Apparatus Control Using Predicted and Unobservable Environmental Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional apparatus control devices, such as air conditioning control systems, fail to accurately calculate control values due to the lack of consideration for unobservable values that cannot be directly observed by sensors, leading to inadequate environmental control.
Innovation Solution
An apparatus control device that utilizes two learning models, one for predicting observable values and another for estimating unobservable values, to calculate control values by substituting these values into equations of state for various control methods, thereby selecting an appropriate control value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only observable values from sensors are used for control calculation, then the device complexity is reduced, but the control accuracy deteriorates due to lack of unobservable value consideration
Solution Approach 1:
The patent introduces a learning model as an intermediary component that estimates unobservable values (such as thermal load) based on observable sensor data. This mediator bridges the gap between simple sensor measurements and the need for comprehensive environmental understanding, enabling accurate control calculations without directly measuring unobservable quantities.
Solution Approach 2:
The patent replaces direct physical measurement mechanisms with a computational estimation approach using learning models. Instead of installing additional sensors to measure unobservable values like thermal load, the system uses software-based learning models to infer these values from existing sensor data, substituting mechanical measurement systems with intelligent algorithms.
2Reliability
If multiple control methods are evaluated with both observable and unobservable values, then the control quality is improved, but the calculation time increases
Solution Approach 1:
The patent performs preliminary estimation of unobservable values using learning models before the control calculation process. By pre-computing these estimated values and making them available when needed, the system avoids time-consuming calculations during critical control moments, thus reducing overall calculation time while maintaining control quality.
Data Source
AI summary
An apparatus control device includes an observed value acquiring unit that acquires, from a sensor that observes an environment in which a control target apparatus is installed, an observed value of the environment, an observation predicted value acquiring unit that gives an observed value acquired to a first learning model and acquires an observation predicted value that is a future observed value from the sensor from the first learning model, and an unobservable value acquiring unit that gives an observed value acquired to a second learning model and acquires an unobservable value that is a value not directly observed by the sensor from the second learning model. In addition, the apparatus control device includes a control value calculating unit that calculates a control value of the control target apparatus using the observed value acquired, the observation predicted value acquired, and the unobservable value acquired.