A hydropower station production data low-latency high-reliability transmission communication method and system

Abstract

The application relates to the technical field of electric power communication, and particularly provides a hydropower station production data low-latency high-reliability transmission communication method and system.The method comprises the following steps: a sending end sends a probe message with a higher priority than a labeled data frame; a measured transmission latency is corrected into a net link latency of each link according to a transmission control parameter, and a link state table is constructed; the labeled data frame is distributed to a time-delay-sensitive queue and a reliability-sensitive queue according to a data level, main transmission links and standby transmission links are simultaneously sent, and the standby transmission links are kept in a built chain state; atomization pre-switching is performed before the main transmission links are interrupted according to the time sequence change of the net link latency; a receiving end removes, checks and recovers the labeled data frame and a redundant frame in an adaptive de-duplication window, and generates a retransmission request; and the forward error correction redundancy and the interleaving depth are jointly adjusted according to a packet loss degree. The application realizes low-latency high-reliability transmission of hydropower station production data.

Description

Technical Field

[0001] This invention relates to the field of power communication technology, specifically to a low-latency, high-reliability transmission communication method and system for hydropower station production data. Background Technology

[0002] With the continuous improvement of automation levels in hydropower stations, a large amount of production data is generated during the operation of hydropower stations. This includes protection and control data such as governor, excitation system, and gate control, as well as operation monitoring data such as unit vibration, sway, and temperature. Protection and control data have high requirements for transmission latency, while operation monitoring data have high requirements for transmission reliability.

[0003] In existing technologies, hydropower station production data is typically transmitted via an internal fiber optic ring network, employing a static primary / backup link redundancy protection method: the backup link remains idle and cold-standby during normal primary link operation, triggering a switchover only after a complete link failure. Because the switchover process involves sequential link establishment, negotiation, and parameter distribution, it is time-consuming, and protection and control data is easily lost during this period. Using a unified transmission strategy and retransmission mechanism for all types of data means that retransmission traffic for monitoring data during link congestion will consume channel resources, leading to increased latency for protection and control data. This creates a conflict between low-latency requirements and high-reliability requirements.

[0004] Even with link quality assessment and adaptive adjustment mechanisms, there are still insurmountable drawbacks: Adaptive adjustment itself alters link traffic. When the sender increases forward error correction redundancy or adjusts transmission bandwidth quotas, the transmission delay measured by probe packets is mixed with the queuing delay introduced by the adjustment behavior itself. The sender may misjudge the adjustment result as link degradation, repeatedly triggering handovers, resulting in ping-pong handovers. Electromagnetic interference generated by the start-up and shutdown of high-power equipment and the opening and closing of gates in the factory is sudden, causing packet loss to accumulate in a series over time. The existing forward error correction mechanism uses fixed block coding. When data frames in the same group are lost consecutively under sudden interference, error correction recovery fails for the entire group, degenerating into a large number of retransmissions. The retransmission traffic further crowds the channel, creating a vicious cycle of packet loss and congestion. The bandwidth of the backup link is much smaller than that of the fiber optic ring network. If all services are simply switched to the backup link when the main link degrades, the backup link will immediately become congested, and the goal of low latency will not be achieved. Summary of the Invention

[0005] This invention proposes a low-latency, high-reliability transmission communication method and system for hydropower station production data. It effectively solves the problems in existing hydropower station production data transmission where adaptive adjustment behavior and link awareness interfere with each other, leading to ping-pong switching; sudden packet loss due to electromagnetic interference causes fixed forward error correction to fail; and backup link switching or congestion causes the low-latency target to fail.

[0006] The technical solution adopted by this invention is as follows: This solution provides a low-latency, high-reliability transmission communication method for hydropower station production data, the method comprising the following steps:

[0007] Step S1: Obtain production data collected by each production terminal of the hydropower station, and determine the terminal type of each production terminal according to its role in the production control process; determine the data level of the production data according to the terminal type; encapsulate the production data with transmission tags, the transmission tags including data level, serial number, collection time and source terminal identifier, and obtain tagged data frames after encapsulation.

[0008] Step S2: The transmission of hydropower station production data is completed by the sending end via the main transmission link and the backup transmission link. The sending end sends probe messages via the main transmission link and the backup transmission link according to a preset probe cycle. The sending priority of the probe messages is higher than that of the tagged data frames sent via the main transmission link and the backup transmission link. The sending end obtains the measured transmission delay, packet loss rate and delay jitter of each link based on the probe messages. The sending end obtains its current transmission control parameters, including forward error correction redundancy, interleaving depth and queue occupancy rate. Based on the current forward error correction redundancy, interleaving depth and queue occupancy rate of the sending end, the queuing delay correction and interleaving buffer delay correction are determined, and the queuing delay correction and interleaving buffer delay correction are subtracted from the measured transmission delay to obtain the net link delay of each link. Based on the net link delay, packet loss rate and delay jitter, the link availability score of each link is determined, and a link status table is constructed based on the link availability score of each link.

[0009] Step S3: Based on the data level of the tagged data frames, allocate them to the delay-sensitive queue and the reliability-sensitive queue; the delay-sensitive queue buffers tagged data frames with high transmission delay requirements, and the reliability-sensitive queue buffers tagged data frames with high transmission reliability requirements; the tagged data frames in the delay-sensitive queue are normally replicated as redundant frames, and the primary transmission link and the backup transmission link simultaneously send tagged data frames and redundant frames. The sending priority of the redundant frames is the same as that of the corresponding tagged data frames, that is, both are lower than the sending priority of the probe messages, so that the backup transmission link remains established; the tagged data frames in the reliability-sensitive queue are grouped and encoded according to the current forward error correction redundancy, and interleaved according to the current interleaving depth before being sent through the primary transmission link.

