Power deterministic network routing and packet scheduling method and system

Abstract

The application provides a power deterministic network routing and packet scheduling method and system, the method comprising: determining the feasible path set from the source to the destination node for each deterministic flow according to the network and traffic model; constructing the three-dimensional scheduling variable consisting of the path, time offset and CSQF period specification set for each flow; jointly solving the three-dimensional variable of each flow to obtain the routing and packet scheduling scheme satisfying the resource, sending, delay and period constraint with the minimum objective function, the objective function being used to realize the double load balancing of the link and period window; and calculating the network remaining resource based on the obtained scheme and allocating the transmission resource for the non-deterministic flow according to the network remaining resource. Through the three-dimensional joint scheduling of the routing, time and period and the double load balancing of the link and window, the application can significantly reduce the jitter under the strict delay constraint, effectively improve the network resource utilization and the reliability of the deterministic service.

Description

Technical Field

[0001] This application relates to the field of power system communication network technology, and in particular to a method and system for power deterministic network routing and packet scheduling. Background Technology

[0002] With the large-scale grid connection of new energy sources, new power systems place higher demands on the flexibility, determinism, and reliability of communication. However, traditional best-effort IP networks suffer from uncertain packet latency and jitter, making it difficult to meet the stringent end-to-end deterministic transmission requirements of emerging services such as remote control, remote monitoring, and precise load control, potentially leading to control failures and system instability. To address the deterministic problem in IP networks, Deterministic Networking (DetNet) technology has emerged. One of its core traffic scheduling mechanisms, Cycle-Specified Queuing and Forwarding (CSQF), evolved from Cyclic Queuing and Forwarding (CQF) in Time-Sensitive Networking (TSN), providing bounded latency for services through multi-queue design and periodic scheduling.

[0003] Currently, research on flow scheduling for deterministic networks includes: schemes using four circular queues and specific algorithms to cope with latency variations; methods combining pre-routing and injected time planning to achieve bounded latency; cross-domain scheduling methods based on network calculus to calculate worst-case latency; and flow scheduling methods based on Time-AwareShaper (TAS) in asynchronous TSNs at the edge.

[0004] However, the main problems with the above-mentioned existing technologies include: most solutions only focus on the latency limit of deterministic services, ignoring the jitter problem caused by changes in transmission latency, and cannot guarantee stable deterministic transmission; they fail to make full use of the period specification feature of CSQF to jointly optimize routing selection, injection time planning and periodic scheduling, resulting in limited solution space and low flow schedulability and network resource utilization. Summary of the Invention

[0005] In view of this, embodiments of this application provide a method and system for deterministic power network routing and packet scheduling to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of this application provides a method for deterministic power network routing and packet scheduling, comprising: Based on the network topology model and flow model of the power deterministic network, for each deterministic flow with end-to-end delay constraints in the deterministic flow set in the flow model, a feasible path from the source node to the destination node is determined for each deterministic flow, forming a set of feasible paths for each deterministic flow. Each of the deterministic flows is constructed with its own three-dimensional scheduling decision variables; wherein, the three-dimensional scheduling decision variables include a route path selected from the set of feasible paths, an injection time offset, and a CSQF period specification set for indicating that the deterministic flow is scheduled to be forwarded at each hop of its route path; With minimizing the objective function as the optimization objective, the three-dimensional scheduling decision variables of each deterministic flow are jointly solved to obtain the routing and packet scheduling schemes that satisfy resource constraints, transmission constraints, end-to-end delay constraints, and period specification constraints for each deterministic flow; wherein, the objective function is used to achieve dual load balancing of links and periodic windows based on continuous CSQF periods; The remaining network resources of the power deterministic network are calculated based on the routing and packet scheduling scheme, and transmission resources are allocated to the nondeterministic flows in the traffic model based on the remaining network resources.

[0007] In some embodiments of this application, before determining the feasible path from the source node to the destination node corresponding to each of the deterministic flows, the method further includes: Construct a network topology model for the power deterministic network; wherein the network topology model includes: a directed graph, wherein the points in the directed graph represent transmission nodes in the power deterministic network, and the directed edges in the directed graph represent a unidirectional communication link between two transmission nodes; the transmission nodes include source nodes, relay nodes, and destination nodes.

[0008] In some embodiments of this application, before determining the feasible path from the source node to the destination node corresponding to each of the deterministic flows, the method further includes: Construct a flow model for the power deterministic network; wherein the flow model includes: a deterministic flow set for storing deterministic flows and a nondeterministic flow set for storing nondeterministic flows; The deterministic flow is defined by a first tuple, which includes: the source node, destination node, end-to-end latency limit, transmission period, maximum tolerable jitter, total amount of data transmitted in one period, set of feasible paths, routing and packet scheduling scheme, binary variables, and priority; the binary variable takes a value of 1 to indicate that the flow has been successfully scheduled, otherwise it takes a value of 0. The deterministic flow has a higher priority than the non-deterministic flow; The nondeterministic flow is defined by a second tuple, which includes: the source node of the flow, the destination node, the total amount of data sent in a period, the set of feasible routes, and the priority.

[0009] In some embodiments of this application, before determining the feasible path from the source node to the destination node corresponding to each of the deterministic flows, the method further includes: Construct a transmission node model for the power deterministic network; wherein the transmission node model represents that: each output port is configured with multiple priority queues, wherein the highest priority queues are dedicated to scheduling the deterministic flow and perform periodic round-robin transmission following the CSQF mechanism or CSQF extended rules; the remaining queues are used to schedule the nondeterministic flow; The resource constraint is used to indicate that, within any CSQF cycle, the total amount of deterministic stream data scheduled in any queue of any transmission node port does not exceed the capacity of that queue. The sending constraint is used to indicate that, in any CSQF cycle, the total amount of deterministic stream data scheduled to be sent to any queue does not exceed the maximum amount of data that the queue can send based on the port rate in one CSQF cycle; The CSQF extension rules include: when a data packet of a deterministic flow arrives at the transmission node in an unexpected CSQF period due to transmission delay, and the transmission node supports PIFO queue management, the transmission node makes a determination based on the period specification set of the deterministic flow to which the data packet belongs and the current period; if it is determined that there exists a deterministic queue that satisfies the end-to-end delay constraint of the deterministic flow and is the next to be converted to the sending state, and the sum of the total amount of data to be sent in the current deterministic queue and the data amount of the data packet satisfies the sending constraint, then the operation of inserting the data packet into the tail of the deterministic queue is performed; otherwise, the data packet is discarded.

[0010] In some embodiments of this application, the step of determining feasible paths from the source node to the destination node for each deterministic flow with end-to-end delay constraints within the deterministic flow set in the flow model, based on the network topology model and flow model of the power deterministic network, to form a set of feasible paths for each deterministic flow, includes: For each deterministic flow within the deterministic flow set in the flow model of the power deterministic network, the K-shortest path algorithm is used to calculate the first K shortest paths from the source node to the destination node of the deterministic flow in the network topology model of the power deterministic network, forming an initial path set; Based on the end-to-end delay constraint of each deterministic flow and the preset single-hop transmission delay, all paths that satisfy the end-to-end delay constraint are selected from the initial path set as feasible paths, forming a set of feasible paths for each deterministic flow; wherein, a path satisfies the end-to-end delay constraint if the product of the number of hops of the path and the single-hop transmission delay is less than or equal to the end-to-end delay constraint.

