Large ship monitoring system and distributed data service method thereof

Abstract

The invention provides a large ship monitoring system and a distributed data service method thereof, and the method comprises the steps: enabling collection units belonging to the same business to be divided into a group, and distributing the collection units to a work station which is mainly responsible for the business; unified node numbering and unicast IP address definition are carried out on all workstations, an independent number and an independent IP are allocated to each workstation, and the number is used as an identity identification code of the workstation and is used for carrying out identity identification in data synchronization; the middle layer adopts a network layer structure based on the redundant Ethernet, data exchange is realized among workstations by adopting udp multicast protocol messages, and meanwhile, a sending and confirming mechanism is added in a protocol application layer. According to the scheme provided by the invention, the ship monitoring system data service method which is more reliable, more efficient and larger in scale is realized.

Description

Technical Field

[0001] The invention belongs to the technical field of large ship monitoring systems, and in particular relates to a large ship monitoring system and a distributed data service method thereof. Background Art

[0002] At present, with the continuous deepening of automation, informatization and intelligent technologies in the industrial field, ship systems have also undergone profound changes in digital transformation, especially for large ocean-going ships in equipment operation, management, maintenance and other aspects. Because the supporting systems and equipment are large in scale and complex in functions, they are usually configured as systematic automated monitoring equipment to meet the needs of operating managers for centralized monitoring and control of the status of various operating equipment, sensor devices, etc. of the ship.

[0003] The ship monitoring system generally includes a three-layer system: the bottom layer is the data acquisition equipment on the field side, which is used to collect and convert data signals from various system devices; the middle layer is the network equipment, which provides reliable network services for all system devices, which can be serial networks, various bus networks or Ethernet; the upper layer is the application equipment, including workstations and operation panels at various operating positions, which are used to realize user interaction for various application functions.

[0004] The ship monitoring system has become the neural network of the entire ship. In the context of the industry's development towards intelligence, unmanned operations, and fewer people operating the system, a ship monitoring system with high stability, high reliability, and high efficiency will provide more solid data support and control responses for intelligent and unmanned systems.

[0005] At present, the ship monitoring system mainly adopts two methods for data collection and transmission: server / client type and direct connection type. The bottom layer usually adopts distributed data collection mode to connect signals and external interfaces by multiple acquisition units. For the server / client mode, an independent system-level server will be configured. The network interface of the server first receives, analyzes, processes or issues instructions for all distributed acquisition devices through network devices, and then the workstations at various operating positions apply for data services from the server through the network as clients. In this way, the server becomes the core of the entire system and undertakes the data request reception of all distributed acquisition units and the data request reply of all workstations. Therefore, the server shoulders the heavy task of data interaction processing, and the stable operation of the server directly determines the reliability of the entire system; for the direct connection type, the configuration of the system server is cancelled, and each distributed acquisition unit is used as a server to provide the service of the accessed data. Each workstation, as a client, requests data services from each acquisition unit through network devices. This method places high requirements on the performance of the acquisition unit and also limits the scale of the monitoring system. With the increase in the number of clients and acquisition units, the data requests of the network will increase exponentially. Especially for large ship monitoring systems, there are too many nodes and the restrictions are particularly obvious. At the same time, the independent operation mode brings disadvantages to the consistency of data calculation, conversion and judgment among workstations.

[0006] Therefore, how to provide a large-scale ship monitoring system and a distributed data service method thereof has become a technical problem that urgently needs to be solved in this field. Summary of the invention

[0007] The purpose of the present invention is to provide a large ship monitoring system and a distributed data service method thereof.

[0008] According to a first aspect of the present invention, there is provided a distributed data service method for a large ship monitoring system, comprising: grouping collection units belonging to the same service into a group and assigning them to a workstation mainly responsible for the service;

[0009] A unified node number and unicast IP address are defined for all workstations, and each workstation is assigned an independent number and IP address. The number serves as the workstation's identification code for identification during data synchronization.

[0010] The middle layer adopts a network layer structure based on redundant Ethernet, and UDP multicast protocol messages are used between workstations to achieve data exchange. At the same time, a sending and confirmation mechanism is added to the protocol application layer.

