obtained by training a basic management model based on a plurality of sample operation parameters, each sample operation parameter comprises a sample environmental parameter, a sample energy production parameter and a sample load parameter, each sample operation parameter carries a first label mark, and the first label mark comprises an actual energy production parameter and an actual load parameter corresponding to the sample operation parameter in the sample future time period; And determining an energy management strategy corresponding to the target energy system in the future time period according to the current energy production parameter, the current load parameter, the energy price parameter, the future energy production parameter and the future load parameter.
2. The method for processing an energy system according to claim 1, wherein, The determining, according to the historical environmental parameter, a future environmental parameter corresponding to the target energy system in a future time period includes: And processing the historical environment parameters through a time sequence prediction model to obtain future environment parameters corresponding to the target energy system in a future time period, wherein the time sequence prediction model is obtained based on a sample historical environment parameter training basic sequence model corresponding to a plurality of sample historical time periods, each sample historical environment parameter carries a second labeling label, and the second labeling label is an actual future environment parameter corresponding to the sample historical environment parameter in the future time period.
3. The energy system processing method according to claim 2, wherein the time series prediction model is obtained by: performing iterative training on the basic sequence model through corresponding sample history environment parameters in the plurality of sample history time periods; After each round of iterative training, obtaining predicted future environmental parameters corresponding to the round of iterative training; Based on the predicted future environmental parameters corresponding to the round of iterative training and the second labeling labels carried by the sample history environmental parameters corresponding to the round of training, obtaining environmental prediction loss; Optimizing the base sequence model by the environmental prediction loss; and stopping training under the condition that the basic sequence model meets the preset condition to obtain the time sequence prediction model.
4. The method for processing an energy system according to claim 1, wherein, The target management model is obtained through training the following steps: performing iterative training on the basic management model through the plurality of sample operation parameters; After each round of iterative training, obtaining a predicted energy production parameter and a predicted load parameter corresponding to the round of iterative training; Based on the predicted energy production parameter and the predicted load parameter corresponding to the round of iterative training and a first labeling tag carried by the sample operation parameter corresponding to the round of training, obtaining energy prediction loss and load prediction loss; Optimizing the basic management model through the energy prediction loss and the load prediction loss; And stopping training under the condition that the basic management model meets the preset condition to obtain the target management model.
5. The energy system processing method of claim 1, further comprising: acquiring target operation data of the target energy system; processing the target operation data through a target evaluation model to obtain a system evaluation result, wherein the target evaluation model is obtained by training a basic evaluation model based on a plurality of sample operation data, each sample operation data carries a third labeling label, and the third labeling label comprises an actual evaluation result corresponding to the sample operation data; And generating an optimization strategy aiming at the target energy system based on the system evaluation result.
6. The method for processing an energy system according to claim 5, wherein, The target evaluation model is obtained through training the following steps: performing iterative training on the basic evaluation model through the plurality of sample operation data; after each round of iterative training, obtaining a prediction evaluation result corresponding to the round of iterative training; obtaining an evaluation prediction loss based on a prediction evaluation result corresponding to the round of iterative training and a third labeling label carried by sample operation data corresponding to the round of training; optimizing the basic evaluation model through the evaluation prediction loss; And stopping training under the condition that the basic evaluation model meets the preset condition to obtain the target evaluation model.
7. The energy system processing method of claim 1, further comprising: Acquiring system construction parameters corresponding to a target region, wherein the system construction parameters comprise at least one of renewable energy output curves, load requirements, power grid access conditions, policy regulations and geographic conditions; Processing the system construction parameters through an artificial intelligent platform to obtain a system construction scheme corresponding to the target region; And constructing the target energy system according to the system construction scheme.