[0011] In some embodiments of this application, the step of constructing the respective three-dimensional scheduling decision variables for each of the deterministic flows includes: For each of the aforementioned deterministic flows, perform the following operations respectively: Select a feasible path from the set of feasible paths of the deterministic flow as the routing path in the three-dimensional scheduling decision variables; An integer value is determined for the deterministic flow within the interval [0, P) as the injection time offset in the three-dimensional scheduling decision variable, where P is the transmission period of the deterministic flow; Generate an ordered set of integers of the same length as the number of hops in the routing path, as the period specification set in the three-dimensional scheduling decision variables; the period specification set satisfies the period specification constraint; Wherein, the period specification constraint is used to indicate that: the i-th integer in the period specification set represents the CSQF period number offset of the deterministic flow being scheduled and forwarded at the i-th hop of the routing path, and for any two adjacent period offsets in the period specification set... and ,satisfy , This represents the total number of CSQF cycles. This indicates the modulo operation.

[0012] In some embodiments of this application, the step of jointly solving the three-dimensional scheduling decision variables of each deterministic flow with the goal of minimizing the objective function and satisfying the end-to-end delay constraint to obtain the routing and grouping scheduling schemes for each deterministic flow includes: Initialize the set of scheduled flows to an empty set, the set of unscheduled flows to contain all the aforementioned deterministic flows, and set the maximum number of iterations; Repeat the preset taboo search iteration steps until the maximum number of iterations is reached or the unscheduled flow set is empty, and output the historical best routing and group scheduling scheme as the routing and group scheduling scheme; The tabu search iteration steps include: In the current iteration, if the set of scheduled flows is not empty, a perturbation is performed based on a preset release probability; if performed, at least one deterministic flow is selected from the set of scheduled flows, its network resources are released, and it is moved to the set of unscheduled flows. For each deterministic flow in the set of unscheduled flows, a routing and packet scheduling scheme is determined for the deterministic flow from the selectable range of the three-dimensional scheduling decision variables, which simultaneously satisfies resource constraints, transmission constraints, end-to-end delay constraints, and period specification constraints; Calculate the value of the objective function based on the routing and packet scheduling schemes of all deterministic flows in the current scheduled flow set; update the tabu table based on the value of the objective function, and update the historically optimal routing and packet scheduling scheme.

[0013] In some embodiments of this application, the objective function is shown in the following formula (I): Formula (1) in, The objective function is... The sum of network resources occupied by all scheduled deterministic flows is used to optimize network resource utilization while satisfying the end-to-end delay constraints. The standard deviation of the load rate of all routing links in the power deterministic network is used to measure the load balance between links. This is the average of the variances of the load rates for each period window across all links, used to measure the load balance between period windows. and These are preset constant coefficients.

[0014] In some embodiments of this application, The calculation formula is shown in Formula (II): Formula (II) Where F represents the set of deterministic flows, For the i-th deterministic flow A flag indicating whether the scheduling was successful; For the i-th deterministic flow Path hop count; For the i-th deterministic flow The amount of data; For the i-th deterministic flow The actual end-to-end delay.

[0015] In some embodiments of this application, The calculation formula is shown in formula (III): Formula (3) in, The total number of links in the power deterministic network. Let the load rate of the j-th link be . This represents the average load rate across all links.

[0016] In some embodiments of this application, The calculation formula is shown in formula (iv): Formula (IV) in, Number of ports; A set of periodic windows; Let be the load of the k-th port of the j-th transmission node in the t-th period window; Let be the average load of the k-th port of the j-th transmission node over all periodic windows.

[0017] A second aspect of this application provides a power deterministic network routing and packet scheduling apparatus, comprising: The feasible path acquisition module is used to determine the feasible path from the source node to the destination node for each deterministic flow with end-to-end time delay constraints in the deterministic flow set in the power deterministic network based on the network topology model and flow model, thereby forming a set of feasible paths for each deterministic flow. The decision variable construction module is used to construct three-dimensional scheduling decision variables for each of the deterministic flows; wherein, the three-dimensional scheduling decision variables include a routing path selected from the set of feasible paths, an injection time offset, and a CSQF periodic specification set used to indicate that the deterministic flow is scheduled to be forwarded at each hop of its routing path; The joint scheduling solution module is used to perform joint scheduling solution on the three-dimensional scheduling decision variables of each deterministic flow with the goal of minimizing the objective function, so as to obtain the routing and packet scheduling schemes that satisfy resource constraints, transmission constraints, end-to-end delay constraints and period specification constraints for each deterministic flow; wherein, the objective function is used to achieve dual load balancing of links and periodic windows based on continuous CSQF periods; The remaining resource allocation module is used to calculate the remaining network resources of the power deterministic network based on the routing and packet scheduling scheme, and to allocate transmission resources to the nondeterministic flow in the traffic model based on the remaining network resources.

[0018] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the power deterministic network routing and packet scheduling method.

[0019] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the power deterministic network routing and packet scheduling method described above.

[0020] The fifth aspect of this application provides a computer program product comprising a computer program that, when executed by a processor, implements the power deterministic network routing and packet scheduling method.

[0021] The sixth aspect of this application provides a power deterministic network system, comprising: multiple transmission nodes supporting the CSQF forwarding mechanism and electronic devices in a power deterministic network; The electronic device includes a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the power deterministic network routing and packet scheduling method.

[0022] The power deterministic network routing and packet scheduling method provided in this application, based on the network topology model and traffic model of the power deterministic network, determines feasible paths from the source node to the destination node for each deterministic flow with end-to-end delay constraints within the deterministic flow set in the traffic model, forming feasible path sets for each deterministic flow; constructs three-dimensional scheduling decision variables for each deterministic flow; wherein the three-dimensional scheduling decision variables include the routing path selected from the feasible path set, the injection time offset, and a CSQF period specification set used to indicate that the deterministic flow is scheduled and forwarded at each hop of its routing path; with minimizing the objective function as the optimization objective, the three-dimensional scheduling decision variables of each deterministic flow are jointly solved for scheduling, resulting in routing and packet scheduling schemes for each deterministic flow that satisfy resource constraints, transmission constraints, end-to-end delay constraints, and period specification constraints; wherein the objective function is used to achieve dual load balancing of links and periodic windows based on continuous CSQF periods; calculates the remaining network resources of the power deterministic network based on the routing and packet scheduling schemes, and based on the network... The remaining network resources are allocated to nondeterministic flows in the traffic model. By introducing a periodic window and making periodic window load balancing one of the core optimization objectives, it forms a heavy load balancing system together with traditional link load balancing, which can significantly reduce jitter in deterministic data transmission and provide stable deterministic quality of service. The three-dimensional joint optimization of routing path selection, injection time offset planning, and CSQF periodic specified set scheduling can significantly improve network resource utilization and flow scheduling success rate. Utilizing the queue redundancy characteristics of the CSQF mechanism, periodic specified constraints are used to explicitly plan a periodic specified set for each flow. It can effectively absorb transmission delay variations and enhance network robustness; it performs joint scheduling and resource reservation for deterministic services to ensure their hard service quality (latency, jitter, zero packet loss); on this basis, it accurately calculates and outputs the remaining network resources for the transmission of nondeterministic (best-effort) services, enabling the coexistence of deterministic and nondeterministic services and efficient resource sharing; and through the design of three-dimensional joint scheduling and dual load balancing, it can achieve low jitter, high resource utilization, and strong robust deterministic power network transmission while ensuring strict latency limits, providing reliable communication support for new power systems.