[0010] Step S4: Based on the timing changes of the net link delay of the primary transmission link in the link status table, calculate the delay trend slope of the primary transmission link to obtain the predicted delay of the primary transmission link; when the predicted delay reaches the preset switching threshold, perform an atomic pre-switching before the primary transmission link is interrupted; the atomic pre-switching includes: completing the link establishment confirmation of the backup transmission link, reallocating the transmission bandwidth quota of the delay-sensitive queue and the reliability-sensitive queue according to the link availability score of the backup transmission link, resetting the forward error correction redundancy and interleaving depth according to the packet loss rate and delay jitter of the backup transmission link, and suspending the transmission of the lowest data-level tagged data frames.

[0011] Step S5: The receiving end determines the adaptive deduplication window based on the difference between the net link delay of the primary transmission link and the net link delay of the backup transmission link, as well as the delay jitter. Within the adaptive deduplication window, the receiving end deduplicates the tagged data frames and redundant frames that arrive simultaneously according to the sequence number to obtain the reassembled data frames. The integrity of the reassembled data frames is checked to obtain the check result. Based on the check result, the receiving end determines the data frames that failed to be received and restores the data frames that failed to be received according to the redundant data corresponding to the forward error correction redundancy. Only when the adaptive deduplication window of the corresponding data frame is closed and still not restored, a retransmission request carrying bandwidth occupancy weight is generated and sent to the sending end.

[0012] Step S6: The sending end determines the packet loss statistics for each data level based on the retransmission request, adjusts the forward error correction redundancy and interleaving depth jointly based on the packet loss clustering degree in the packet loss statistics, adjusts the transmission bandwidth quota for each queue in conjunction with the link availability score in the link status table, and uses the adjustment results as the control parameters for the next round of transmission, outputting the production data transmission quality assessment results.

[0013] This invention also proposes a low-latency, high-reliability transmission communication system for hydropower station production data, deployed in a transmission architecture consisting of a transmitter, a receiver, a main transmission link, and a backup transmission link. The system includes a data hierarchical encapsulation module, a net latency link sensing module, a dual-path transmission control module, an atomization pre-switching module, an adaptive deduplication and retransmission module, and a closed-loop adjustment module.

[0014] The data classification and encapsulation module acquires production data collected from each production terminal, determines the terminal type and data level of each production data, and completes the encapsulation of transmission tags, resulting in a tagged data frame.

[0015] The net delay link perception module sends probe messages to obtain the measured transmission delay, packet loss rate, and delay jitter of each link based on the probe messages; it acquires transmission control parameters, including forward error correction redundancy, interleaving depth, and queue occupancy rate; it corrects the measured transmission delay of each link based on the transmission control parameters to obtain the net link delay; it determines the link availability score of each link based on the net link delay, packet loss rate, and delay jitter; and it constructs a link status table based on the link availability score of each link.

[0016] The dual-path transmission control module receives tagged data frames, writes them into the delay-sensitive queue and the reliability-sensitive queue according to data level, copies the tagged data frames in the delay-sensitive queue to obtain redundant frames and sends them simultaneously through the main transmission link and the backup transmission link, performs forward error correction group coding according to forward error correction redundancy and interleaving according to interleaving depth on the tagged data frames in the reliability-sensitive queue and sends them through the main transmission link, sets the transmission bandwidth quota for the delay-sensitive queue and the reliability-sensitive queue, and writes the transmission bandwidth quota, forward error correction redundancy and interleaving depth into the transmission channel configuration;

[0017] The atomic pre-switching module reads the timing changes of the net link delay of the main transmission link from the link status table, calculates the predicted delay, and compares the predicted delay with the preset switching threshold. When the predicted delay reaches the preset switching threshold, an atomic pre-switching is performed before the main transmission link is interrupted. In the atomic pre-switching, the transmission bandwidth quota is reallocated and the forward error correction redundancy and interleaving depth are reset, and the information is sent to the dual-path transmission control module to update the transmission channel configuration.

[0018] The adaptive deduplication and retransmission module determines the adaptive deduplication window, performs deduplication, integrity verification and error correction and recovery on tagged data frames and redundant frames that arrive at the same time, and generates a retransmission request carrying bandwidth occupancy weight for data frames that have not been recovered after the adaptive deduplication window is closed.

[0019] The closed-loop adjustment module obtains the packet loss clustering degree based on the retransmission request, jointly adjusts the forward error correction redundancy and interleaving depth, and adjusts the transmission bandwidth quota in conjunction with the link availability score in the link status table. The adjusted transmission control parameters are fed back to the dual-path transmission control module and the net delay link perception module, respectively. The transmission control parameters fed back to the net delay link perception module are used for net link delay correction in the next detection cycle, and the production data transmission quality assessment results are output.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] (1) In view of the problem of ping-pong handover caused by mutual interference between adaptive adjustment behavior and link awareness in the prior art, the present invention sends probe messages with higher priority than tagged data frames, so that the measured transmission delay of the probe messages is not affected by the fluctuation of production data transmission traffic; by determining the queuing delay correction amount and the interleaving buffer delay correction amount according to the current forward error correction redundancy, interleaving depth and queue occupancy rate of the sending end, and deducting the above two correction amounts from the measured transmission delay, the net link delay is obtained after removing the delay introduced by the sending end's own adjustment behavior; the net link delay is used as a unified input for constructing the link state table, judging the pre-handover timing and closed-loop adjustment, so that the link state judgment is no longer polluted by the adjustment behavior, cutting off the positive feedback path of "adjustment changes traffic, traffic pollutes measurement, measurement mis-triggers adjustment", and overcoming the ping-pong handover problem of repeated triggering of handover. This effect requires the cooperation of three factors: detection mechanism, delay correction, and unified input. Without queuing delay correction and interleaving buffer delay correction, the measured transmission delay is still mixed with the delay introduced by the adjustment behavior itself. Without the link state table with net link delay as the unified input, the closed-loop adjustment of forward error correction redundancy, interleaving depth, and transmission bandwidth quota will still be misjudged as link degradation.

