Industrial Controller Memory Segmentation for Power Loss Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial controllers face challenges in quickly resuming control after power loss due to the limitations of volatile RAM, which requires data to be copied to non-volatile memory within a short time frame, and existing battery backup or capacitive storage solutions are costly and bulky, unable to meet the increasing data retention needs.
Innovation Solution
A method that classifies data objects as static or dynamic, storing static objects in non-volatile memory and dynamic objects in volatile memory, with changes to static objects recorded in non-volatile memory, allowing only frequently changed dynamic objects to be copied to non-volatile memory during power loss and restored upon power return, minimizing the data to be persisted.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If battery backup or capacitive storage is used to retain data during power loss, then data retention reliability is improved, but device size and expense increase
Solution Approach 1:
The patent segments data objects into static and dynamic categories based on their change frequency. Static data objects (unchanged or rarely changed) are stored in non-volatile memory, while dynamic data objects (frequently changed) are stored in volatile memory. This segmentation allows the system to avoid using battery backup or capacitive storage for all data, thereby reducing device size and expense while maintaining data retention reliability for critical static data.
Solution Approach 2:
The patent applies different storage qualities to different parts of the data set. Non-volatile memory is used specifically for static data objects that require persistence, while volatile memory is used for dynamic data objects that change frequently. This local differentiation of storage quality allows the system to achieve reliable data retention only where necessary, avoiding the need for expensive and bulky battery backup systems for all data.
2Reliability
If battery backup or capacitive storage is used to retain data during power loss, then data retention reliability is improved, but device expense increases
Solution Approach 1:
The patent segments data objects into static and dynamic categories based on their change frequency. Static data objects (unchanged or rarely changed) are stored in non-volatile memory, while dynamic data objects (frequently changed) are stored in volatile memory. This segmentation allows the system to avoid using battery backup or capacitive storage for all data, thereby reducing device size and expense while maintaining data retention reliability for critical static data.
Solution Approach 2:
The patent applies different storage qualities to different parts of the data set. Non-volatile memory is used specifically for static data objects that require persistence, while volatile memory is used for dynamic data objects that change frequently. This local differentiation of storage quality allows the system to achieve reliable data retention only where necessary, avoiding the need for expensive and bulky battery backup systems for all data.
3Reliability
If all data is copied from volatile memory to non-volatile memory during power loss, then data retention is improved, but the time required to complete the copy within the short voltage window becomes excessive
Solution Approach 1:
The patent extracts only the necessary data (static data objects) for copying to non-volatile memory during power loss, leaving dynamic data objects in volatile memory. Since static data objects are unchanged or rarely changed, they are already in a stable state suitable for persistence. This extraction approach significantly reduces the amount of data that needs to be copied during the short voltage window, thereby improving data retention reliability without excessive copy time.
Solution Approach 2:
The patent applies partial action by copying only static data objects to non-volatile memory during power loss, rather than copying all data. This partial copying approach is sufficient to maintain system reliability because static data objects represent the stable, persistent state of the system, while dynamic data objects can be regenerated or restored from their sources after power recovery.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
The subject matter disclosed herein describes a method of allocating and persisting memory in an industrial controller without requiring a battery backup or a large capacitive storage system. Each data object is identified as static or dynamic. Static objects are further classified by whether frequent access of that data object is required. Each of the data objects is stored in non-volatile memory. The dynamic data objects and static data objects requiring frequent access are stored in volatile memory. A record of static data objects is maintained in non-volatile memory and a record of dynamic data objects is maintained in volatile memory. Upon power loss, the present value of each dynamic data object is copied to non-volatile memory. When power is restored, the values of both the dynamic data objects and the static data objects that require frequent access at run-time are copied from non-volatile memory to volatile memory.