[0011] Optionally, the method further comprises:

[0012] In addition to being allocated to a workstation that is primarily responsible for the business of the collection unit, each collection unit is allocated to a predefined workstation according to task priority, based on the principle that higher-priority tasks have higher priority.

[0013] Optionally, the middle layer adopts a network layer structure based on redundant Ethernet including:

[0014] All interactive data is connected through dual redundant Ethernet networks. The two redundant networks are completely independent and do not adopt a master-slave mode. Instead, they work simultaneously. The sender sends messages on both networks at the same time. The receiver selects the messages based on their priority order. Messages that arrive first are parsed, while messages that arrive later are discarded.

[0015] Optionally, the UDP multicast protocol message includes: a heartbeat message, a data message, a reception confirmation message and a retransmission request message.

[0016] Optionally, the distributed data synchronization heartbeat message includes:

[0017] After each workstation is turned on, it broadcasts and sends heartbeat messages on the network at a regular interval to mark its own online status; this mainly includes the heartbeat message identification number, site number, timestamp, working status, number of data sets currently responsible, the latest message sequence number of each data set currently sent for the acquisition message, the latest message sequence number of each data set currently sent for the control message, and a check code; at the same time, it receives heartbeat messages from predefined workstations on the network to determine the current operating status of each workstation in the entire system; based on the operating status and task priority settings, it determines its own communication task with the acquisition unit and executes it; at the same time, it verifies the latest data message sequence number in the target heartbeat message with the latest message received from the target.

[0018] Optionally, the data message includes: a collection data message and a control data message;

[0019] The data collection message is a change-triggered and time-triggered mechanism; it is used for the workstation to send the status data of the field side uploaded from the collection unit to the predefined workstation;

[0020] The interface description file of the acquisition unit uniformly divides the interface numbers of different acquisition units and also serves as the data set number of the interface. All data in the data set are pre-arranged according to the data type and address order. The acquisition data message will be sent in groups based on the data set. The sent data includes input analog and digital signals. If a single packet cannot send all the data of the interface, the data of the interface will be divided into several data fields according to the data address order in the interface description file and sent in sequence. It mainly includes data message identification number, sending end site number, timestamp, data set number, data field number, data set acquisition message sequence number, data length, data field content and check code.

[0021] The control data message is an event-triggered mechanism; the workstation for issuing instructions sends the instruction data to the predetermined workstation, and the priority workstation issues the control instruction data to the field side;

[0022] The control data message is also sent in groups based on the data set. The sent data includes output analog and digital signals, and is also sent as a whole or in parts according to the data volume. It mainly includes the data message identification number, the sending site number, the timestamp, the data set number, the data field number, the control message sequence number of the data set, the data length, the data field content and the check code.

[0023] Optionally, the reception confirmation message is:

[0024] When the receiving workstation completes the reception and verification of a data message, it returns confirmation information to the sender to indicate that the data message has been successfully received. The structure of the reception confirmation message mainly includes: confirmation message identification number, confirming party site number, timestamp, confirmed data set number, confirmed message sequence number and check code.

[0025] Optionally, the resend request message is:

[0026] When the receiving workstation determines that there is a missing data packet based on the latest message or heartbeat message received, it applies to the sender for resending the latest message or heartbeat message according to the missing packet situation; the structure of the resend request message mainly includes: resend request message identification number, request site number, timestamp, requested data set number, requested message sequence number and check code.

[0027] Optionally, the UDP multicast protocol is a custom multicast protocol, that is, the reliable transmission of data is ensured by confirming the reception of the sent message and retransmitting the message, specifically including:

[0028] When the sender workstation determines that the data domain in the data set it manages has changed or the sending cycle has been reached, it generates a new message of the data set; the message sequence number in the new message is incremented and updated each time it is sent, and the message content is stored in the cache while sending the message; when the receiver identifies it as a collection message based on the message identification number and verifies that the message format is complete, if the verification is passed, it parses it according to the data address sequence of the interface description file and updates the local database, and sends a reception confirmation message at the same time; when the sender confirms that all workstations have completed the reception based on the confirmation message, the message content is released in the cache;

[0029] Packet loss is determined when the following three situations occur:

[0030] The first case: When the workstation receives a data set with non-continuous message sequence numbers;

[0031] The second situation: the latest data set message number in the received heartbeat message is inconsistent with the latest message number of the data set parsed by the station;

[0032] The third situation: when receiving the message number in the confirmation message of the relevant data set sent by the predefined workstation, the workstation has not received the collection message.

