Abnormality Detection Control with Protected Feature Data Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing control devices for predictive maintenance do not adequately protect know-how related to abnormality detection processing, as they only conceal the program but not the data during processing, leaving the data vulnerable to unauthorized access.
Innovation Solution
A control device with a feature extraction unit, processing unit, and determination unit that calculates and stores feature amounts and scores, along with a first data storage unit for restricted access and a second data storage unit for public state values, utilizing an authority management unit to restrict access and ensure security through identification and password verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If access to all data in the control device is allowed for verification purposes, then the validity of abnormality detection can be confirmed, but the know-how related to feature amounts and processing data is exposed
Solution Approach 1:
The patent segments data into two categories: state values (public) and processing data including feature amounts (secret). This segmentation allows differential access control where verification can be performed on state values while processing data remains protected. The feature extraction unit and processing unit generate and store this segmented data structure, enabling selective disclosure for validation without exposing proprietary algorithms or intermediate calculations.
Solution Approach 2:
The authority management unit acts as an intermediary that mediates access requests between external devices and the data storage unit. It verifies authentication information and selectively permits access to only state values while blocking access to processing data. This intermediary mechanism enables validity confirmation through controlled access to non-sensitive data without exposing the know-how contained in processing data.
2Reliability
If access to processing data is restricted to protect know-how, then security is improved, but the ability to verify abnormality detection validity is reduced
Solution Approach 1:
The data storage unit stores state values and processing data separately with different access permissions. State values can be accessed by external devices for verification purposes, while processing data remains restricted. This segmentation structure allows the system to maintain security over know-how while still permitting validation of the abnormality detection functionality through access to input state values and output determination results.
Solution Approach 2:
The authority management unit serves as an intermediary that enables controlled verification. It allows external devices to access state values and determination results for validity confirmation while blocking access to the processing data containing feature amounts and algorithmic information. This mediated access resolves the contradiction by providing just enough information for verification without exposing proprietary know-how.
3Ease of operation
If all data is made accessible for transparency, then verification ease is improved, but the protective mechanism for know-how is weakened
Solution Approach 1:
The patent implements segmentation of data accessibility by distinguishing between state values (accessible) and processing data (protected). This allows external devices to easily verify abnormality detection by accessing state values and determination results without requiring access to protected processing data. The segmented structure provides verification ease for non-sensitive operations while maintaining strong protection for proprietary know-how.
Solution Approach 2:
The authority management unit acts as an intermediary that facilitates easy verification of non-sensitive data while automatically protecting sensitive information. It provides a simple interface for accessing state values and determination results for validation purposes, while transparently blocking access to processing data. This intermediary mechanism achieves ease of operation for verification without compromising the reliability of know-how protection.
Data Source
AI summary
A control device includes: a feature extraction unit that calculates one or more feature amounts from one or more state values; a processing unit that calculates a score based on the one or plurality of feature amounts calculated by the feature extraction unit with reference to a learning model; a determination unit that generates a determination result indicating whether any abnormality has occurred in a monitoring target based on the score; a first data storage unit that stores at least one of data related to processing in the feature extraction unit and data related to processing in the processing unit; a second data storage unit that stores an arbitrary state value capable of being referred to by the control device; and an authority management unit that restricts access to the first data storage unit.