[0022] (2) To address the problem of fixed forward error correction failing due to sudden packet loss caused by electromagnetic interference, this invention determines the packet loss statistics for each data level based on retransmission requests, and jointly adjusts the forward error correction redundancy and interleaving depth based on the packet loss clustering in the packet loss statistics: the forward error correction redundancy ensures that the recovery capability of the error correction group matches the average packet loss level, and the interleaving depth distributes the data frames lost in bursts to different error correction groups, so that forward error correction maintains the recovery capability of the entire group under sudden packet loss scenarios, avoiding degradation into a large number of retransmissions; at the same time, the delay change introduced after the interleaving depth adjustment is included in the interleaving buffer delay correction and participates in the determination of net link delay, so that error correction enhancement will not be misjudged as link degradation. The forward error correction redundancy, interleaving depth and net link delay correction form a mutually input coupling relationship, breaking the vicious cycle of sudden packet loss, retransmission traffic crowding the channel and further congestion, and solving the problem of fixed block coding failing as a whole group under sudden packet loss.

[0023] (3) To address the problem of low-latency targets being lost due to congestion during backup transmission link switching, this invention uses the normal replication of tagged data frames in the latency-sensitive queue as redundant frames. The primary and backup transmission links simultaneously send tagged data frames and redundant frames, ensuring that the backup transmission link always remains in a linked state, and no link establishment negotiation is required during pre-switching. By binding link establishment confirmation, reallocation of transmission bandwidth quotas according to the availability score of the backup transmission link, resetting the forward error correction redundancy and interleaving depth according to the packet loss rate and latency jitter of the backup transmission link, and pausing the transmission of tagged data frames with the lowest data level into an atomic pre-switching, it takes effect all at once before the primary transmission link is interrupted, avoiding the simultaneous influx of redundant frames and error correction encoded data into the backup transmission link during switching and preventing the bandwidth of tagged data frames with low latency requirements from being squeezed during switching. In conjunction with the generation of retransmission requests carrying bandwidth occupancy weights only after the adaptive deduplication window is closed, repeated retransmissions when redundant frames are still in transit are avoided. The above mechanisms work together to achieve lossless switching before link interruption, overcoming the technical biases of cold standby of backup transmission links and congestion during switching. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating a low-latency, high-reliability transmission communication method for hydropower station production data according to the present invention.

[0025] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0027] Example 1:

[0028] Please see Figure 1 This embodiment provides a low-latency, high-reliability transmission communication method for hydropower station production data, the method comprising the following steps:

[0029] Step S1: Obtain production data collected by each production terminal of the hydropower station, classify the terminal type and determine the data level according to the role of each production terminal in the production control process, and encapsulate the transmission tag to obtain tagged data frames.

[0030] Step S2: The sending end sends probe messages with higher priority than tagged data frames through the main transmission link and the backup transmission link to obtain the measured transmission delay, packet loss rate and delay jitter of each link; combined with the current transmission control parameters of the sending end, the queuing delay correction and interleaving buffer delay correction are determined, and the queuing delay correction and interleaving buffer delay correction are subtracted from the measured transmission delay to obtain the net link delay of each link, and then the link availability score of each link is determined and the link status table is constructed.

[0031] Step S3: Write the tagged data frames into the delay-sensitive queue and the reliability-sensitive queue according to the data level; duplicate the tagged data frames in the delay-sensitive queue to obtain redundant frames, and send the tagged data frames and redundant frames simultaneously on the main transmission link and the backup transmission link to keep the backup transmission link in the established state; group and encode the tagged data frames in the reliability-sensitive queue according to the forward error correction redundancy in the current transmission control parameters, and then interleave them according to the current interleaving depth in the transmission control parameters before sending them through the main transmission link;

[0032] Step S4: Calculate the delay trend slope based on the timing change of the net link delay of the main transmission link in the link status table to obtain the predicted delay of the main transmission link; when the predicted delay reaches the preset switching threshold, perform an atomic pre-switching before the main transmission link is interrupted, and complete the link establishment confirmation of the backup transmission link, the reallocation of the transmission bandwidth quota of each queue, the reset of the forward error correction redundancy and interleaving depth, and the suspension of the transmission of the lowest data level tagged data frame in one go.

[0033] Step S5: The receiving end determines the adaptive deduplication window based on the difference between the net link delay of the main transmission link and the net link delay of the backup transmission link, as well as the delay jitter. Within the adaptive deduplication window, the tagged data frames and redundant frames that arrive simultaneously are deduplicated according to the sequence number in the transmission tag to obtain the reconstructed data frame. After integrity verification and recovery, a retransmission request carrying bandwidth occupancy weight is generated for the data frame that has not been recovered after the adaptive deduplication window is closed and sent to the sending end.

