Automatic fault detection and recovery system which provides stability and continuity of operation in an industrial multiprocessor control

• US 4,377,000 A
• Filed: 06/18/1980
• Issued: 03/15/1983
• Est. Priority Date: 05/05/1980
• Status: Expired due to Term

• Current Assignee: Westinghouse Process Control, Inc. (Emerson Electric Company)
• Original Assignee: Westinghouse Electric Corporation (Paramount Global (f/k/a ViacomCBS Inc.))
• Inventors: Staab, Carl J.
• Primary Examiner(s): ATKINSON, CHARLES
• Application Number:
  US06/160,411
• Time in Patent Office: 1,000 Days
• Field of Search: 371/11, 364/101, 364/119, 364/200, 364/900, 364/184, 364/186, 364/187
• US Class Current: 714/55
• CPC Class Codes: G06F 11/1417 Boot up procedures
• CPC Class Codes: G06F 11/2025 using centralised failover control functionality
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component

Abstract

A multiprocessor computer control system for a process includes at least two central processors and respectively associated input/output systems. Control system faults are detected and a bootstrap microprocessor is triggered into its fault detector and recovery mode. The output control signals are placed on and central processors are stopped and restarted and then it is determined which if any of them is to be inoperative. A reconfigured control system is defined with the inoperative unit(s) excluded. The microprocessor switches the input/output systems as necessary to structure the control system in reconfigured form so that process control can resume without disturbance.

Claims

1. A multiprocessor computer control system comprising at least a first central processor and a second central processor, a plurality of sensors coupled to equipment in a production system, means for connecting said sensors to first and second process input/output systems, a plurality of controllable devices for operating the production system equipment, switch means including at least first switch means for coupling said first process input/output system to said first central processor and second switch means for coupling said second process input/output system to said second central processor, and means for detecting control system faults and providing control system recovery in reconfigured form as necessary with substantially no process disturbance, said fault detecting and recovery means including a system bootstrap microprocessor controller coupled to each said central processor, means forming a part of said microprocessor controller and said central processors for sensing any of a plurality of control system failures and for thereby triggering said microprocessor controller into a recovery mode of operation, means for stopping said central processors and placing process control signals in a hold state upon command by said microprocessor controller, means for detecting the integrity of each central processor, said central processors including means for defining reconfigured central processor hardware so that process control duties are loaded only on operational central processor hardware, means including said microprocessor controller for operating said switch means as required so that only operational central processor hardware is coupled to the process equipment, means for restarting said system after completion of control system reconfiguration if any, and means for returning said microprocessor controller to a monitor mode after completion of the fault detection and recovery mode.

2. A system as set forth in claim 1 wherein each of said central processors has associated therewith an input/output system which is switched on a unit basis when reconfiguration is required.

3. A system as set forth in claim 1 wherein said microprocessor controller includes means for registering and executing any input/output switching action determined to be required by the central processors.

4. A system as set forth in claim 1 wherein said system fault detecting means includes means forming a part of said microprocessor controller for responding to a manual demand signal for fault detection operation and for thereby triggering said recovery mode of operation.

5. A system as set forth in claim 1 wherein said microprocessor controller includes sequencing means for stepping the whole system through its fault detector and recovery mode of operation once such mode is triggered, said microprocessor controller coupled to said central processors and said input/output switching means to provide for execution of various steps in the fault detection and recovery sequence.