[0023] Additional advantages, objectives, and features of this application will be set forth in part in the description which follows, and will in part become apparent to those skilled in the art upon review of the following description, or may be learned by practice of the application. The objectives and other advantages of this application can be realized and obtained by means of the structures specifically pointed out in the specification and drawings.

[0024] Those skilled in the art will understand that the purposes and advantages that can be achieved with this application are not limited to those specifically described above, and that the above and other purposes that this application can achieve will be more clearly understood from the following detailed description. Attached Figure Description

[0025] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, do not constitute a limitation thereof. The components in the drawings are not drawn to scale but are merely for illustrating the principles of this application. For ease of illustration and description of certain parts of this application, corresponding portions in the drawings may be enlarged, i.e., may appear larger relative to other components in an exemplary device actually manufactured according to this application. In the drawings: Figure 1 This is a schematic diagram of the first process of a power deterministic network routing and group scheduling method in one embodiment of this application.

[0026] Figure 2 This is a schematic diagram of a second process for a power deterministic network routing and group scheduling method in one embodiment of this application.

[0027] Figure 3 This is a flowchart illustrating the tabu search iteration steps in a power deterministic network routing and group scheduling method according to an embodiment of this application.

[0028] Figure 4 This is a schematic diagram of the structure of a power deterministic network routing and packet scheduling device in one embodiment of this application.

[0029] Figure 5 This is a schematic diagram of the scheduling process of the power deterministic network routing and group scheduling method in an application example of this application.

[0030] Figure 6 This is a schematic diagram of the architecture of a power deterministic network system in one embodiment of this application. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and their descriptions are used to explain this application, but are not intended to limit it.

[0032] It should also be noted that, in order to avoid obscuring this application with unnecessary details, only the structures and / or processing steps closely related to the solution according to this application are shown in the accompanying drawings, while other details that are not closely related to this application are omitted.

[0033] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0034] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0035] In the following description, embodiments of the present application will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0036] It should be noted that Deterministic Networking (DetNet) is a technology framework developed by the IETF working group to address the problem of deterministic transmission in IP networks. It aims to provide end-to-end deterministic services with bounded latency, low jitter, and zero packet loss for real-time applications in wide area networks. DetNet combines mechanisms such as clock synchronization, resource reservation, path selection, and traffic shaping to provide dedicated deterministic paths and bandwidth for data streams, ensuring the predictability of data packets during transmission.

[0037] DetNet's core mechanism lies in traffic scheduling. Through path constraints and bandwidth reservation, DetNet provides fixed transmission resources for critical service flows, avoiding competition with non-deterministic service flows. Simultaneously, DetNet can utilize Periodically Specified Circular Queuing Forwarding (CSQF) to allocate independent scheduling queues for deterministic service flows on transmission nodes and employ periodic scheduling to precisely control the transmission time slots of data packets, thereby providing end-to-end deterministic services for high real-time requirements.

[0038] CSQF evolved from CQF in Layer 2 Time-Sensitive Networking (TSN) technology. In TSN, clocks are strictly synchronized, and the network's time period is divided into even and odd periods. CQF provides deterministic latency and jitter for services through fixed-period even and odd queues. Data packets are sent from upstream nodes and received at downstream nodes within one period, and are forwarded to the next hop in the next period. Therefore, its end-to-end latency depends only on the period size T and the number of hops H.

[0039] To adapt to the characteristics of Layer 3 IP / MPLS networks, DetNet extended CQF to CSQF. Since link latency and processing latency are significant in large-scale networks, clock cycles are difficult to synchronize strictly and can only achieve frequency synchronization. Therefore, CQF cycles cannot be strictly aligned, and reserving more bandwidth as a guard band would significantly reduce effective bandwidth. Thus, in CSQF, each output interface uses three or more queues to absorb jitter, with only one queue in the transmission state during the same cycle. When a data packet arrives at a network node, different queues can be selected to delay for one or more cycles before forwarding to the next hop. Its end-to-end latency still depends only on the cycle size and the number of hops in the path.

[0040] Research on DetNet stream scheduling mainly includes: 1) Analyze the situation where changes in transmission delay cause packets to arrive early or late, and propose using four circular queues and the Paternoster algorithm to satisfy deterministic packet transmission; 2) Combining pre-routing and injection time planning (ITP) to shape traffic at the network edge, a bounded end-to-end delay and zero packet loss transmission method is proposed to meet the requirements of large-scale network flow scheduling. 3) A DetNet and TSN cross-domain scheduling method that satisfies the end-to-end worst-case latency requirement by calculating the worst-case latency of the flow through the DetNet domain based on the network computation curve; 4) A deterministic flow scheduling method based on TAS in an edge-side asynchronous TSN by introducing network computation service curves and constraint programming.

[0041] However, while Scheme 1) uses four circular queues to meet CSQF requirements, it assumes that the transmitting node can send packets equivalent to two cycles of data within one cycle, which is not suitable for the DetNet scenario. Furthermore, the algorithm only proposes a queue switching mechanism without incorporating injection time and routing for scheduling, which can lead to flow convergence collisions and link congestion in the core network for high-priority deterministic flows. Scheme 2) combines routing and injection time for flow scheduling, but ignores the issue of packets arriving at the next hop prematurely or delayed due to link and processing delays, thus failing to guarantee deterministic packet transmission. Scheme 3) considers the packet delay caused by processing delays and uses network calculus curves to calculate the worst-case delay, but it only guarantees bounded delays for deterministic flows and not jitter requirements for deterministic transmission, and it does not combine routing and injection time for flow scheduling. Scheme 4) above uses network calculus curves to calculate worst-case flow scheduling in asynchronous TSNs at the edge, which can guarantee deterministic delay and jitter of the flow. However, the gate list configuration of TAS is too complex in large-scale networks, and each hop in DetNet is not strictly synchronous. This method cannot meet the deterministic jitter requirement in CSQF.

[0042] In other words, existing technologies have two main problems: First, they only consider the latency limits of deterministic services, while ignoring jitter and nondeterministic transmission problems caused by changes in transmission latency; second, they do not fully utilize the period specification feature of CSQF and combine it with methods such as injection time planning and joint routing scheduling to expand the solution space, thereby improving flow schedulability and network resource utilization.