[0033] According to a second aspect of the present invention, there is provided a large ship monitoring system, comprising the distributed data service method for the large ship monitoring system as described in any one of the first aspects of the present invention.

[0034] The beneficial effects brought by the present invention are as follows:

[0035] It can be seen from the above scheme that the embodiment of the present invention provides a large-scale ship monitoring system and its distributed data service method, which has the following beneficial effects: from the perspective of system structure, it breaks the limitation of relying on the server as the core node, and realizes a flexible data service method through distributed interaction relying on reliable multicast, task classification, and redundant Ethernet technology. It does not rely on any single point, and the single point failure of any workstation can be replaced by other workstations, which greatly improves the reliability of the system. The data interaction on the field side can also significantly improve the efficiency of data interaction. At the same time, it also breaks the limitation of the direct connection between the workstation and the acquisition unit on the system scale and performance, forming a flexible architecture that can be cut and expanded. Under the guarantee of superior reliability and flexibility, it can provide the data terminal foundation for the application requirements of interoperability, mutual backup and distributed control between different workstations of the upper-level application, and will also more effectively enhance the platform integration capability of the large-scale ship monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of a large ship monitoring system provided according to an embodiment.

[0037] Description of reference numerals:

[0038] 1. The first workstation; 2. The second workstation; 3. The third workstation; 4. The fourth workstation; 5. The first acquisition unit; 6. The second acquisition unit; 7. The third acquisition unit; 8. The fourth acquisition unit; 9. The first ring network switch; 10. The second ring network switch; 11. The third ring network switch; 12. The fourth ring network switch. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] According to a first aspect of the present invention, there is provided a distributed data service method for a large ship monitoring system, comprising: grouping collection units belonging to the same service into a group and assigning them to a workstation mainly responsible for the service;

[0041] A unified node number and unicast IP address are defined for all workstations, and each workstation is assigned an independent number and IP address. The number serves as the workstation's identification code for identification during data synchronization.

[0042] The middle layer adopts a network layer structure based on redundant Ethernet, and UDP multicast protocol messages are used between workstations to achieve data exchange. At the same time, a sending and confirmation mechanism is added to the protocol application layer.

[0043] In some embodiments, the method further comprises:

[0044] In addition to being allocated to a workstation that is primarily responsible for the business of the collection unit, each collection unit is allocated to a predefined workstation according to task priority, based on the principle that higher-priority tasks have higher priority.

[0045] All workstations have unified configuration description files for the communication task priority of each acquisition unit and interface description files for each acquisition unit to ensure coordination and consistency among all workstations. At the same time, each acquisition unit has only one workstation to interact with it. When one of the workstations is in an abnormal state, the acquisition unit it is in charge of will be continued by other workstations.

[0046] In some embodiments, the middle layer adopts a network layer structure based on redundant Ethernet including:

[0047] All interactive data is connected through dual redundant Ethernet networks. The two redundant networks are completely independent and do not adopt a master-slave mode. Instead, they work simultaneously. The sender sends messages on both networks at the same time. The receiver selects the messages based on their priority order. Messages that arrive first are parsed, while messages that arrive later are discarded.

[0048] This solution can better provide network stability guarantee for distributed data synchronization, and will not affect system operation in the event of any network failure or any node network card failure.

[0049] In some embodiments, the UDP multicast protocol message includes: a heartbeat message, a data message, a reception confirmation message and a retransmission request message.

[0050] The heartbeat message of distributed data synchronization includes:

[0051] After each workstation is turned on, it broadcasts and sends heartbeat messages on the network at a regular interval to mark its own online status; this mainly includes the heartbeat message identification number, site number, timestamp, working status, number of data sets currently responsible, the latest message sequence number of each data set currently sent for the acquisition message, the latest message sequence number of each data set currently sent for the control message, and a check code; at the same time, it receives heartbeat messages from predefined workstations on the network to determine the current operating status of each workstation in the entire system; based on the operating status and task priority settings, it determines its own communication task with the acquisition unit and executes it; at the same time, it verifies the latest data message sequence number in the target heartbeat message with the latest message received from the target.

