Big data miner for multi-dimensional load characteristic analysis

Abstract

The invention relates to a big data miner for multi-dimensional load characteristic analysis. According to the invention, a power consumer consumption behavior mode analysis module is connected with a load characteristic index analysis module, a consumption behavior mode factor and influence mechanism analysis module and a load curve abnormality diagnosis module, the consumption behavior mode factor and influence mechanism analysis module is connected with a power consumer demand response baseline calculation module and a demand response potential analysis module, the power consumer demand response baseline calculation module is connected with a power consumer demand response effect evaluation module, and the load curve abnormality diagnosis module is connected with an emergency load forecasting module. The big data miner provided by the invention for the multi-dimensional load characteristic analysis can be used for deep mining of load characteristics of a power consumer under the big data environment, can provide scientific theoretical support for two-way interactive demand response technology, and gives full play to the operating efficiency.

Description

technical field [0001] The invention relates to the technical field of power systems and automation thereof, in particular to a big data excavator for multi-dimensional load characteristic analysis. Background technique [0002] The analysis of load characteristics is the basis of smart grid research. There are two main types of traditional power system analysis methods for power consumption characteristics: one is based on the analysis of influencing factors, that is, the dependent variables are extracted one by one under the premise that the remaining variables remain unchanged, and qualitative or quantitative descriptions are made. The extent of its influence on the independent variable, but the objects are all regional power grids, which is far less than the research accuracy and depth of the power supply side, and cannot meet the needs of implementing intelligent demand-side management; Analyze the power consumption characteristics of a certain type of users in detail, ...

AI Technical Summary

Problems solved by technology

[0002] The analysis of load characteristics is the basis of smart grid research. There are two main types of traditional power system analysis methods for power consumption characteristics: one is based on the analysis of influencing factors, that is, the dependent variables are extracted one by one under the premise that the remaining variables remain unchanged, and qualitative or quantitative descriptions are made. The extent of its influence on the independent variable, but the objects are all regional power grids, which is far less than the research accuracy and depth of the power supply side, and cannot meet the needs of implementing intelligent demand-side management; Analyze the power consumption characteristics of a certain type of users in detail, and obtain the impact of various types of users on the power consumption characteristics of the regional power grid qualitatively or quantitatively

But its accuracy is not high, and it is difficult to carry out load forecasting based on the analysis results

[0003] Generally speaking, the data utilization rate of the existing load characteristic analysis technology is low, the analysis results are rough, the accuracy is not high, and it cannot go deep into the user level, which is not conducive to the refinement of demand response strategy formulation

Images