[0034] Step S6: The sending end determines the packet loss statistics for each data level based on the retransmission request, adjusts the forward error correction redundancy and interleaving depth jointly based on the packet loss clustering degree in the packet loss statistics, adjusts the sending bandwidth quota in combination with the link availability score, uses the adjusted transmission control parameters for the next round of transmission, and outputs the production data transmission quality assessment results.

[0035] In one specific implementation, a hydropower station is used as an example. The production terminals of this hydropower station include a governor terminal, an excitation terminal, a gate control terminal, a unit monitoring terminal, and a video terminal. The governor terminal collects the guide vane opening degree and unit speed; the excitation terminal collects the excitation current and turbine terminal voltage; and the gate control terminal collects the gate opening degree and opening / closing status. The above three types of terminals perform protective control in the production control process, and their terminal type is control terminal. The unit monitoring terminal collects unit vibration, sway, bearing temperature, and oil temperature, and its terminal type is monitoring terminal. The video terminal collects video data from the powerhouse and dam area, and its terminal type is auxiliary terminal.

[0036] The data level of each production data is determined according to the terminal type: the data level of production data from control terminals is determined to be Level 1, resulting in Level 1 data; the data level of production data from monitoring terminals is determined to be Level 2, resulting in Level 2 data; and the data level of production data from auxiliary terminals is determined to be Level 3, resulting in Level 3 data. Tagged data frames of Level 1 data are allocated to the latency-sensitive queue, while tagged data frames of Level 2 and Level 3 data are allocated to the reliability-sensitive queue.

[0037] The transmitting end is deployed at the station control layer communication management unit, and the receiving end is deployed at the centralized control center data receiving gateway. The main transmission link uses an in-station fiber optic ring network, and the backup transmission link uses a dedicated power wireless network. The bandwidth of the backup transmission link is less than that of the main transmission link.

[0038] Under normal operating conditions, the backup transmission link is in an established state and routinely carries redundant frames in the delay-sensitive queue. The redundant frames occupy less than 30% of the total bandwidth of the backup transmission link, maintaining a lightly loaded established state. Probe messages are sent with higher priority than tagged data frames in the delay-sensitive and reliability-sensitive queues, ensuring that the queuing delay of probe messages is unaffected by fluctuations in production data transmission traffic.

[0039] During operation, a section of the fiber optic ring network experienced micro-bending loss in the optical cable due to slope construction. If the measured transmission delay is directly used to determine the link status, the measured transmission delay will be misjudged as continuously deteriorating because the previous closed-loop adjustment has improved the forward error correction redundancy and interleaving depth. This leads to the inclusion of delays corresponding to queuing delay corrections and interleaving buffer delay corrections in the measured transmission delay, resulting in repeated handovers. In this implementation, the transmitting end subtracts the queuing delay corrections and interleaving buffer delay corrections from the measured transmission delay to obtain the net link delay: the net link delay for Level 1 data remains stable at around 4 milliseconds, unaffected by adjustment actions, and no erroneous handovers occur. The net link delay of the main transmission link increases from 4 milliseconds to 9 milliseconds and then to 14 milliseconds over several consecutive detection cycles. The transmitting end calculates the delay trend slope based on the change in net link delay over time to obtain the predicted delay, which reaches the preset handover threshold of 20 milliseconds.

[0040] The sending end then performs an atomic pre-switching: confirming that the backup transmission link is established; transferring the transmission bandwidth quota for Level 1 data in the latency-sensitive queue from the primary transmission link to the backup transmission link, and prioritizing the allocation of the remaining bandwidth of the primary transmission link to the reliability-sensitive queue; increasing the forward error correction redundancy from 5% to 10% and the interleaving depth from 1 to 2 based on the packet loss rate and latency jitter of the backup transmission link; suspending the transmission of Level 3 data to release the bandwidth of the backup transmission link. These operations take effect all at once within the same decision cycle, preventing congestion on the backup transmission link during the switchover.

[0041] Approximately 40 seconds later, the fiber optic cable in that section was completely severed. Since pre-switching and parameter resetting had been completed, primary data continued to be transmitted via the backup transmission link, and the transmission delay of the reassembled data frames remained within the preset delay requirements; no data loss occurred.

[0042] During operation, the opening and closing of gates within the plant generated sudden electromagnetic interference, causing the reliability-sensitive queue to lose multiple adjacent data frames in a short period of time. The transmitting end, based on retransmission requests, statistically analyzed packet loss at each data level, determined that the packet loss clustering had reached the burst threshold, and increased the interleaving depth from 2 to 4. This dispersed the continuously lost data frames into different error correction groups. The receiving end directly recovered most of the lost data frames using the redundant data corresponding to the forward error correction redundancy. Simultaneously, the delay variation introduced by the increased interleaving depth was included in the interleaving buffer delay correction and participated in the determination of the net link delay, without causing any misjudgment during pre-handover.

[0043] Therefore, the hydropower station's production data transmission process ensured both low latency and high reliability under abnormal operating conditions, including both fiber optic cable interruption and sudden electromagnetic interference.

[0044] Example 2:

[0045] This embodiment is based on Embodiment 1 and further explains the process of generating tagged data frames in step S1.

[0046] A terminal type table is pre-configured, which records the terminal type of each production terminal. The terminal types are classified according to the role of the production terminal in the production control process.

[0047] Query the terminal type table for each production terminal; pre-configure the level mapping table, which records the correspondence between terminal type and data level, where control terminals correspond to level 1, monitoring terminals correspond to level 2, and auxiliary terminals correspond to level 3; determine the data level of production data by querying the level mapping table based on the terminal type.

[0048] Assign serial numbers to production data; production data under the same source terminal identifier are numbered sequentially according to the time of collection, while production data under different source terminal identifiers are numbered independently.