[0052] The data message includes: a collection data message and a control data message;

[0053] The data collection message is a change-triggered and time-triggered mechanism; it is used for the workstation to send the status data of the field side uploaded from the collection unit to the predefined workstation;

[0054] The interface description file of the acquisition unit uniformly divides the interface numbers of different acquisition units and also serves as the data set number of the interface. All data in the data set are pre-arranged according to the data type and address order. The acquisition data message will be sent in groups based on the data set. The sent data includes input analog and digital signals. If a single packet cannot send all the data of the interface, the data of the interface will be divided into several data fields according to the data address order in the interface description file and sent in sequence. It mainly includes data message identification number, sending end site number, timestamp, data set number, data field number, data set acquisition message sequence number, data length, data field content and check code.

[0055] The control data message is an event-triggered mechanism; the workstation for issuing instructions sends the instruction data to the predetermined workstation, and the priority workstation issues the control instruction data to the field side;

[0056] The control data message is also sent in groups based on the data set. The sent data includes output analog and digital signals, and is also sent as a whole or in parts according to the data volume. It mainly includes the data message identification number, the sending site number, the timestamp, the data set number, the data field number, the control message sequence number of the data set, the data length, the data field content and the check code.

[0057] Because the data address is predefined, the receiver can directly parse each data packet based on it without having to wait for the split data to be received completely before parsing, which greatly improves the data parsing efficiency.

[0058] The reception confirmation message is:

[0059] When the receiving workstation completes the reception and verification of a data message, it returns confirmation information to the sender to indicate that the data message has been successfully received. The structure of the reception confirmation message mainly includes: confirmation message identification number, confirming party site number, timestamp, confirmed data set number, confirmed message sequence number and check code.

[0060] The resend request message is:

[0061] When the receiving workstation determines that there is a missing data packet based on the latest message or heartbeat message received, it applies to the sender for resending the latest message or heartbeat message according to the missing packet situation; the structure of the resend request message mainly includes: resend request message identification number, request site number, timestamp, requested data set number, requested message sequence number and check code.

[0062] The UDP multicast protocol is a custom multicast protocol, which ensures reliable data transmission by confirming the receipt of sent messages and retransmitting them, including:

[0063] When the sender workstation determines that the data domain in the data set it manages has changed or the sending cycle has been reached, it generates a new message of the data set; the message sequence number in the new message is incremented and updated each time it is sent, and the message content is stored in the cache while sending the message; when the receiver identifies it as a collection message based on the message identification number and verifies that the message format is complete, if the verification is passed, it parses it according to the data address sequence of the interface description file and updates the local database, and sends a reception confirmation message at the same time; when the sender confirms that all workstations have completed the reception based on the confirmation message, the message content is released in the cache;

[0064] Packet loss is determined when the following three situations occur:

[0065] The first case: When the workstation receives a data set with non-continuous message sequence numbers;

[0066] The second situation: the latest data set message number in the received heartbeat message is inconsistent with the latest message number of the data set parsed by the station;

[0067] The third situation: when receiving the message number in the confirmation message of the relevant data set sent by the predefined workstation, the workstation has not received the collection message.

[0068] After the receiving workstation determines that the message is lost or the received message verification fails, it sends a retransmission request to the network. The sending workstation loads the message from the cache and resends it according to the data set number and message sequence number in the retransmission request message.

[0069] If the sender has sent the latest message sequence number through the heartbeat message but still has not received the confirmation message from other workstations and has not received the resend request message, it will automatically resend it once and then release the message from the cache.

[0070] For the resent message received by the receiving workstation, if an updated message has been received, the data in the resent message will be inserted into the history record according to the timestamp; if the resent message is the latest message, the current local real-time database will be updated according to the new message and recorded in the history library at the same time.

[0071] According to a second aspect of the present invention, there is provided a large ship monitoring system, comprising the distributed data service method for the large ship monitoring system described in any one of the first embodiments.

