Safety Computing Device Failure Detection via Memory Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing safety controllers require complex and costly duplex circuit configurations to detect both hardware and software failures, which complicates the system and increases costs.

Innovation Solution

A safety computing device with a simplex circuit configuration that includes a processor and memory with separate memory areas for input and output data, using redundancy codes and computation diagnosis units to detect failures without a duplex hardware structure.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a duplex circuit configuration is used to detect hardware and software failures, then the reliability of failure detection is improved, but the device complexity and cost increase

Engineering Contradiction:
Improvefailure detection capabilityVSAvoidcircuit configuration complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates a virtual copy of the computation process by storing input data in two different memory areas and executing the same program twice. This software-based copying approach replaces the need for physical duplex hardware circuits, achieving failure detection capability while maintaining a simple circuit configuration. The two execution results are then collated to detect software failures, and hardware failures are detected through computation diagnosis.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical/duplex hardware circuit system with a software-based computation system. Instead of using duplicate physical circuits for comparison, the invention uses a single processor to execute computations twice with different memory areas and collates the results through software logic. This substitution dramatically simplifies the hardware while maintaining the ability to detect both hardware and software failures.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If a duplex circuit configuration is used to detect hardware and software failures, then the reliability of failure detection is improved, but the cost increases

Engineering Contradiction:
Improvefailure detection capabilityVSAvoidmanufacturing cost
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The patent creates a virtual copy of the computation process by storing input data in two different memory areas and executing the same program twice. This software-based copying approach replaces the need for physical duplex hardware circuits, achieving failure detection capability while maintaining a simple circuit configuration. The two execution results are then collated to detect software failures, and hardware failures are detected through computation diagnosis.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical/duplex hardware circuit system with a software-based computation system. Instead of using duplicate physical circuits for comparison, the invention uses a single processor to execute computations twice with different memory areas and collates the results through software logic. This substitution dramatically simplifies the hardware while maintaining the ability to detect both hardware and software failures.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Device complexity

If a simplex circuit configuration is used to simplify the system, then the device complexity and cost are reduced, but the capability to detect both hardware and software failures is compromised

Engineering Contradiction:
Improvecircuit configuration simplicityVSAvoidfailure detection capability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent segments the memory into two distinct memory areas (first memory area and second memory area) with different addresses. This segmentation allows the system to store the same input data in two separate locations, enabling the execution control unit to perform duplicate computations independently. The segmentation of memory space is key to achieving both simplicity and comprehensive failure detection capability in the simplex configuration.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The execution control unit performs periodic duplicate computations by executing the same program process twice with input data from different memory areas. This periodic execution of the same computation with different memory references enables the system to detect software failures through result collation while maintaining a simple hardware structure. The periodic nature of this dual execution is what provides the failure detection capability without requiring complex hardware.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentUS9417943B2Safety computing device, safety input device, safety output device, and safety controller
Publication Date: 2016.08.16 MITSUBISHI ELECTRIC CORP
  • US9417943B2 patent drawing
  • US9417943B2 patent drawing
  • US9417943B2 patent drawing

AI summary

A safety computing device includes a processor and a memory. The memory includes a first memory area and a second memory area having an address different from the first memory area. The processor includes an execution control unit performing a first process including the program process on input data written in the first memory area, and a second process including the program process on input data written in the second memory area and addition of redundancy code to output data written in the second memory area, a result collating unit collating output data to which redundancy code is added in the first and second processes, a computation diagnosis unit diagnosing presence or absence of failure in the processor and the memory, and an abnormality processing unit that, when an abnormality is detected by at least one of redundancy check, collation, and diagnosis, stops outputting output data.