[0049] The data level, sequence number, acquisition time, and source terminal identifier are encapsulated into a transmission tag, and the transmission tag is attached to the production data header to obtain a tagged data frame; wait for queue allocation in step S3.

[0050] Example 3:

[0051] This embodiment, based on Embodiment 1, further explains the process of determining net link latency and link availability score in step S2. The tagged data frames are generated using the method described in Embodiment 2.

[0052] The sending end sends probe messages to the main transmission link and the backup transmission link according to a preset probe period. The probe messages carry a sending timestamp. The receiving end fills in the acknowledgment timestamp on the received probe messages and returns them along the same path. The probe messages and tagged data frames share the sending channel at the sending end and the parsing channel at the receiving end. The processing priority of the probe messages in the sending channel and the parsing channel is higher than the processing priority of the tagged data frames in the delay-sensitive queue and the reliability-sensitive queue.

[0053] The sending end determines the round-trip time of each probe message on each link based on the sending timestamp and the backfilled response timestamp carried in the probe message, and determines the measured transmission delay of each link based on the round-trip time; the sending end determines the packet loss rate of each link based on the number of probe messages that have not returned along the original path, and determines the delay jitter of each link based on the fluctuation range of the measured transmission delay over multiple consecutive probe cycles.

[0054] The transmitting end reads its current transmission control parameters to obtain the current queue occupancy rate and the current interleaving depth, and determines the queuing delay correction and interleaving buffer delay correction accordingly: the queuing delay correction is the product of the current queue occupancy rate and the channel transmission duration of a single tagged data frame; the interleaving buffer delay correction is the product of the current interleaving depth minus 1 and the channel transmission duration of a single tagged data frame; the channel transmission duration of a single tagged data frame is pre-calculated based on the bandwidth of each link and the average length of multiple tagged data frames.

[0055] The net link delay for each link is determined using the following formula:

[0056] ;

[0057] in, Indicates net link latency; This indicates the measured transmission delay of the link; This indicates the amount of queuing delay correction; This indicates the amount of interleaving buffer delay correction;

[0058] Based on the net link latency, packet loss rate, and latency jitter of each link, a weighted sum is obtained after normalization to obtain the link availability score for each link. The normalized net link latency, packet loss rate, and latency jitter constitute the latency, reliability, and stability components of the link availability score, respectively. Lower net link latency results in a higher latency component, lower packet loss rate results in a higher reliability component, and lower latency jitter results in a higher stability component. In this embodiment, the sum of the weights corresponding to the latency, reliability, and stability components is 1, with a weight of 0.5 for the latency component, 0.3 for the reliability component, and 0.2 for the stability component.

[0059] The link availability score, net link latency, packet loss rate, and latency jitter of each link are associated with the corresponding link and written into the link status table. Each record in the link status table corresponds to the primary transmission link and the backup transmission link, respectively. The link status table is updated with the detection results of each detection cycle and is used for subsequent pre-switching decisions and closed-loop adjustment.

[0060] It should be noted that the link status table uses net link delay, rather than measured transmission delay, as the delay metric. Since the closed-loop regulation adjusts the transmission control parameters, it changes the production data transmission flow. If the measured transmission delay is used directly as the basis for judgment, the delay introduced by the regulation behavior itself will be misjudged as link degradation, leading to ping-pong switching. Using net link delay as a unified input cuts off this positive feedback path.

[0061] Example 4:

[0062] This embodiment, based on Embodiment 1, further explains the process in step S3 of allocating tagged data frames to the delay-sensitive queue and the reliability-sensitive queue, and of simultaneously sending tagged data frames and redundant frames on the primary transmission link and the backup transmission link. Specifically, the tagged data frames are generated using the method described in Embodiment 2, and the link state table is generated using the method described in Embodiment 3.

[0063] Read the data level from the transmission tag of the tagged data frame, write the tagged data frame of level 1 data into the delay-sensitive queue, and write the tagged data frames of level 2 and level 3 data into the reliability-sensitive queue.

[0064] Sending bandwidth quotas are set for latency-sensitive queues and reliability-sensitive queues respectively, and the sending bandwidth quotas are dynamically updated with closed-loop adjustment.

[0065] For each tagged data frame in the latency-sensitive queue, a redundant frame is obtained by copying it. The redundant frame carries the same sequence number and acquisition time as the corresponding tagged data frame. The tagged data frame is sent through the main transmission link, and the redundant frame is sent through the backup transmission link. The transmission state in which the main transmission link and the backup transmission link send tagged data frames and redundant frames simultaneously is recorded as the dual-path normal concurrent state. The transmission priority of the redundant frame is the same as that of the corresponding tagged data frame, that is, both are lower than the transmission priority of the probe message.

[0066] In this embodiment, the hot standby ratio is pre-set, and the bandwidth occupied by redundant frames on the backup transmission link is limited to the product of the bandwidth of the backup transmission link and the hot standby ratio. The hot standby ratio is 30%. By adopting the above-mentioned dual-path normal concurrent state, the established state of the backup transmission link is actually verified in each detection cycle. During the atomic pre-switching, there is no need to re-execute the link establishment confirmation. Only the forward error correction redundancy and interleaving depth need to be reset.

[0067] For each tagged data frame in the reliability-sensitive queue, several consecutive tagged data frames are grouped and encoded according to the current forward error correction redundancy to obtain the forward error correction redundancy check block corresponding to each group of data frames; the data frames of each error correction group are interleaved according to the current interleaving depth and then sent through the main transmission link.