[0072] like Figure 1 As shown, the upper layer of the system is configured with 4 workstation computers as the operating platform of the distributed data service and application system. The middle layer is divided into two groups by 4 ring network switches, including the first ring network switch 9, the second ring network switch 10, the third ring network switch 11 and the fourth ring network switch 12, to build a dual redundant ring network to provide a reliable data link. The bottom layer is composed of 4 sets of data acquisition units for field data acquisition and control. All workstations and acquisition units are connected to the dual redundant ring network through two independent network interfaces. It is assumed that each acquisition unit includes 500 bytes of acquisition data, and the specific implementation method is as follows.

[0073] Assign station numbers 1 / 2 / 3 / 4 to the first workstation 1, B, C, and D, and configure the task priorities for data interaction between the workstation and the acquisition unit as follows:

[0074] First collection unit 5: first workstation 1> second workstation 2> third workstation 3> fourth workstation 4;

[0075] Second collection unit 6: second workstation 2 > third workstation 3 > fourth workstation 4 > first workstation 1;

[0076] The third collection unit 7: the third workstation 3> the fourth workstation 4> the first workstation 1> the second workstation 2;

[0077] Fourth collection unit 8: fourth workstation 4 > first workstation 1 > second workstation 2 > third workstation 3.

[0078] According to the above configuration, when all workstations and collection units are running online, the workstations are divided into first workstation 1→first collection unit 5, second workstation 2→second collection unit 6, third workstation 3→third collection unit 7, and fourth workstation 4→fourth collection unit 8.

[0079] The data of each acquisition unit is taken as an independent data set, namely data set 1 / 2 / 3 / 4. Each data set contains 500 bytes. Assuming that each data message sends a maximum of 200 bytes, each data set is divided into three data fields.

[0080] Message structure design:

[0081] (1) The message identifier uses two bytes and is designed as follows.

[0082] Heartbeat message identifier: 011A;

[0083] Collected data message identifier: 021A;

[0084] Control data message identifier: 021B;

[0085] Confirmation message identifier: 031A;

[0086] Resend request message identifier: 041A.

[0087] (2) The timestamp uses the Unix format, with a length of 6 bytes to represent the UTC time in milliseconds;

[0088] (3) The working status in the heartbeat message is represented by 1 byte, 01 starting, 02 running, 03 fault / stop;

[0089] (4) The number of data sets that the workstation is currently responsible for in the heartbeat message is represented by 1 byte;

[0090] (5) The site number, data set number, and data field number are all represented by 1 byte;

[0091] (6) Data length is represented by 1 byte;

[0092] (7) The message sequence number is represented by 4 bytes. When the sequence number overflows, it will automatically return to zero and start counting again.

[0093] (8) The check code uses CRC-32 cyclic redundancy check, and uses 0x04C11DB7 to generate the polynomial. The result consists of 4 bytes.

[0094] First, each workstation sends the following heartbeat message to the network at regular intervals after it is online:

[0095]

[0096] As shown above, it is a heartbeat message in the running state with the site number of 2 (the second workstation 2). The message is divided into 8 parts. The first part is the heartbeat message identifier 011A, the second part is the site number 2, the third part is the timestamp UTC: 2024-08-30 18:04:20 497ms, the fourth part is the workstation running status (running), the fifth part is that the current workstation is only responsible for the collection task of one data set, the sixth part is the responsible data set number 2, the seventh part is the latest data number sent corresponding to the data set number 14580, and the eighth part is the CRC-32 check code.

[0097] When the workstation completes data acquisition with the acquisition unit, it performs data synchronization and sends data messages according to the cycle and data domain changes. An example is as follows:

[0098]

[0099] As shown above, the collection data message sent by the site number is 2 (the second workstation 2). The message is divided into 9 parts. The first part is the collection data message identifier 021A, the second part is the site number 2, the third part is the timestamp UTC: 2024-08-30 18:04:20 520ms, the fourth part is the sent data set number 2, the fifth part is the data field number 03 of the corresponding data set, the sixth part is the message number 14581 of the data set message, the seventh part is the data length 100 bytes, the eighth part is the actual 100 bytes of data, and the ninth part is the CRC-32 check code.