[0043] Based on this, in order to address the problems of existing power deterministic network flow scheduling methods ignoring jitter caused by transmission delay variations and failing to fully utilize the periodic characteristics of CSQF for joint optimization, making it difficult to simultaneously guarantee low latency, low jitter, and high resource utilization, this application provides a power deterministic network routing and packet scheduling method, a power deterministic network routing and packet scheduling device for executing the power deterministic network routing and packet scheduling method, a physical device, a computer-readable storage medium, a computer program product, and a power deterministic network system. When allocating resources for deterministic flows, the method considers dual load balancing of links and periods, and adjusts the forwarding period under deterministic latency conditions, thereby reducing jitter caused by link delay and processing delay, and eliminating the impact of deterministic flows arriving in the wrong period on the network.

[0044] The following examples will provide a detailed description.

[0045] Based on this, embodiments of this application provide a power deterministic network routing and packet scheduling method that can be implemented by a power deterministic network routing and packet scheduling device. See [link to relevant documentation]. Figure 1 The power deterministic network routing and packet scheduling method specifically includes the following: Step 100: Based on the network topology model and flow model of the power deterministic network, for each deterministic flow with end-to-end delay constraints in the deterministic flow set in the flow model, determine the feasible path from the source node to the destination node for each deterministic flow, forming a set of feasible paths for each deterministic flow.

[0046] Understandably, the feasible path set refers to the set of all potential transmission paths that are physically connected and meet the maximum tolerable latency requirements of the service flow; it is one of the fundamental solution spaces for subsequent optimization. The traffic model is an abstraction describing the characteristics and classifications of data flows transmitted in the network, used to distinguish services with different service requirements. The source node is the network device that generates and sends the service flow data. The destination node is the network device that receives the service flow data. The feasible path is a potential transmission route from the source node to the destination node that satisfies basic connectivity and whose theoretical transmission latency does not exceed the end-to-end latency constraint specified by the flow.

[0047] In one or more embodiments of this application, the deterministic flow refers to a data flow with periodic characteristics that requires the network to provide bounded latency, low jitter, and zero packet loss guarantees. The set of deterministic flows is the collection of all deterministic flows. The end-to-end latency constraint is the maximum time limit allowed for the transmission of data packets of a deterministic flow from the source node to the destination node.

[0048] Specifically, in step 100, the power deterministic network routing and packet scheduling device reads the network topology (such as nodes and links) and the declared deterministic service flow information. For each service flow, the power deterministic network routing and packet scheduling device runs the K-shortest path algorithm to find the top K paths with the shortest delay from the origin to the destination. Subsequently, based on the end-to-end delay limit, all paths that do not meet the delay requirements are filtered out, and the remaining paths constitute the feasible path set for that flow.

[0049] Step 200: Construct three-dimensional scheduling decision variables for each of the deterministic flows; wherein the three-dimensional scheduling decision variables include a routing path selected from the set of feasible paths, an injection time offset, and a CSQF period specification set for indicating that the deterministic flow is scheduled to be forwarded at each hop of its routing path.

[0050] It should be noted that the CSQF period specification set is a core concept in the period-specified circular queuing forwarding mechanism CSQF. It is an ordered integer sequence that specifies the precise transmission period number of the data packet at each hop in the transmission path. It is a key control parameter for achieving end-to-end deterministic latency and low jitter.

[0051] Specifically, in step 200, the power deterministic network routing and packet scheduling device defines three decision values ​​to be determined for each deterministic flow: 1) selecting a final path from the set of feasible paths; 2) selecting a start time point (i.e., injecting a time offset) within the flow's transmission period; 3) generating a period specification set, which is a list, where the i-th number in the list indicates in which period-specified cyclic queuing forwarding (CSQF) period the data packets of the flow should be scheduled to be sent when they pass through the i-th device on the selected path.

[0052] Step 300: With minimizing the objective function as the optimization objective, perform joint scheduling solution on the three-dimensional scheduling decision variables of each deterministic flow to obtain the routing and packet scheduling schemes that satisfy resource constraints, transmission constraints, end-to-end delay constraints and period specification constraints for each deterministic flow; wherein, the objective function is used to achieve dual load balancing of links and periodic windows based on continuous CSQF periods.

[0053] It should be noted that the joint scheduling solution refers to the process of simultaneously and collaboratively determining the three-dimensional decision variables for all flows in the network, rather than allocating them in isolation or sequentially. The aim is to obtain a globally optimal or near-optimal scheduling scheme. The dual load balancing is the core optimization objective. The first layer is traditional link load balancing, which avoids overloading some links. The second layer is the periodic window load balancing proposed in this application, which aims to distribute traffic evenly across different time windows, thereby directly reducing queuing delays and jitter caused by traffic aggregation over time.

[0054] Specifically, in step 300, the power deterministic network routing and packet scheduling device, guided by minimizing the objective function, performs a global comprehensive calculation of the three-dimensional scheduling decision variables for all flows. The solution process must ensure that the final scheme formed by the three decision values ​​selected for each flow simultaneously satisfies four categories of constraints: queue capacity limitations (resource constraints), port transmission capacity limitations (transmission constraints), total flow latency not exceeding limits (end-to-end latency constraints), and adjacent hop forwarding cycle offsets within [1,2] cycles (cycle specification constraints). The objective function in this step is specifically designed to simultaneously optimize load balancing between links and load balancing between CSQF cycle windows.

[0055] Step 400: Calculate the remaining network resources of the power deterministic network based on the routing and packet scheduling scheme, and allocate transmission resources to the nondeterministic flows in the traffic model based on the remaining network resources.

[0056] Specifically, in step 400, after calculating a deterministic scheduling scheme and reserving the necessary resources for all deterministic flows, the power deterministic network routing and packet scheduling device accurately calculates the remaining unused bandwidth resources on each link in the network. Subsequently, these remaining resources are dynamically allocated to nondeterministic flows that do not require strict latency guarantees, thereby maximizing the overall utilization of network resources.

[0057] In one or more embodiments of this application, nondeterministic flow may also be referred to as best-effort service, i.e. BE flow.

[0058] As described above, the power deterministic network routing and packet scheduling method provided in this application introduces a periodic window and uses periodic window load balancing as one of the core optimization objectives. Together with traditional link load balancing, it constitutes heavy load balancing, which can significantly reduce jitter in deterministic data transmission and provide stable deterministic quality of service. By performing three-dimensional joint optimization of routing path selection, injection time offset planning, and CSQF periodic specified set scheduling, it can significantly improve network resource utilization and flow scheduling success rate. Utilizing the queue redundancy characteristics of the CSQF mechanism, it specifies the period for each flow through periodic constraints. The designated set can effectively absorb transmission delay variations and enhance network robustness; it performs joint scheduling and resource reservation for deterministic services to ensure their hard service quality (latency, jitter, zero packet loss). On this basis, it accurately calculates and outputs the remaining network resources for the transmission of nondeterministic (best-effort) services, enabling the coexistence of deterministic and nondeterministic services and efficient resource sharing. Furthermore, through the design of three-dimensional joint scheduling and dual load balancing, it can achieve low jitter, high resource utilization, and strong robustness in power deterministic network transmission while ensuring strict latency limits, providing reliable communication support for new power systems.