[0068] Among them, the forward error correction redundancy check block is used to directly recover the tagged data frames lost in the same error correction group at the receiving end; the interleaving depth represents the interleaving interval of data frames in adjacent error correction groups when they are sent through the main transmission link, and is used to distribute the data frames that are lost in a sudden series to different error correction groups.

[0069] Example 5:

[0070] This embodiment, based on Embodiment 1, further explains the atomic pre-switching process in step S4. Specifically, the link state table is generated using the method described in Embodiment 3, and the latency-sensitive queue and reliability-sensitive queue are generated using the method described in Embodiment 4.

[0071] Read the net link delay of the main transmission link in multiple consecutive detection cycles from the link status table, perform linear fitting on the change of net link delay with time in each detection cycle, and determine the slope of the fitted line as the delay trend slope of the main transmission link.

[0072] Pre-set the pre-switching lead time and calculate the predicted delay of the main transmission link according to the following formula:

[0073] ;

[0074] in, Indicates the predicted latency; This represents the net link latency for the most recent probe cycle; Indicates the slope of the time delay trend; Indicates the pre-switching lead time;

[0075] The predicted latency is compared with a preset handover threshold. When the predicted latency reaches the preset handover threshold, an atomic pre-handover is performed all at once within the same decision cycle, including the following operations:

[0076] Operation 1: Send a link establishment confirmation request to the backup transmission link; preset the availability threshold. Since the backup transmission link is in a dual-path normal concurrent established state, the link establishment confirmation only needs to verify that the link availability score of the backup transmission link reaches the availability threshold, without renegotiating the link establishment parameters.

[0077] Operation 2: Transfer the transmission bandwidth quota of Level 1 data in the latency-sensitive queue from the primary transmission link to the backup transmission link, and reallocate the remaining bandwidth of the primary transmission link to the reliability-sensitive queue; the reallocation is based on the link availability scores of the backup transmission link and the primary transmission link respectively.

[0078] Operation 3: Reset the forward error correction redundancy and interleaving depth according to the packet loss rate and packet loss clustering of the backup transmission link; where packet loss clustering is the proportion of the number of adjacent lost data frames to the total number of lost data frames, and adjacent loss means that the lost data frames are in consecutive positions in the transmission sequence;

[0079] Operation 4: Suspend the transmission of the lowest-level tagged data frames, i.e., Level 3 data, to free up bandwidth on the backup transmission link;

[0080] The above four operations are executed as a whole and take effect at once; if only some of the operations are executed, the backup transmission link will become congested at the moment of switching due to the simultaneous influx of primary data, redundant frames and forward error correction redundancy check blocks.

[0081] After the switchover is completed, the redundant frame transmission channel on the main transmission link is retained; a stable observation period and a recovery threshold are preset. During the stable observation period, the net link latency is used as the criterion. When the link availability score of the main transmission link recovers to above the recovery threshold and remains at the preset duration, the tagged data frames are switched back to the main transmission link for transmission.

[0082] The preset handover threshold is higher than the recovery threshold, and the pre-handover lead time and the stable observation period are both no less than an integer multiple of the detection period, so as to avoid repeated handover when the net link latency fluctuates around the preset handover threshold.

[0083] Example 6:

[0084] This embodiment, based on Embodiment 1, further explains the process of determining the adaptive deduplication window and generating retransmission requests in step S5. Specifically, tagged data frames and redundant frames are generated using the method described in Embodiment 4, and the link state table is generated using the method described in Embodiment 3.

[0085] The receiving end reads the net link delay of the primary transmission link from the link state table. Net link delay of backup transmission links Delay jitter of the main transmission link Delay jitter of backup transmission links Pre-set margin coefficient and basic window The adaptive deduplication window is determined according to the following formula:

[0086] ;

[0087] in, This represents an adaptive deduplication window; in this embodiment... Take 1.5, Take 5 milliseconds;

[0088] Within the adaptive deduplication window, the receiving end retains the earliest arriving data frame as the valid frame and discards the remaining duplicate frames to obtain the recombined data frame for data frames with the same sequence number under the same source terminal identifier.

[0089] Perform integrity verification on the reconstructed data frame to obtain the verification result; determine the tagged data frames that failed or were missing based on the verification result, and restore them first based on the forward error correction redundancy check block corresponding to the same group of data frames.

[0090] Data frames that have not been recovered by the forward error correction redundancy check block after the adaptive deduplication window is closed are confirmed as truly lost, and a retransmission request is generated for the truly lost data frames.

[0091] The retransmission request carries the sequence number, source terminal identifier, and data level of the corresponding data frame, and carries the bandwidth occupancy weight according to the data level; the bandwidth occupancy weight of each data level is preset, the bandwidth occupancy weight corresponding to the first-level data is greater than the bandwidth occupancy weight corresponding to the second-level data, and the bandwidth occupancy weight corresponding to the second-level data is greater than the bandwidth occupancy weight corresponding to the third-level data.

[0092] The retransmission request is sent to the sending end; the sending end only retransmits the data frame specified in the retransmission request, and the retransmission request for first-level data enters the transmission channel with priority over the retransmission requests for second-level and third-level data.

[0093] By constraining the generation of retransmission requests to occur after the adaptive deduplication window is closed, the receiver is prevented from misjudging packet loss and triggering repeated retransmissions while redundant frames are still in transit on the backup transmission link, thus reducing the occupation of channel resources by invalid retransmissions.