[0100] When other workstations receive data synchronization messages from other workstations, they parse and verify them and then send a reception confirmation message to the network to the sender. For example:

[0101]

[0102] As shown above, the confirmation message replied by the site number is 4 (the fourth workstation 4), the message is divided into 6 parts, the first part is the confirmation message identifier 031A, the second part is the confirmation site number 4, the third part is the timestamp UTC: 2024-08-30 18:04:20 600ms, the fourth part is the data set 2 of the confirmation reception message, the fifth part is the sequence number 14581 of the confirmation reception message, and the sixth part is the CRC-32 check code.

[0103] If a workstation determines that a packet is lost or a message verification error occurs, it will send a retransmission request message to the network, as shown below:

[0104]

[0105] As shown above, it is a retransmission request message sent by site number 1 (first workstation 1). The message is divided into 6 parts. The first part is the confirmation message identifier 041A, the second part is the confirmation site number 1, the third part is the timestamp UTC: 2024-08-30 18:04:20 640ms, the fourth part is the data set 2 of the request retransmission message, the fifth part is the request retransmission message sequence number 14581, and the sixth part is the CRC-32 check code.

[0106] Special situation handling:

[0107] When the system is initially started, the first workstation to start running (without receiving heartbeats from other workstations) collects data according to the distribution of task priorities. When subsequent workstations are started, they first receive network data based on the heartbeat situation in the network, complete the startup state after synchronizing the data messages of all workstations, and hand over tasks with other workstations according to the task priority, and perform collection and data publishing work. During the handover process, the data set message sequence number is continuously updated.

[0108] During operation, when a workstation loses its heartbeat or sends a heartbeat status of failure / stop, other workstations take over the collection unit that the workstation is responsible for according to the task priority.

[0109] When a collection unit fails or is unable to receive and send data, the workstation responsible for interacting with it will send a message to other workstations indicating that the corresponding data set of the collection unit is in an "unknown" state. Other workstations will mark and prompt the data in the data set as "unknown" in their respective data services, and issue a communication failure prompt at the same time.

[0110] Dual-ring Ethernet has extremely reliable performance. A single-point interruption in any ring will not affect the data exchange of that ring. The overall effectiveness of any ring can still support the operation of the system in the other ring.

[0111] In summary, the solution proposed in the present invention breaks the limitation of relying on the server as the core node from the perspective of system structure, and realizes a flexible data service method through distributed interaction relying on reliable multicast, task classification, and redundant Ethernet technology. It does not rely on any single point, and any single point failure of any workstation can be replaced by other workstations, which greatly improves the reliability of the system. Relying on each workstation to complete the data interaction on the field side can also significantly improve the efficiency of data interaction. At the same time, it also breaks the limitation of the direct connection between the workstation and the acquisition unit on the scale and performance of the system, forming a flexible architecture that can be cut and expanded. Under the guarantee of superior reliability and flexibility, it can provide the data terminal foundation for the application requirements of interoperability, mutual backup and distributed control between different workstations of the upper-level application, and will also more effectively enhance the platform integration capability of large-scale ship monitoring systems.

[0112] The above are preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A distributed data service method for a large ship monitoring system, characterized in that: include: Grouping the collection units belonging to the same business into one group and assigning them to the workstations mainly responsible for the business; A unified node number and unicast IP address are defined for all workstations, and each workstation is assigned an independent number and IP address. The number serves as the workstation's identification code for identification during data synchronization. The middle layer adopts a network layer structure based on redundant Ethernet, and UDP multicast protocol messages are used between workstations to achieve data exchange. At the same time, a sending and confirmation mechanism is added to the protocol application layer.

2. The distributed data service method for a large ship monitoring system according to claim 1, characterized in that: The method further comprises: In addition to being allocated to a workstation that is primarily responsible for the business of the collection unit, each collection unit is allocated to a predefined workstation according to task priority, based on the principle that higher-priority tasks have higher priority.

3. The distributed data service method for a large ship monitoring system according to claim 1, characterized in that: The middle layer adopts a network layer structure based on redundant Ethernet, including: All interactive data is connected through dual redundant Ethernet networks. The two redundant networks are completely independent and do not adopt a master-slave mode. Instead, they work simultaneously. The sender sends messages on both networks at the same time. The receiver selects the messages based on their priority order. Messages that arrive first are parsed, while messages that arrive later are discarded.

