Cleansed Power Grid Data Prediction Using Weather Bucketing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Electric power grid operators face challenges in effectively using data from supply and load meters to make accurate predictions of future power provision and consumption due to differences in data protocols, units, and frequent changes at endpoint locations, leading to reliance on guesswork based on personal experience rather than analytical methods.

Innovation Solution

An apparatus and method that processes initial supply and load meter data by assigning values to buckets based on weather conditions, deriving bounds, identifying outliers, and generating naive models to fill gaps, merging cleansed data, and using linear regression to create power provision and consumption models for predictive analytics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If operators rely on personal experience and assumptions to predict power provision and consumption, then the process is simple and requires no complex analysis, but the prediction accuracy deteriorates significantly

Engineering Contradiction:

Improvesimplicity of prediction processVSAvoidprediction accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary data processing system that acts as a mediator between raw meter data and prediction models. This system cleanses, integrates, and standardizes data from multiple sources (supply meters, load meters, weather stations) before feeding it to prediction algorithms, thereby improving prediction accuracy without requiring operators to directly handle complex data processing

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/manual approach of expert-based guessing with automated computational systems. Machine learning models and statistical algorithms substitute for human intuition and experience, objectively analyzing historical data patterns to generate predictions with quantifiable accuracy metrics

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Quantity of substance

If operators collect and use large quantities of meter data for predictions, then more information is available for analysis, but the data quality and usability deteriorate due to protocol differences, missing values, and outliers

Engineering Contradiction:

Improvequantity of dataVSAvoiddata quality

Core Design Contradiction:

Quantity of substanceVSReliability

Solution Approach 1:

The patent segments the data processing task into distinct stages: data collection from multiple sources, data cleansing (handling missing values and outliers), data integration (combining supply and load data), and model training. This segmentation allows each stage to be optimized independently, maintaining data quality while processing large volumes

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts and removes harmful elements from the data set through systematic cleansing procedures. Missing values are identified and handled through imputation or removal, outliers are detected using statistical methods and either corrected or excluded, and inconsistent data from different protocols are standardized, thereby extracting only the reliable signal from the noisy raw data

Inventive Principle:

Principle #2Taking out (Extraction)

3Adaptability or versatility

If operators integrate data from multiple sources with different protocols and units, then the comprehensiveness of analysis improves, but the complexity of data processing increases

Engineering Contradiction:

Improvecomprehensiveness of data integrationVSAvoiddata processing complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal data processing framework that handles multiple data sources (supply meters, load meters, weather stations) with different protocols and units through a single integrated system. This framework provides standardized interfaces for data ingestion, automatic protocol detection and conversion, and unified data structures that accommodate diverse input formats, thereby managing complexity while maintaining comprehensiveness

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10418811B2Electric power grid supply and load prediction using cleansed time series data

Publication Date: 2019.09.17 SAS INSTITUTE INC

AI summary

An apparatus includes a processor to: assign each value of each set of values of an initial supply meter data and of an initial load meter data to one of multiple buckets based on weather conditions and/or time and date; for each bucket, generate upper and lower bounds of power provision and power consumption values, and use the upper and lower bounds to identify outlier values assigned to the bucket; for each set of values within the initial supply meter data and within the initial load meter data, generate a naive model from the non-outlier values, and use interpolation and the naive model to fill in gaps, thereby generating cleansed supply meter data and cleansed load meter data; and store the cleansed supply meter data and cleansed load meter data together as merged meter data for use in making predictions.