[0059] To further address the fundamental problem of infeasible or inefficient scheduling schemes due to unclear network connectivity descriptions, and to ensure that the entire scheduling algorithm is built on an accurate and reliable network foundation, this application provides a power deterministic network routing and group scheduling method, see [link to relevant documentation]. Figure 2 The method for deterministic power network routing and packet scheduling includes the following steps prior to step 100: Step 010: Construct the network topology model of the power deterministic network; wherein, the network topology model includes: a directed graph, wherein the points in the directed graph represent transmission nodes in the power deterministic network, and the directed edges in the directed graph represent a unidirectional communication link between two transmission nodes; the transmission nodes include source nodes, relay nodes and destination nodes.

[0060] Specifically, the network topology model can be a directed graph G(V,E), where V represents the set of transmission nodes, encompassing the source, relay, and destination nodes of all flows in the network. E represents the set of directed edges, with each directed edge representing a unidirectional communication link between two transmission nodes. Specifically, for any two directly connected transmission nodes... and The physical links between them are represented in the model as two directed edges: and ; Nodes may be interconnected through underlying network technologies such as Time-Sensitive Networking (TSN) or Multiprotocol Label Switching (MPLS). To simplify the model, these connections are abstracted as virtual full-duplex links with deterministic transmission delays.

[0061] As can be seen from the above description, the power deterministic network routing and group scheduling method provided in this application provides accurate and unambiguous structured input for subsequent path calculation and resource scheduling by clearly defining and constructing a network topology model based on a directed graph. This solves the fundamental problem that scheduling schemes are infeasible or inefficient due to unclear network connection relationships, and ensures that the entire scheduling algorithm is built on an accurate and reliable network foundation.

[0062] To further address the issues of inconsistent parameter descriptions and unclear service level distinctions among different types of service flows under a unified scheduling framework, this application provides a power deterministic network routing and packet scheduling method, see [link to relevant documentation]. Figure 2 The method for deterministic power network routing and packet scheduling includes the following steps prior to step 100: Step 020: Construct the flow model of the power deterministic network; wherein the flow model includes: a set of deterministic flows for storing deterministic flows and a set of nondeterministic flows for storing nondeterministic flows.

[0063] Specifically, deterministic streams are periodic and carry services with strict time-sensitive requirements, needing to guarantee maximum end-to-end latency, jitter, and packet loss rate. Nondeterministic streams (i.e., BE streams) are non-periodic random data transmissions with unknown arrival times and data volumes, and do not promise any performance metrics regarding latency, jitter, or packet loss rate.

[0064] The deterministic flow is defined by a first tuple, which includes: the source node, destination node, end-to-end latency limit, transmission period, maximum tolerable jitter, total amount of data transmitted in one period, set of feasible paths, routing and packet scheduling scheme, binary variables, and priority; the value of the binary variable is 1 when the flow is successfully scheduled, otherwise it is 0.

[0065] Specifically, deterministic flow Defined as .in, and Indicates the source and destination nodes of the stream. This is the upper limit of end-to-end delay (i.e., the maximum end-to-end delay). For the sending period, To maximize the tolerance for jitter, This represents the total amount of data sent within a period. `pathset` is a set of feasible paths generated by an algorithm stored in a red-black tree; its element `path` is represented as an array of paths. ]. This refers to routing and packet scheduling schemes (also known as flow scheduling schemes). Binary variables. A value of 1 indicates that the flow has been successfully scheduled; otherwise, it is 0. The priority (i.e., flow priority) is defined as the priority of deterministic flows, which have a higher priority than nondeterministic flows.

[0066] The nondeterministic flow is defined by a second tuple, which includes: the source node, the destination node, the total amount of data sent within a period, the set of feasible routes, and the priority. BE flows have no periodicity or strict latency jitter limitations, requiring only "best-effort" transmission; therefore, a BE flow is defined as follows: .

[0067] It should be noted that the set of nondeterministic flows refers to the collection of all best-effort type service flows in the network that do not require strict latency and jitter guarantees. The first tuple and the second tuple are structured data objects defined to describe the complete characteristics of a flow, consisting of multiple predefined attribute fields, each storing parameters for a specific aspect of the flow. The first tuple describes deterministic flows and includes more stringent QoS parameters; the second tuple describes nondeterministic flows with simpler parameters. The maximum tolerable jitter is the maximum acceptable end-to-end latency fluctuation range for the service, a key indicator for measuring transmission stability. The binary variable is a flag variable with a value of 0 or 1. In this model, a value of 1 indicates that the flow has been successfully allocated resources by the algorithm and included in the scheduling scheme; a value of 0 indicates that it has not been scheduled or the scheduling has failed.

[0068] The scheduling objective of Periodically Specified Circular Queuing Forwarding (CSQF) is to generate a scheduling scheme for each deterministic flow. The scheme is a triple {offset, path, csset}. Here, is the injection time offset of the stream, which can not exceed the sending period of the stream. For routing schemes, The period specification set indicates in which period each hop of the packet should be sent. The algorithm combines injection time planning, routing selection, and the period specification set for flow scheduling, expanding the solution space and significantly improving flow schedulability and network resource utilization. After generating a deterministic flow scheduling scheme, the number of nondeterministic flows that can be accommodated is calculated based on the remaining network resources.

[0069] Specifically, in step 020, before the power deterministic network routing and packet scheduling device performs path calculation and scheduling, the system or management unit needs to classify and model all service data that needs to be transmitted in the network. First, a deterministic flow set and a non-deterministic flow set are created as the two main components of the flow model. For each service requiring deterministic service (such as distribution network remote control), its technical requirements are extracted, and a deterministic flow record is created. This record is stored in a first tuple data structure, containing the source node, destination node, maximum allowed end-to-end delay (end-to-end delay upper limit), periodic transmission time interval (transmission period), maximum tolerable delay variation (maximum tolerable jitter), total amount of data to be transmitted in each period, an initially empty set of feasible paths, a routing and packet scheduling scheme to be calculated, a binary variable initially set to 0 (used to mark whether it was ultimately successfully scheduled), and a high priority value. For other ordinary data services, a non-deterministic flow record is created and stored in a second tuple, containing its source node, destination node, data volume, set of feasible paths, and a lower priority value. The system explicitly states that deterministic flows always have a higher priority than any non-deterministic flows.

[0070] As can be seen from the above description, the power deterministic network routing and packet scheduling method provided in this application solves the problem of chaotic parameter descriptions and unclear service level distinctions for different types of service flows under a unified scheduling framework by accurately defining the structured composition and priority rules of the traffic model. It provides clear and computable input for subsequent scheduling algorithms, and can ensure that high-priority deterministic services can always obtain resource guarantees first, thereby realizing the orderliness and overall efficiency of network resource allocation.

[0071] To further address the issue of packets arriving at unexpected times due to transmission delay jitter potentially being dropped or causing cascading delays, this application provides a power deterministic network routing and packet scheduling method, see [link to relevant documentation]. Figure 2 The method for deterministic power network routing and packet scheduling includes the following steps prior to step 100: Step 030: Construct the transmission node model of the power deterministic network; wherein, the transmission node model is used to represent: each output port is configured with multiple priority queues, wherein the multiple queues with the highest priority are dedicated to scheduling the deterministic flow and periodically rotating and sending according to the CSQF mechanism or CSQF extended rules; the remaining queues are used to schedule the nondeterministic flow.

