Combinatorial solution determination system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Black-box optimization problems, such as selecting an air conditioning system with minimum power consumption, require significant computational time due to the inability to formulate an objective function, leading to inefficient device selection and increased costs.
Innovation Solution
A combinatorial solution determination system utilizing an iterated local search method, necessary-series-data extraction, sparse estimation, and extreme value statistics to reduce computational costs by extracting relevant data and optimizing system configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a black-box optimization problem is solved using traditional simulation methods, then the optimal solution can be derived, but a large amount of computation time is required
Solution Approach 1:
The system performs preliminary actions by extracting necessary series data from historical data before the main optimization process. This pre-processing step identifies and stores critical data patterns that will be reused during simulation, reducing the need to process complete historical datasets repeatedly and thereby reducing computation time while maintaining solution optimality
Solution Approach 2:
The system extracts only the necessary series data required for optimization from the complete historical dataset, separating essential information from redundant data. This extraction process identifies key parameters and time series that directly impact the objective function, allowing the optimization to proceed with a reduced dataset while preserving solution accuracy
2Measurement precision
If complete series data is used for simulation, then accurate evaluation indices can be calculated, but the computational cost increases
Solution Approach 1:
The system extracts necessary series data that contains the essential information for accurate evaluation index calculation, removing redundant data points that do not contribute significantly to the accuracy. This selective extraction maintains measurement precision while reducing the data volume that requires computational processing
Solution Approach 2:
The system applies partial action by using only the portion of historical data that is necessary for accurate evaluation, rather than processing the complete dataset. The necessary-series-data extraction unit identifies the minimum required data subset that preserves accuracy for calculating evaluation indices, reducing computational cost while maintaining sufficient precision
3Reliability
If all historical data is processed to satisfy constraint conditions, then comprehensive constraint verification is achieved, but data processing time increases
Solution Approach 1:
The system extracts necessary series data that is specifically relevant for constraint condition verification, separating constraint-critical data from other historical data. This extraction ensures that only the data points that can potentially violate or satisfy constraints are processed, maintaining reliable constraint verification while reducing overall data processing time
Solution Approach 2:
The system applies local quality by treating different portions of historical data differently based on their relevance to constraint conditions. The necessary-series-data extraction unit identifies specific time periods, parameters, or data characteristics that are locally critical for constraint satisfaction, processing these with higher priority and detail while using simplified processing for less critical data portions
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided is a combinatorial solution determination system capable of efficiently deriving a solution of a black-box optimization problem. A combinatorial solution determination system (190) includes a candidate solution generation unit (110), a simulation unit (120), an evaluation index calculation unit (130), a solution determination unit (140), and a necessary-series-data extraction unit (150). The simulation unit calculates simulation data using information related to candidate combinatorial solutions generated by the candidate solution generation unit; and series data. The evaluation index calculation unit calculates evaluation indices based on the simulation data. The solution determination unit determines a combinatorial solution having high evaluation from among the plurality of candidate combinatorial solutions based on the evaluation indices. The necessary-series-data extraction unit extracts, from within the series data (first series data), second series data required for calculating the evaluation indices with predetermined accuracy. The necessary-series-data extraction unit extracts, from within the first series data, third series data required for verifying suitability of a predetermined constraint condition. The necessary-series-data extraction unit combines the second series data and the third series data to acquire necessary series data. After the necessary-series-data extraction unit extracts the necessary series data, the simulation unit calculates the simulation data using information related to the candidate combinatorial solutions and the necessary series data.