CN-122022000-A - Energy consumption prediction method, device, electronic equipment and computer readable storage medium

CN122022000ACN 122022000 ACN122022000 ACN 122022000ACN-122022000-A

Abstract

According to the energy consumption prediction method, the device, the electronic equipment and the computer readable storage medium, through collecting the energy consumption data, the environment data and the production scheduling data generated by actual production in the current collection period, the association relation among the environment, the production scheduling and the actual production energy consumption can be obtained, so that the prediction result can accord with the characteristics of the current production environment, meanwhile, the energy consumption in the prediction period can be predicted based on the specific production in the prediction period by combining with the production scheduling data in the prediction period, the prediction result can be adjusted based on the production scheduling condition, the association between the energy consumption prediction and the actual production is ensured, and the accuracy of the energy consumption prediction is improved.

Inventors

FAN RUIQIANG
ZHU CHENGCHENG
YANG TAOLI
LI LIANG
Xue Chuanli

Assignees

重庆蓝电汽车科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251223

Claims (10)

1. An energy consumption prediction method, characterized in that the energy consumption prediction method comprises: Acquiring actual energy consumption data, production environment data and production scheduling data of a current acquisition period, wherein the production scheduling data comprises actual production scheduling data in the current acquisition period and production scheduling data in a prediction period; generating input data comprising the actual energy consumption data, the production environment data and the production scheduling data; Acquiring a trained energy consumption prediction model, and inputting the input data into the trained energy consumption prediction model; And outputting an energy consumption prediction result of the prediction period based on the input data by the trained energy consumption prediction model.
2. The energy consumption prediction method of claim 1, wherein acquiring the scheduling data for a current acquisition cycle comprises: acquiring source scheduling plan data, and determining the time length of a time unit in the current acquisition period and the prediction period; determining a throughput per unit time based on the source scheduling data; Determining a target throughput for the time unit based on the throughput per unit time and the length of time; synthesizing said target throughput for each of said time units to obtain a periodic production profile; generating said production scheduling data comprising said periodic production profile.
3. The energy consumption prediction method of claim 1, wherein acquiring the scheduling data for a current acquisition cycle comprises: Acquiring source scheduling plan data; determining production parameters in the source scheduling plan data; Generating said production data comprising said production parameters.
4. The energy consumption prediction method of claim 1, wherein acquiring the production environment data for a current acquisition cycle comprises: acquiring historical workshop environment parameters, historical external environment parameters and historical actual energy consumption data in historical input data; fitting the historical workshop environment parameters, the historical external environment parameters and the historical actual energy consumption data to obtain a temperature energy consumption relation; Acquiring current workshop environment parameters acquired by a workshop environment sensor and external environment parameters acquired by a current external environment sensor in a current acquisition period; generating the production environment data including the current plant environment parameters, the current external environment sensors, and the temperature energy consumption relationship.
5. The energy consumption prediction method according to claim 1, wherein the energy consumption prediction model comprises an input layer, a long-short-term memory layer, an attention mechanism layer and an output layer, and wherein the training-completed energy consumption prediction model outputs the prediction result of the prediction period based on the input data comprises: inputting the input data to the input layer so that the input layer outputs a time sequence corresponding to the input data; inputting the time sequence into the long-period memory layer so that the long-period memory layer outputs the time step characteristics corresponding to the time sequence; Inputting the time step feature to the attention mechanism layer so that the attention mechanism layer outputs a feature vector corresponding to the time step feature, wherein the attention mechanism layer distributes a first weight for the production data, distributes a second weight for the production environment data and distributes a third weight for the actual energy consumption data, and the first weight and the second weight are larger than the third weight; And inputting the feature vector to the output layer, so that the output layer maps the feature vector to the prediction period to obtain the energy consumption prediction result.
6. The energy consumption prediction method according to claim 1, characterized in that the energy consumption prediction method further comprises: when the updating moment is reached, training sample data acquired in a current updating sample period are acquired, wherein the training sample data comprise the actual energy consumption data, the production environment data and the actual production scheduling data acquired in the current updating sample period; and updating the trained energy consumption prediction model through the training sample data.
7. The energy consumption prediction method according to claim 6, wherein the updating of the trained energy consumption prediction model by the training sample data includes: Acquiring actual scheduling data in the training samples; Matching a loss weight corresponding to the actual production data, wherein the loss weight is positively correlated with the production output of the actual production data; determining a loss function corresponding to the loss weight; and updating the trained energy consumption prediction model through the training sample data and the loss function.
8. An energy consumption prediction apparatus, characterized in that the energy consumption prediction apparatus comprises: The first acquisition module is used for acquiring actual energy consumption data, production environment data and production scheduling data of a current acquisition period, wherein the production scheduling data comprise actual production scheduling data in the current acquisition period and production scheduling data in a prediction period; the first generation module is used for generating input data comprising the actual energy consumption data, the production environment data and the production scheduling data; The second acquisition module is used for acquiring the energy consumption prediction model after training, and inputting the input data into the energy consumption prediction model after training; And the first output module is used for outputting an energy consumption prediction result of the prediction period based on the input data through the trained energy consumption prediction model.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the energy consumption prediction method according to any one of claims 1 to 7.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the energy consumption prediction method according to any of claims 1 to 7.