[0072] It's important to note that a priority queue is a data structure used to buffer data packets to be sent at the output port. Queues have a strict sending priority order, with higher-priority queues sending packets first. Round-robin sending refers to a scheduling method under the CSQF mechanism where multiple queues dedicated to deterministic flows take turns gaining sending rights according to a fixed, periodic schedule. Queue capacity is the maximum amount of data a queue can buffer. Port rate is the maximum theoretical data transmission capacity of a network port per unit of time, usually measured in bits per second (bps). An unexpected CSQF cycle is a situation where the actual CSQF cycle number of a deterministic flow packet arriving at a network node is inconsistent with the cycle number pre-assigned for that hop in the global scheduling scheme, usually caused by accumulated transmission delay fluctuations. The target deterministic queue with the earliest sending time refers to the queue that gains sending rights earliest among the next or more deterministic queues that are about to gain sending rights and can accommodate the delayed packet in the CSQF round-robin schedule.

[0073] The CSQF extension rules include: when a data packet of a deterministic flow arrives at the transmission node in an unexpected CSQF period due to transmission delay, and the transmission node supports PIFO queue management, the transmission node makes a determination based on the period specification set of the deterministic flow to which the data packet belongs and the current period; if it is determined that there exists a deterministic queue that satisfies the end-to-end delay constraint of the deterministic flow and is the next to be converted to the sending state, and the sum of the total amount of data to be sent in the current deterministic queue and the data amount of the data packet satisfies the sending constraint, then the operation of inserting the data packet into the tail of the deterministic queue is performed; otherwise, the data packet is discarded.

[0074] The resource constraint is used to indicate that, within any CSQF cycle, the total amount of deterministic stream data scheduled in any queue of any transmission node port does not exceed the capacity of that queue.

[0075] The sending constraint is used to indicate that, within any CSQF cycle, the total amount of deterministic stream data scheduled to be sent to any queue does not exceed the maximum amount of data that queue can send based on the port rate within a CSQF cycle.

[0076] In other words, the CSQF mechanism is the traditional CSQF rule. When a data packet of a deterministic flow arrives at the transmission node in an unexpected CSQF cycle, the data packet will be discarded directly.

[0077] The CSQF extension rule is a new CSQF rule that pushes packets to the tail of the deterministic queue based on PIFO, without affecting deterministic flows that should be forwarded in that cycle. If the total number of packets in the queue exceeds the data transmission limit for one cycle when the queue is about to become a sending queue, the excess packets (i.e., deterministic packets that arrived in unexpected CSQF cycles) will be discarded at the end of the cycle. The CSQF extension rule also imposes certain requirements on transmission nodes. This rule helps eliminate the uncertainty caused by transmission delays and is also linked to the dual load balancing algorithm design (if the cycle load is not heavy, nodes have a better chance of receiving and forwarding deterministic packets that arrived in unexpected CSQF cycles). However, this CSQF extension rule is not mandatory; the algorithm can function normally even without it.

[0078] In one specific implementation, the transmission node model designed in this application has eight priority queues, from queue 7 to queue 0. Queues 5-7 are used for deterministic services, and are turned on or off in turn according to the extended rules of the CSQF mechanism. The remaining queues are used for BE services, and BE services are transmitted when the deterministic service queue in the current cycle is empty. Without loss of generality, the port rate is denoted as c.

[0079] Deterministic queues follow a Push-In-First-Out (PIFO) rule. When a deterministic flow arrives during an error period due to variations in transmission delay, the transmission node determines whether the flow can arrive on time based on the period specification set and the remaining hops. If it can, it is pushed to the tail of the fastest forwarding queue, provided that queue constraints are met. Utilizing the multi-queue redundancy mechanism of CSQF, the forwarding period can be adjusted while meeting the delay deadline, eliminating the impact of flows arriving during error periods on other flows and ensuring the deterministic order of the network.

[0080] Let the total number of CSQF cycles be T, and let the p-th queue of the k-th port of the j-th transmission node be... The queue size is When the period is t, deterministic flow If the corresponding resource block is occupied, the mapping value will be... Set to 1 if the current value is 1, otherwise set to 0. When the period is t, the p-th queue of the k-th port of the j-th transmission node is opened, and the mapping value is then... Set to 1, otherwise set to 0.

[0081] As described above, the power deterministic network routing and packet scheduling method provided in this application solves the problem of packets being dropped or causing cascading delays due to unexpected arrival times caused by transmission delay jitter by constructing a refined transmission node queue model and clarifying the elastic forwarding rules of CSQF. This model effectively absorbs network jitter using queue redundancy and intelligent insertion mechanisms, ensuring reliable, low-jitter transmission of deterministic flows even in non-strict time synchronization environments.

[0082] To further address the challenge of efficiently and accurately pre-calculating all feasible transmission routes for service flows with strict latency limits in large and complex network topologies, this application provides a power deterministic network routing and packet scheduling method, see [link to relevant documentation]. Figure 2 Step 100 in the power deterministic network routing and packet scheduling method specifically includes the following: Step 110: For each deterministic flow in the deterministic flow set in the flow model of the power deterministic network, the K-shortest path algorithm is used to calculate the first K shortest paths from the source node to the destination node of the deterministic flow in the network topology model of the power deterministic network, forming an initial path set.

[0083] It's important to note that the K-shortest path algorithm is a graph theory algorithm used to find the top K shortest paths (sorted by total path cost) from a starting point to a destination in a directed graph. Here, "shortest" typically refers to the fewest hops or the smallest total link weight. The initial path set refers to the set of candidate paths that are topologically feasible and relatively optimal, obtained through preliminary calculations by the algorithm; it forms the basis for further refined selection.

[0084] Specifically, in step 110, for each service flow within the deterministic flow set in the traffic model, the power deterministic network routing and packet scheduling device runs the K-shortest path algorithm on the constructed network topology model (directed graph), with its source and destination nodes as the starting and ending points. This algorithm finds the first K shortest paths (e.g., hop count or link cost) from the starting point to the ending point. These paths constitute an initial path set for the flow, containing several relatively optimal potential transmission routes.

[0085] Step 120: Based on the end-to-end delay constraint of each deterministic flow and the preset single-hop transmission delay, select all paths that satisfy the end-to-end delay constraint from the initial path set as feasible paths, forming a set of feasible paths for each deterministic flow; wherein, a path satisfies the end-to-end delay constraint if the product of the number of hops of the path and the single-hop transmission delay is less than or equal to the end-to-end delay constraint.

[0086] It should be noted that single-hop transmission delay refers to the fixed time delay incurred when a data packet is transmitted over a unidirectional communication link. This typically includes signal propagation delay and device processing delay, but in this model, it is simplified to a preset constant value. The feasible path set refers to the set of all paths selected from the initial path set that satisfy both topological connectivity and service flow delay constraints. This set will serve as the decision range for subsequently selecting a specific route (path) for this flow.

[0087] Specifically, in step 120, after obtaining the initial path set, the power deterministic network routing and packet scheduling device performs a delay feasibility check on each path based on the flow-specific end-to-end delay constraints and the network's preset single-hop transmission delay parameters. The check formula is: path hop count × single-hop transmission delay ≤ end-to-end delay constraint. The expected transmission delay of each path is calculated, and only paths whose calculation results satisfy the above inequality are retained. All retained paths constitute the final feasible path set for the deterministic flow.

