CN104636219A - Synchronous event response method for three-machine hot backup real-time system - Google Patents

Abstract

The invention provides a method for synchronous event response of a three-machine hot backup real-time system. The three-machine hot backup real-time system includes three independently operated single machines, which communicate and exchange data with each other at regular intervals in each control cycle. A stand-alone machine runs the following steps: S1. Query whether there is an event that needs to be processed; S2. When the event that needs to be processed arrives, record the event status of the machine; S3. Exchange information with other stand-alone machines after recording the event state, and check whether other stand-alone machines There is event status; S4, choose to execute: S41, if in step S3 all three stand-alone machines are found to have event status, then start to respond; or S42, if in step S3 it is found that there are stand-alone machines without event status, then wait for a control cycle. The invention decides whether to use data information according to the status of received events through information exchange between single machines, and waits for a beat when the information is not synchronized, thereby improving the reliability and completeness of the hot backup system.

Description

A method for synchronous event response in real-time system with three-machine hot backup

technical field

The invention relates to a multi-machine hot backup control system computer on a spacecraft, in particular to a method for synchronous event response based on a three-machine hot backup real-time system.

Background technique

The system control scheme of transferring from the moon to the orbit and returning to the earth accurately has the characteristics of high technical difficulty and high requirements for reliability and safety. Among them, the guidance, navigation and control computer application software, as an important part of the guidance, navigation and control subsystem, plays a decisive role in the success of the return program.

The returner control computer adopts a three-machine hot backup structure. During the return and reentry process, there will be various asynchronous events that need to be processed by the software at any time, such as the injection of data and instructions on the ground, the change of the structure of the cabin, and the forwarding of various data by the digital tube broadcast. Due to the interference of the flight environment, such as severe vibration, entering the electromagnetic blind zone, etc., it is difficult for the stand-alone software of each system in the three-pole hot backup structure to maintain command-level synchronization during operation. The difference in the above results in the event that the three machines are not synchronized.

In order to improve the reliability and security of the software, it is necessary to respond synchronously to the asynchronous events during the return and re-entry process to ensure the synchronous operation of the three machines, thereby improving the reliability of the software operation and ensuring the successful completion of the scientific research of returning to the earth from the lunar orbit test task.

Contents of the invention

The technical problem to be solved by the present invention is: to provide a method for time synchronization response based on the real-time system of the three-machine hot backup, to perform synchronous response processing on the asynchronous events in the return and re-entry process, to ensure the synchronous operation of the three machines, thereby improving the software operational reliability.

The technical solution of the present invention is: a method for synchronous event response based on the three-machine hot backup real-time system, the three-machine hot backup real-time system includes three independently operated stand-alone machines, and the three stand-alone machines communicate regularly with each other in each control cycle, Exchanging data, each control cycle and each stand-alone operation includes the following steps: S1. Query whether there is an event that needs to be processed; S2. When the event that needs to be processed arrives, record the event status of the machine; S3. After recording the event status, communicate with other Single machines exchange information with each other, and check whether other single machines have an event state; S4, choose to perform one of the following two actions: S41, if in step S3, it is found that all three single machines have an event state, then start to respond to events and process data; Or S42, if it is found in step S3 that one or two stand-alone machines have no event status, wait for one control cycle.

It also includes, S5. If step S42 was executed in the last control cycle and one control cycle has been waited, respond to events and process data in the current cycle.

Further, there is a waiting count in the stand-alone machine, and the waiting count is incremented by 1 when waiting for one control cycle in step S42, and returns to zero after step S5.

Compared with the prior art, the present invention has the advantages of: through the information exchange between the stand-alone machines, it is decided whether to use the data information according to the state of the received event, and it waits for a beat when the information is not synchronized, so as to improve the reliability and reliability of the hot backup system. completeness.

Description of drawings

FIG. 1 shows a step diagram of a method for synchronizing event response based on a three-machine hot backup real-time system of the present invention.

Detailed ways

The invention provides a method for responding to a real-time system synchronization event in a three-machine hot backup system of an active rendezvous and docking spacecraft, which comprehensively considers the fault-tolerant state and determines whether to use data information according to the state of the received event. Only use the received data when it is ensured that all three machines have received the data. The invention comprehensively considers the state information of the three machines, and improves the reliability and completeness of the hot backup system. This method is suitable for a multi-machine hot backup redundant system that requires high timing but has complex functions and a large amount of fault-tolerant information.

Referring to shown in Fig. 1, the three-machine hot backup real-time system of the present invention is installed on the returning aircraft, and includes three identical stand-alone machines, and the three stand-alone machines communicate regularly and exchange data between each control cycle (every beat) . Each stand-alone machine in the three-machine hot backup system performs the following steps:

(1) Query whether a random event occurs.

Query whether there are random events coming that need to be processed. If an event occurs, it is not processed immediately, but the event status of the machine is recorded first.

(2) The state of the three machines switching.

The stand-alone machines exchange information with each other to check whether there is an event arrival status.

(3) Perform synchronous processing

If all three machines find a new random event, they will start to respond to the event, respond to uplink information, process data, and set the data processing completion flag to be valid; otherwise, if one or two machines do not receive it, wait for the count to increase. 1. Delay the response by one shot to wait for the single machine that does not query the asynchronous event in this shot, so as to ensure that the three machines process the event synchronously, and the overall response delay to the event will not be very long, which still meets the system requirements. If it is judged that it has been waiting for one shot, the data can be processed at this time, or it can be waited until all the stand-alone machines have event status; after the data processing is completed, the value of the waiting count is cleared.

Tests show that after applying the algorithm of the present invention, the timing synchronization of the three-computer hot backup real-time system is good, the safety and reliability of system operation are improved, and the stability of the system under complex timing is guaranteed.

The content that is not described in detail in the specification of the present invention belongs to the well-known technology of those skilled in the art.

Claims (3)

1. a method for three-machine hot backup real-time system synchronous event response, described three-machine hot backup real-time system comprises three stand-alone machines of independent operation, it is characterized in that, three described stand-alone machines are mutually timing in each control cycle Communication and data exchange, each control cycle and each stand-alone operation include the following steps: S1. Query whether there is an event that needs to be processed; S2. When the event that needs to be processed arrives, record the event status of the machine; S3. After recording the event state, exchange information with other stand-alone machines, and inquire whether other stand-alone machines have the event state; S4. Choose to perform one of the following two actions: S41. If it is found in step S3 that all three stand-alone machines have the event status, start to respond to the event and process data; or S42. If it is found in step S3 If one or two single machines do not have the event status, wait for a control cycle. 2. according to the method for three-computer hot backup real-time system synchronous event response of claim 1, it is characterized in that, also comprising, S5. If step S42 was executed in the last control cycle and one control cycle has been waited, respond to the event and process data in the current cycle. 3. according to the method for the three-machine hot backup real-time system synchronous event response of claim 1, it is characterized in that, The stand-alone machine is provided with a waiting count, which is incremented by 1 when waiting for one control cycle in the step S42, and returns to zero after the step S5.