Description

Energy consumption prediction method, device, electronic equipment and computer readable storage medium Technical Field The present invention relates to the field of production management, and in particular, to an energy consumption prediction method, an apparatus, an electronic device, and a computer readable storage medium. Background However, in the actual production process, the production environment is complex, and various factors can influence the finally generated energy consumption, so that the energy consumption is difficult to accurately predict by the energy consumption prediction method which only depends on the historical energy consumption data. Disclosure of Invention The invention mainly aims to provide an energy consumption prediction method, an energy consumption prediction device, electronic equipment and a computer readable storage medium, and aims to solve the problem that an energy consumption prediction method depending on historical energy consumption data in the prior art is difficult to accurately predict energy consumption. To achieve the above object, the present invention provides an energy consumption prediction method, including the steps of: Acquiring actual energy consumption data, production environment data and production scheduling data of a current acquisition period, wherein the production scheduling data comprises actual production scheduling data in the current acquisition period and production scheduling data in a prediction period; generating input data comprising the actual energy consumption data, the production environment data and the production scheduling data; Acquiring a trained energy consumption prediction model, and inputting the input data into the trained energy consumption prediction model; And outputting an energy consumption prediction result of the prediction period based on the input data by the trained energy consumption prediction model. Optionally, acquiring the scheduling data of the current acquisition cycle includes: acquiring source scheduling plan data, and determining the time length of a time unit in the current acquisition period and the prediction period; determining a throughput per unit time based on the source scheduling data; Determining a target throughput for the time unit based on the throughput per unit time and the length of time; synthesizing said target throughput for each of said time units to obtain a periodic production profile; generating said production scheduling data comprising said periodic production profile. Optionally, acquiring the scheduling data of the current acquisition cycle includes: Acquiring source scheduling plan data; determining production parameters in the source scheduling plan data; Generating said production data comprising said production parameters. Optionally, acquiring the production environment data of the current acquisition cycle includes: acquiring historical workshop environment parameters, historical external environment parameters and historical actual energy consumption data in historical input data; fitting the historical workshop environment parameters, the historical external environment parameters and the historical actual energy consumption data to obtain a temperature energy consumption relation; Acquiring current workshop environment parameters acquired by a workshop environment sensor and external environment parameters acquired by a current external environment sensor in a current acquisition period; generating the production environment data including the current plant environment parameters, the current external environment sensors, and the temperature energy consumption relationship. Optionally, the energy consumption prediction model comprises an input layer, a long-short-term memory layer, an attention mechanism layer and an output layer, and the energy consumption prediction model after training is used for outputting the energy consumption prediction result of the prediction period based on the input data comprises the following steps: inputting the input data to the input layer so that the input layer outputs a time sequence corresponding to the input data; inputting the time sequence into the long-period memory layer so that the long-period memory layer outputs the time step characteristics corresponding to the time sequence; Inputting the time step feature to the attention mechanism layer so that the attention mechanism layer outputs a feature vector corresponding to the time step feature, wherein the attention mechanism layer distributes a first weight for the production data, distributes a second weight for the production environment data and distributes a third weight for the actual energy consumption data, and the first weight and the second weight are larger than the third weight; And inputting the feature vector to the output layer, so that the output layer maps the feature vector to the prediction period to obtain the energy consumption prediction result. Option