[0088] As described above, the power deterministic network routing and packet scheduling method provided in this application solves the problem of efficiently and accurately pre-calculating all feasible transmission routes for service flows with strict delay limits in large and complex network topologies through a two-stage path calculation method using K-shortest path and delay constraint screening. This method avoids the huge overhead of exhaustively enumerating all possible paths, quickly focusing on high-quality candidate paths that simultaneously meet the dual conditions of short path and sufficient delay, laying an efficient and reliable foundation for subsequent joint scheduling solutions.

[0089] To further address the shortcomings of existing scheduling methods in terms of flexibility when dealing with variations in transmission delay, and their inability to effectively plan hop-by-hop forwarding timing of packets while ensuring the delay ceiling, this application provides a power deterministic network routing and packet scheduling method. (See also...) Figure 2 Step 200 in the power deterministic network routing and packet scheduling method specifically includes the following: For each of the aforementioned deterministic flows, perform the following operations respectively: Step 210: Select a feasible path from the set of feasible paths of the deterministic flow as the routing path in the three-dimensional scheduling decision variables.

[0090] Specifically, for the currently processed deterministic flow, a path can be pre-selected from the previously calculated set of feasible paths for that flow as a candidate routing path. This selection operation is a necessary definition for constructing the three-dimensional variables. The final path selected will be determined in subsequent joint scheduling steps after comprehensively considering factors such as global load balancing.

[0091] Step 220: Determine an integer value for the deterministic flow within the interval [0, P) as the injection time offset in the three-dimensional scheduling decision variable, where P is the transmission period of the deterministic flow.

[0092] Specifically, an integer value can be selected as the injection time offset for the deterministic flow within its transmission period P. This value defines the time offset of the start of the first data packet transmission from the source node relative to the start of its transmission period. For example, if P = 100 microseconds, the offset can take an integer value between 0 and 99 microseconds.

[0093] Step 230: Generate an ordered set of integers of the same length as the number of hops in the routing path, as the period specification set in the three-dimensional scheduling decision variables; the period specification set satisfies the period specification constraint.

[0094] Wherein, the period specification constraint is used to indicate that: the i-th integer in the period specification set represents the CSQF period number offset of the deterministic flow being scheduled and forwarded at the i-th hop of the routing path, and for any two adjacent period offsets in the period specification set... and ,satisfy , This represents the total number of CSQF cycles. This indicates the modulo operation.

[0095] Specifically, an ordered list of integers of length equal to the number of hops in the selected route path can be generated as the period specification set for the flow. This set must satisfy the period specification constraint. Specifically, the i-th integer in the set (called the period offset) indicates which CSQF cycle a packet in the flow will be scheduled to forward on the i-th hop device of the path. This period number is the period number offset relative to the flow start time (defined by offset). The constraint requires that for any two adjacent offsets in the set, the difference modulo the total number of CSQF cycles N must be between 1 and 2. This means that packets are allowed to be forwarded to the next, or the next-next CSQF cycle, between adjacent hops, thus providing limited flexibility to absorb link delay jitter.

[0096] It's important to note that the cycle number offset is an integer value representing the CSQF cycle number in which an event (such as packet forwarding) occurs, and it's the offset relative to a certain time base (such as the start time of a flow). For example, an offset of 3 indicates that the 3rd CSQF cycle occurs after the base time. The modulo operation is a mathematical operation used to find the remainder after dividing two integers. In this constraint, it ensures that the cycle offset cycles within the range of 0 to (N-1), reflecting the cyclical nature of CSQF cycles. The total number of CSQF cycles N refers to the total number of discrete cycles contained in a complete scheduling cycle in the CSQF scheduling mechanism. For example, if N=8, the cycle number offset cycles between 0 and 7.

[0097] As described above, the power deterministic network routing and packet scheduling method provided in this application solves the problem that existing scheduling methods lack flexibility in dealing with changes in transmission delay and cannot effectively plan the hop-by-hop forwarding sequence of packets while ensuring the upper limit of delay, by precisely defining three-dimensional decision variables and setting periodic constraints. This design provides each flow with a clear and restricted degree of freedom for periodic hops, enabling the scheduling algorithm to absorb network jitter by adjusting the forwarding period. Thus, without affecting end-to-end latency, it lays a precise and controllable scheduling foundation for achieving low-jitter transmission.

[0098] To further address the challenges of traditional methods easily getting trapped in local optima and struggling to find high-quality feasible scheduling schemes under complex constraints and high-dimensional solution spaces, this application provides a power deterministic network routing and group scheduling method, see [link to relevant documentation]. Figure 2 Step 300 in the power deterministic network routing and packet scheduling method specifically includes the following: Step 310: Initialize the scheduled flow set to an empty set, the unscheduled flow set to contain all the aforementioned deterministic flows, and set the maximum number of iterations.

[0099] It should be noted that the scheduled flow set is used to temporarily store deterministic flows that have been assigned feasible scheduling schemes (i.e., whose three-dimensional variable values ​​are determined) in the current iteration during the algorithm's execution. The unscheduled flow set is used to temporarily store deterministic flows that have not yet been assigned feasible scheduling schemes, or those that were previously assigned but subsequently released. The maximum number of iterations is a preset positive integer used to control the upper limit of the tabu search algorithm's execution, preventing infinite loops, and is one of the conditions for the algorithm's termination. When the algorithm starts running, two sets are created: the scheduled flow set (initially empty) and the unscheduled flow set (initially containing all deterministic flows that need to be scheduled). At the same time, a maximum number of iterations is set as one of the algorithm's termination conditions.

[0100] Step 320: Repeat the preset tabu search iteration steps until the maximum number of iterations is reached or the unscheduled flow set is empty, and output the historical best routing and group scheduling scheme as the routing and group scheduling scheme.

[0101] It should be noted that the tabu search iterative steps refer to the standard sequence of operations performed by the tabu search algorithm in a single loop. This typically includes a perturbation and construction phase to generate new solutions, and an evaluation and update phase to assess and record the new solutions. The tabu search algorithm is a metaheuristic optimization algorithm. It introduces a tabu list to record recent search history, preventing the algorithm from repeatedly accessing already explored solutions in the short term. This effectively helps it escape local optima and guides the search towards new, promising regions to find the globally optimal or near-optimal solution.

[0102] Among them, see Figure 3 The tabu search iteration steps specifically include the following: Step 321: In the current iteration, if the set of scheduled flows is not empty, determine whether to perform a disturbance based on the preset release probability; if so, select at least one deterministic flow from the set of scheduled flows, release the network resources it occupies, and move it to the set of unscheduled flows.

[0103] Understandably, the release probability is a preset probability threshold (between 0 and 1) used to control the frequency with which the algorithm performs perturbation operations during iterations. It is a key parameter guiding the algorithm to balance concentrated search (utilizing the current good solution) and dispersed search (exploring new regions). In tabu search, perturbation refers to intentionally altering a portion of the current solution to escape the current local optimum. In this embodiment, it specifically refers to filtering out a portion of the scheduled flow from the current solution in the hope of finding a better overall combination in subsequent constructions.

