Server, air conditioner and method for controlling thereof
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing air conditioner technologies face inefficiencies in predicting cooling tendencies due to insufficient user setting information and the overhead of generating models for each unit, making it difficult to accurately predict user cooling needs.
Innovation Solution
A server system that groups air conditioners with similar usage patterns and uses a cooling capacity prediction model to predict cooling capacity, allowing for optimized operation without requiring extensive user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a model is generated for each air conditioner to predict cooling tendency, then prediction accuracy for individual users is improved, but the overhead of batch processing for learning significantly increases
Solution Approach 1:
The patent merges multiple individual air conditioner models into a single unified model that processes data from multiple units. Instead of training separate models for each air conditioner (which increases overhead), the system combines datasets from multiple units and trains one model that learns from aggregated patterns, thereby maintaining prediction accuracy while significantly reducing batch processing overhead.
Solution Approach 2:
The patent creates a universal model that serves multiple air conditioners simultaneously. This single model is designed to handle prediction tasks for various users and units, making it multi-functional. The model learns from diverse data sources and can generalize to predict cooling tendencies across different air conditioners without requiring separate specialized models for each unit.
2Ease of operation
If user setting information is collected directly from a small number of users, then data collection is simple, but the user's setting information for learning by the AI model is insufficient
Solution Approach 1:
The patent combines setting information from multiple air conditioners and multiple users into a unified dataset. By aggregating data across different units and users, the system accumulates sufficient learning material without requiring each individual user to provide extensive settings. The merged dataset enables the AI model to learn from diverse patterns while maintaining ease of data collection.
Solution Approach 2:
The patent introduces an intermediary data aggregation layer that collects settings from multiple air conditioners and consolidates them into a shared learning pool. This intermediary mechanism allows individual users to continue providing simple settings while the system accumulates sufficient data by combining inputs from multiple sources, thereby solving the data insufficiency problem without complicating the collection process.
3Ease of operation
If air conditioners operate with directly user-set parameters, then user control is simple, but accurate prediction of user cooling tendency becomes difficult with insufficient setting information
Solution Approach 1:
The patent introduces an intermediary AI model that acts as a bridge between simple user settings and accurate cooling tendency prediction. The model receives basic user inputs and combines them with aggregated data from multiple air conditioners and users, then generates accurate predictions. This intermediary layer enables the system to maintain simple user control while achieving high prediction accuracy through intelligent data processing and pattern recognition.
Data Source
AI summary
A server is provided. The server includes a communication device configured to receive operation information from an air conditioner, a memory configured to store information of a plurality of groups and a cooling capacity prediction model for predicting a cooling capacity that corresponds to each of the plurality of groups, and a processor to map the air conditioner to one group among the plurality of groups based on the received operation information, and control the communication device to enable the air conditioner to use the cooling capacity prediction model corresponding to the mapped group.