[0094] Example 7:

[0095] This embodiment, based on Embodiment 1, further explains the process of jointly adjusting and closing-loop regulating the forward error correction redundancy and interleaving depth in step S6. Specifically, the link state table is generated using the method described in Embodiment 3, the retransmission request is generated using the method described in Embodiment 6, and the packet loss clustering degree is defined as described in Operation 3 of Embodiment 5.

[0096] The statistical period is set in advance. The sending end counts the number of packets lost for each data level based on the retransmission requests received within the statistical period, and obtains the packet loss statistics for each data level.

[0097] Determine the packet loss clustering degree of the main transmission link based on packet loss statistics. ;

[0098] The forward error correction redundancy is adjusted as follows: a preset packet loss reference threshold, a lower limit, and an upper limit for the forward error correction redundancy are pre-set. When the packet loss rate of the corresponding link in the current statistical period is not greater than the preset packet loss reference threshold, the forward error correction redundancy remains at the lower limit. When the packet loss rate is greater than the preset packet loss reference threshold, the forward error correction redundancy increases linearly between the lower and upper limits as the packet loss rate increases. In this embodiment, the preset packet loss reference threshold is 5%, the lower limit of the forward error correction redundancy is 2%, and the upper limit is 20%.

[0099] The interlacing depth is determined according to the following formula:

[0100] ;

[0101] in, Indicates the interlacing depth; This indicates the preset upper limit of interlacing depth, in this embodiment. Take 8; Indicates rounding up;

[0102] Among them, forward error correction redundancy is used to ensure that the recovery capability of the error correction group matches the packet loss rate within the statistical period; interleaving depth is used to distribute suddenly lost data frames to different error correction groups; and packet loss clustering... When the forward error correction redundancy is increased, it is not enough to recover consecutively lost data frames within the same error correction group. The interleaving depth must be increased simultaneously. The delay change introduced after the interleaving depth adjustment is updated to the interleaving buffer delay correction amount described in Example 3, and used as the input for the net link delay correction in the next detection cycle.

[0103] Based on the link availability scores of the primary and backup transmission links, the transmission bandwidth quotas of the delay-sensitive queue and the reliability-sensitive queue on each link are reallocated proportionally; when the link availability score of the primary transmission link decreases, the transmission bandwidth quota of the backup transmission link is increased; when the link availability score of the primary transmission link recovers, the transmission bandwidth quota of the backup transmission link is reclaimed.

[0104] The adjusted forward error correction redundancy, interleaving depth, and transmission bandwidth quota are used as adjusted transmission control parameters, written into the transmission channel configuration, and fed back to the transmitting end for net link delay correction in the next probe cycle.

[0105] Based on the packet loss statistics for each data level, the net link latency for each link, and the number of retransmission requests, a production data transmission quality assessment result is generated and output. The production data transmission quality assessment result is also used to generate and output alarm information under abnormal operating conditions.

[0106] Example 8:

[0107] This embodiment provides a low-latency, high-reliability transmission communication system for hydropower station production data. The system is deployed in a transmission architecture consisting of a transmitter, a receiver, a main transmission link, and a backup transmission link, and is used to execute the methods described in any of the preceding embodiments. The system includes a data hierarchical encapsulation module, a net latency link sensing module, a dual-path transmission control module, an atomization pre-switching module, an adaptive deduplication and retransmission module, and a closed-loop adjustment module.

[0108] The data classification and encapsulation module acquires production data collected from various production terminals of the hydropower station, classifies terminal types according to the role of each production terminal in the production control process, determines the data level of the production data, encapsulates transmission tags for the production data, and obtains tagged data frames.

[0109] The net delay link sensing module sends probe messages with higher priority than tagged data frames through the main transmission link and the backup transmission link to obtain the measured transmission delay, packet loss rate and delay jitter of each link; combined with the current transmission control parameters of the sending end, it determines the queuing delay correction amount and the interleaving buffer delay correction amount, and subtracts the above two correction amounts from the measured transmission delay to obtain the net link delay of each link, and then determines the link availability score of each link and constructs the link status table;

[0110] The dual-path transmission control module writes tagged data frames into a delay-sensitive queue and a reliability-sensitive queue according to data level; it copies tagged data frames in the delay-sensitive queue to obtain redundant frames, and simultaneously sends tagged data frames and redundant frames through the main transmission link and the backup transmission link, keeping the backup transmission link in an established state; it performs group coding on tagged data frames in the reliability-sensitive queue according to the current forward error correction redundancy, performs interleaving according to the current interleaving depth, and then sends them through the main transmission link; it sets transmission bandwidth quotas for the delay-sensitive queue and the reliability-sensitive queue, and writes the transmission bandwidth quotas, forward error correction redundancy, and interleaving depth into the transmission channel configuration;

[0111] The atomic pre-switching module reads the timing changes of the net link delay of the main transmission link from the link status table, calculates the predicted delay, and compares the predicted delay with the preset switching threshold. When the predicted delay reaches the preset switching threshold, an atomic pre-switching is performed before the main transmission link is interrupted. In the atomic pre-switching, the transmission bandwidth quota is reallocated and the forward error correction redundancy and interleaving depth are reset, and the information is sent to the dual-path transmission control module to update the transmission channel configuration.

[0112] The adaptive deduplication and retransmission module determines the adaptive deduplication window, performs deduplication, integrity verification and recovery on tagged data frames and redundant frames that arrive at the same time, and generates a retransmission request carrying bandwidth occupancy weight for data frames that have not yet been recovered after the adaptive deduplication window is closed.