4. The distributed data service method for a large ship monitoring system according to claim 1, characterized in that: The UDP multicast protocol message includes: a heartbeat message, a data message, a reception confirmation message and a resend request message.

5. The distributed data service method of a large ship monitoring system according to claim 4, characterized in that: The heartbeat message of distributed data synchronization includes: After each workstation is turned on, it broadcasts and sends heartbeat messages on the network at a regular interval to mark its own online status; this mainly includes the heartbeat message identification number, site number, timestamp, working status, number of data sets currently responsible, the latest message sequence number of each data set currently sent for the acquisition message, the latest message sequence number of each data set currently sent for the control message, and a check code; at the same time, it receives heartbeat messages from predefined workstations on the network to determine the current operating status of each workstation in the entire system; based on the operating status and task priority settings, it determines its own communication task with the acquisition unit and executes it; at the same time, it verifies the latest data message sequence number in the target heartbeat message with the latest message received from the target.

6. The distributed data service method for a large ship monitoring system according to claim 4, characterized in that: The data message includes: a collection data message and a control data message; The data collection message is a change-triggered and time-triggered mechanism; it is used for the workstation to send the status data of the field side uploaded from the collection unit to the predefined workstation; The interface description file of the acquisition unit uniformly divides the interface numbers of different acquisition units and also serves as the data set number of the interface. All data in the data set are pre-arranged according to the data type and address order. The acquisition data message will be sent in groups based on the data set. The sent data includes input analog and digital signals. If a single packet cannot send all the data of the interface, the data of the interface will be divided into several data fields according to the data address order in the interface description file and sent in sequence. It mainly includes data message identification number, sending end site number, timestamp, data set number, data field number, data set acquisition message sequence number, data length, data field content and check code. The control data message is an event-triggered mechanism; the workstation for issuing instructions sends the instruction data to the predetermined workstation, and the priority workstation issues the control instruction data to the field side; The control data message is also sent in groups based on the data set. The sent data includes output analog and digital signals, and is also sent as a whole or in parts according to the data volume. It mainly includes the data message identification number, the sending site number, the timestamp, the data set number, the data field number, the control message sequence number of the data set, the data length, the data field content and the check code.

7. The distributed data service method of a large ship monitoring system according to claim 4, characterized in that: The reception confirmation message is: When the receiving workstation completes the reception and verification of a data message, it returns confirmation information to the sender to indicate that the data message has been successfully received. The structure of the reception confirmation message mainly includes: confirmation message identification number, confirming party site number, timestamp, confirmed data set number, confirmed message sequence number and check code.

8. The distributed data service method for a large ship monitoring system according to claim 4, characterized in that: The resend request message is: When the receiving workstation determines that there is a missing data packet based on the latest message or heartbeat message received, it applies to the sender for resending the latest message or heartbeat message according to the missing packet situation; The structure of the resend request message mainly includes: a resend request message identification number, a request site number, a timestamp, a requested data set number, a requested message sequence number and a check code.

9. The distributed data service method of a large ship monitoring system according to claim 4, characterized in that: The UDP multicast protocol is a custom multicast protocol, which ensures reliable data transmission by confirming the receipt of sent messages and retransmitting them, including: When the sender workstation determines that the data domain in the data set it manages has changed or the sending cycle has been reached, it generates a new message of the data set; the message sequence number in the new message is incremented and updated each time it is sent, and the message content is stored in the cache while sending the message; when the receiver identifies it as a collection message based on the message identification number and verifies that the message format is complete, if the verification is passed, it parses it according to the data address sequence of the interface description file and updates the local database, and sends a reception confirmation message at the same time; when the sender confirms that all workstations have completed the reception based on the confirmation message, the message content is released in the cache; Packet loss is determined when the following three situations occur: The first case: When the workstation receives a data set with non-continuous message sequence numbers; The second situation: the latest data set message number in the received heartbeat message is inconsistent with the latest message number of the data set parsed by the station; The third situation: when receiving the message number in the confirmation message of the relevant data set sent by the predefined workstation, the workstation has not received the collection message.

10. A large ship monitoring system, characterized in that: A distributed data service method for a large ship monitoring system comprising the method described in any one of claims 1-9.

Images