Factory Electricity Prediction for Abnormal Energy Use Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to effectively determine the effectiveness of energy-saving measures in factory electricity consumption, making it difficult to detect abnormal consumption promptly and leading to unsatisfactory energy costs.
Innovation Solution
An energy-saving prediction method using an electricity consumption prediction model to forecast reference consumption based on historical data, comparing it with actual consumption, and providing a function to highlight differences, allowing for timely detection of abnormalities and evaluation of energy-saving measures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If factory electricity consumption is monitored across multiple aspects (air-conditioning, air-compression, production, basic), then comprehensive energy management is achieved, but it becomes difficult to determine effectiveness of energy-saving measures and detect abnormal consumption in time
Solution Approach 1:
The patent segments electricity consumption into distinct categories (air-conditioning, air-compression, production, basic) and applies separate prediction models to each category. This segmentation allows for precise monitoring of each aspect while managing complexity through modular prediction models that can be independently trained and adjusted.
Solution Approach 2:
The patent introduces an electricity consumption prediction model as an intermediary that processes raw consumption data and provides predicted values for comparison. This intermediary layer simplifies the analysis by automatically identifying deviations between actual and predicted consumption, making it easier to detect abnormalities without manually analyzing complex multi-aspect data.
2Reliability
If traditional electricity consumption monitoring methods are used, then basic data collection is achieved, but effectiveness of energy-saving measures cannot be determined and abnormal consumption cannot be detected promptly
Solution Approach 1:
The patent implements preliminary action by training prediction models on historical data before actual energy-saving measures are implemented. The models establish baseline expected consumption patterns in advance, enabling immediate detection of abnormalities when actual consumption deviates from predictions, thus reducing the time to detect issues while ensuring reliable evaluation of energy-saving measures.
Solution Approach 2:
The patent establishes a feedback mechanism that continuously compares actual electricity consumption with predicted consumption and provides real-time alerts when deviations exceed thresholds. This feedback loop enables prompt detection of abnormal consumption and reliable assessment of energy-saving measure effectiveness by showing the impact over time.
Data Source
AI summary
An energy-saving prediction method for factory electricity consumption and an electronic apparatus are provided. The method includes the following steps. A reference electricity consumption amount in a unit period is predicted based on factory actual operation information in the unit period by using an electricity consumption prediction model. An actual electricity consumption amount in the unit period is acquired. The reference electricity consumption amount and the actual electricity consumption amount are displayed on an electricity consumption reference interface. A function is provided according to a first difference value between the reference electricity consumption amount and the actual electricity consumption amount.