[0113] The closed-loop adjustment module obtains the packet loss clustering degree based on the retransmission request, jointly adjusts the forward error correction redundancy and interleaving depth, and adjusts the transmission bandwidth quota in conjunction with the link availability score in the link status table. The adjusted transmission control parameters are fed back to the dual-path transmission control module and the net delay link perception module, respectively. The transmission control parameters fed back to the net delay link perception module are used for net link delay correction in the next detection cycle, and the production data transmission quality assessment results are output.

[0114] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A low-latency, high-reliability transmission communication method for hydropower station production data, characterized in that, The method for transmitting production data via a primary transmission link and a backup transmission link includes the following steps: Step S1: Obtain production data collected from each production terminal of the hydropower station, determine the data level, encapsulate the production data with transmission tags, and obtain tagged data frames; Step S2: Send probe messages with higher priority than tagged data frames to obtain the measured transmission delay, packet loss rate and delay jitter of each link; obtain transmission control parameters, including forward error correction redundancy and interleaving depth; correct the measured transmission delay to the net link delay; and construct the link state table. Step S3: Allocate tagged data frames to the delay-sensitive queue and the reliability-sensitive queue according to data level and set the transmission bandwidth quota; the tagged data frames in the delay-sensitive queue are copied into redundant frames and transmitted simultaneously through the main transmission link and the backup transmission link; the tagged data frames in the reliability-sensitive queue are encoded, interleaved and then transmitted. Step S4: Obtain the predicted delay based on the net link delay of the primary transmission link in the link status table. When the predicted delay reaches the preset handover threshold, perform atomic pre-handover. Step S5: Determine the adaptive deduplication window, deduplicate, verify and restore tagged data frames and redundant frames that arrive simultaneously, and generate a retransmission request. Step S6: Determine the packet loss clustering degree based on the retransmission request, and jointly adjust the forward error correction redundancy and interleaving depth, as well as the transmission bandwidth quota; In step S2, the queuing delay correction and interleaving buffer delay correction are determined according to the transmission control parameters. The queuing delay correction and interleaving buffer delay correction are subtracted from the measured transmission delay to obtain the net link delay of each link. The link availability score of each link is determined according to the net link delay, packet loss rate and delay jitter, and a link status table is constructed. In step S3, the redundant frame carries the same transmission label as the corresponding tagged data frame, and the redundant frame is sent through the backup transmission link; the tagged data frames in the reliability sensitive queue are grouped and encoded according to the forward error correction redundancy to obtain the forward error correction redundancy check block, and then interleaved according to the interleaving depth and sent through the main transmission link. In step S4, the atomic pre-switching includes: completing the link establishment confirmation of the backup transmission link; reallocating the transmission bandwidth quota according to the link availability score of the backup transmission link; resetting the forward error correction redundancy and interleaving depth according to the packet loss rate and latency jitter of the backup transmission link; and suspending the transmission of the lowest data-level tagged data frames.

2. The method for low-latency and high-reliability transmission and communication of hydropower station production data according to claim 1, characterized in that: In step S1, a terminal type table that records the function of each production terminal and a level mapping table that records the correspondence between the function of the production terminal and the data level are pre-configured. The function of the production terminal is queried according to the terminal type table, and the data level of the production data is determined according to the level mapping table.

3. The method for low-latency and high-reliability transmission communication of hydropower station production data according to claim 2, characterized in that: In step S4, the slope of the delay trend is calculated based on the change of the net link delay of the main transmission link with time in the link status table, and then the predicted delay of the main transmission link is obtained.

4. The method for low-latency and high-reliability transmission communication of hydropower station production data according to claim 3, characterized in that: In step S5, the tagged data frames that failed or were missing are recovered according to the forward error correction redundancy check block; tagged data frames that are not recovered after the adaptive deduplication window is closed are confirmed to be truly lost. The retransmission request carries the transmission tag of the corresponding tagged data frame and the bandwidth usage weight determined according to the data level.

5. The method for low-latency and high-reliability transmission communication of hydropower station production data according to claim 4, characterized in that: In step S5, an adaptive deduplication window is determined based on the difference in net link delay between the primary transmission link and the backup transmission link, as well as the delay jitter. Specifically, for tagged data frames and redundant frames with the same transmission label, the earliest arriving data frame is retained as the valid frame, and the remaining duplicate frames are discarded.

6. The method for low-latency and high-reliability transmission communication of hydropower station production data according to claim 4, characterized in that: In step S6, a statistical period is preset, and the number of packet losses at each data level is counted according to the retransmission request based on the statistical period. The packet loss clustering degree is determined based on the number of packet losses. When the packet loss rate is greater than the preset packet loss reference threshold, the forward error correction redundancy increases with the increase of the packet loss rate. The delay change introduced by the interleaving depth adjustment is included in the interleaving buffer delay correction amount.

7. A low-latency, high-reliability transmission communication system for hydropower station production data, used to implement the low-latency, high-reliability transmission communication method for hydropower station production data as described in any one of claims 1-6, characterized in that: It includes a data hierarchical encapsulation module, a net latency link awareness module, a dual-path transmission control module, an atomization pre-switching module, an adaptive deduplication and retransmission module, and a closed-loop adjustment module; The data classification and encapsulation module encapsulates transmission tags to obtain tagged data frames; the net delay link awareness module obtains the net link delay of each link and constructs a link status table. The dual-path transmission control module allocates tagged data frames to the delay-sensitive queue and the reliability-sensitive queue and then sends them. The atomic pre-switching module performs atomic pre-switching; the adaptive deduplication and retransmission module performs deduplication, verification, recovery and generates retransmission requests; the closed-loop adjustment module jointly adjusts the forward error correction redundancy and interleaving depth.

Images