8. An energy system processing apparatus, comprising: The first acquisition module is configured to acquire historical environment parameters corresponding to the target energy system in a historical time period; A first determining module configured to determine, according to the historical environmental parameters, future environmental parameters corresponding to the target energy system in a future time period; The first obtaining module is configured to process the future environmental parameter, the current energy production parameter and the current load parameter of the target energy system through a target management model to obtain the future energy production parameter and the future load parameter corresponding to the target energy system in the future time period, the target management model is obtained based on a plurality of sample operation parameter training basic management models, each sample operation parameter comprises a sample environmental parameter, a sample energy production parameter and a sample load parameter, each sample operation parameter carries a first label mark, and the first label mark comprises an actual energy production parameter and an actual load parameter corresponding to the sample operation parameter in the future time period; And the second determining module is configured to determine an energy management strategy corresponding to the target energy system in the future time period according to the current energy production parameter, the current load parameter, the energy price parameter, the future energy production parameter and the future load parameter.
9. An electronic device, comprising: A processor; A memory for storing processor-executable instructions; wherein the processor is configured to implement the steps of the energy system processing method of any one of claims 1 to 7 when executed.
10. A computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the steps of the energy system processing method of any of claims 1 to 7.

Description

Energy system processing method and device, electronic equipment and storage medium Technical Field The disclosure relates to the technical field of energy management, and in particular relates to an energy system processing method, an energy system processing device, electronic equipment and a storage medium. Background When an ac/dc hybrid multi-micro energy system is constructed in a certain area, the system needs to be capable of monitoring and processing a large amount of data in real time, including the production, consumption and storage states of electric power, and a safety mechanism capable of protecting the system from risks such as overload and short circuit must be designed, and stability of voltage and frequency and economy of the whole system are ensured. Disclosure of Invention In order to overcome the problems in the related art, the disclosure provides an energy system processing method, an apparatus, an electronic device and a storage medium, which predict future environmental parameters, and combine current energy production parameters and current load parameters of a target energy system to obtain corresponding future energy production parameters and future load parameters of the target energy system in a future time period, so as to determine corresponding energy management strategies of the target energy system in the future time period according to the current energy production parameters, the current load parameters, the energy price parameters, the future energy production parameters and the future load parameters. So as to be able to effectively cope with the randomness and intermittence of the renewable energy sources, as well as the uncertainty of the user's needs, in order to improve the stability and economy of the target energy system. According to a first aspect of embodiments of the present disclosure, there is provided an energy system processing method, including: acquiring a historical environment parameter corresponding to a target energy system in a historical time period; determining a future environmental parameter corresponding to the target energy system in a future time period according to the historical environmental parameter; Processing the future environmental parameters, the current energy production parameters and the current load parameters of the target energy system through a target management model to obtain the future energy production parameters and the future load parameters corresponding to the target energy system in the future time period, wherein the target management model is obtained by training a basic management model based on a plurality of sample operation parameters, each sample operation parameter comprises a sample environmental parameter, a sample energy production parameter and a sample load parameter, each sample operation parameter carries a first label mark, and the first label mark comprises an actual energy production parameter and an actual load parameter corresponding to the sample operation parameter in the sample future time period; And determining an energy management strategy corresponding to the target energy system in the future time period according to the current energy production parameter, the current load parameter, the energy price parameter, the future energy production parameter and the future load parameter. Optionally, the determining, according to the historical environmental parameter, a future environmental parameter corresponding to the target energy system in a future time period includes: And processing the historical environment parameters through a time sequence prediction model to obtain future environment parameters corresponding to the target energy system in a future time period, wherein the time sequence prediction model is obtained based on a sample historical environment parameter training basic sequence model corresponding to a plurality of sample historical time periods, each sample historical environment parameter carries a second labeling label, and the second labeling label is an actual future environment parameter corresponding to the sample historical environment parameter in the future time period. Optionally, the time series prediction model is obtained by: performing iterative training on the basic sequence model through corresponding sample history environment parameters in the plurality of sample history time periods; After each round of iterative training, obtaining predicted future environmental parameters corresponding to the round of iterative training; Based on the predicted future environmental parameters corresponding to the round of iterative training and the second labeling labels carried by the sample history environmental parameters corresponding to the round of training, obtaining environmental prediction loss; Optimizing the base sequence model by the environmental prediction loss; and stopping training under the condition that the basic sequence model meets the preset condi