[0104] Specifically, in each iteration, the algorithm first checks if the set of scheduled flows is not empty. If it is, a random number is generated and compared with a preset release probability (e.g., 0.6). If the random number is greater than this probability, a perturbation operation is performed: one or more deterministic flows are randomly selected from the set of scheduled flows, removed from the set, and all resources (such as link bandwidth and queue space) occupied by these flows in the simulated network are released. Then, these flows are put back into the set of unscheduled flows.

[0105] Step 322: For each deterministic flow in the set of unscheduled flows, determine a routing and packet scheduling scheme that simultaneously satisfies resource constraints, transmission constraints, end-to-end delay constraints, and period specification constraints from the selectable range of the three-dimensional scheduling decision variables.

[0106] It should be noted that, for each deterministic flow in the unscheduled flow set, the algorithm attempts to combine a specific scheduling scheme from the available range of three-dimensional scheduling decision variables (i.e., selecting a path from its feasible path set, selecting an injection time offset within its transmission period, and generating a period-specified set that satisfies the period-specified constraint). This scheme must simultaneously satisfy resource constraints, transmission constraints, and the flow's own end-to-end latency constraints. If such a feasible scheme can be found for a flow, it is added to the scheduled flow set, and corresponding resources are reserved for it in the simulated network; otherwise, it remains in the unscheduled flow set.

[0107] Step 323: Calculate the value of the objective function based on the routing and packet scheduling schemes of all deterministic flows in the current scheduled flow set; update the tabu table based on the value of the objective function, and update the historically optimal routing and packet scheduling scheme.

[0108] The tabu list is the core data structure of the tabu search algorithm, used to store recently forbidden (i.e., taboo) moves or changes in solutions. Its purpose is to prevent the algorithm from getting stuck in a local loop and to force the exploration of new search directions. The historical optimal routing and group scheduling scheme is the complete set of scheduling schemes that maximize (minimize) the objective function value found during the entire operation of the algorithm up to the current iteration.

[0109] After completing the scheduling attempts for all flows in this iteration, the algorithm simulates the network resource occupancy state based on the scheduling scheme determined by all flows in the currently scheduled flow set and calculates the value of the objective function. Then, the algorithm performs two updates: 1) Update the tabu table, recording operations that caused the objective function value to deteriorate in this iteration (such as releasing or scheduling a specific flow), and prohibiting such operations in the short term to avoid loops; 2) Update the historical best routing and group scheduling schemes, that is, compare the objective function value of the current solution with the historical best value, and replace the historical best record with the current solution if the current solution is better.

[0110] When the loop terminates (reaching the maximum number of iterations or the set of unscheduled flows is empty), the algorithm outputs the recorded historical best routing and group scheduling schemes as the final result.

[0111] As described above, the power deterministic network routing and group scheduling method provided in this application solves the problem that traditional methods are prone to getting trapped in local optima and struggling to find high-quality feasible scheduling schemes under complex constraints and high-dimensional solution spaces by employing a tabu search algorithm for global joint optimization. This algorithm intelligently balances depth search and breadth exploration, efficiently calculating scheduling schemes that satisfy all strict constraints and have excellent load balancing for deterministic service flows in large-scale networks, significantly improving scheduling success rate and network performance.

[0112] Based on the above, in one or more embodiments of this application, the formulas for each constraint condition are given as follows: (1) Resource constraints (also known as resource bandwidth constraints): The amount of resource blocks used by each queue of a transmission node cannot exceed the queue size. in, This refers to the deterministic flow during the CSQF period t. Whether the p-th queue of the k-th port of the j-th transmission node is occupied (1 if occupied, 0 otherwise); It is the i-th deterministic flow The total amount of data that needs to be sent within a sending cycle; It is the capacity of the queue (i.e., the maximum amount of data that can be stored).

[0113] (2) Sending constraints (also known as queue sending constraints): All packets in each queue of the transmission node should be sent within one cycle: in, It refers to the speed of the transmission port. It is the duration of a single CSQF cycle.

[0114] (3) End-to-end delay constraint (also referred to as delay constraint): The end-to-end delay of deterministic flow is only related to the injection time offset, the forwarding period of the last hop, and the size of the CSQF period. It is related to the path transmission delay and must not exceed the flow cutoff time. The transmission delay is a fixed constant; for simplicity, the transmission delay of each hop is denoted as . .

[0115] in, yes The actual end-to-end delay; yes The time offset injected into the scheduling scheme (i.e., the short name for routing and packet scheduling scheme); yes The last element of the period specified set in the scheduling scheme (i.e., the CSQF period offset of the last hop). yes The number of hops for the selected routing path in the scheduling scheme; This is the preset single-hop transmission delay; yes The upper limit of end-to-end latency.

[0116] (4) Period specification constraint: Deterministic flow can only choose to forward with a delay of 1 or 2 periods at each hop.

[0117] in, and yes In the scheduling scheme, the offset between two adjacent CSQF cycles in the cycle specification set (csset); % is the modulo operation; N is the total number of CSQF cycles; yes The first element of the period specified set in the scheduling scheme (i.e., the CSQF period offset of the first hop). It is a set of positive integers.

[0118] To further address the challenge of existing scheduling methods that focus solely on a single objective and fail to simultaneously optimize resource efficiency, spatial load, and temporal load while ensuring latency, this application provides a power deterministic network routing and group scheduling method. Before executing traffic routing scheduling, the algorithm generates a pre-calculated network path book based on the K-shortest path algorithm. When calculating routes, retrieve them from the pre-computed pathbook. arrive All routes that meet the latency requirements are sorted according to route weights designed based on Domain Specific Knowledge (DSK) strategies. DSK refers to professional knowledge specific to a particular domain or discipline. For example, in CSQF-based DetNet routing, prioritizing feasible paths with fewer hops reduces network resources consumed per unit flow. Strategies such as prioritizing routes with lower load rates and scheduling flows with larger packets are more effective in reducing the probability of deterministic flow collisions caused by link and processing delays, thus indirectly reducing flow jitter. Therefore, routing weight Represented as: in, , The constant coefficient, Let be the load rate of the j-th link. The three weighting factors represent: the packet size sent per cycle for this flow, the average load rate of the path, and the relative relationship between the path hop count and the number of network edges. Paths with higher routing weights are more likely to be selected in the subsequent tabu search algorithm.

[0119] The algorithm aims to minimize the jitter of deterministic streaming. Assume that for each transmitting port, n CSQF cycles can be divided into n windows, each containing two adjacent CSQF cycles. Let the load of the t-th cycle window of the k-th port of the j-th transmitting node be denoted as... : in, This represents the load of the k-th port of the j-th transmission node during the t-th period window. A period window consists of two consecutive CSQF periods. It is an indicator value for whether the p-th queue is an active send queue at the beginning of the t-th period window of the CSQF (1 if yes, 0 otherwise). It is an indicator value (0 or 1). When it is 1, it indicates that the deterministic flow is in the (t+1)%Nth CSQF cycle. The data is scheduled to occupy the p-th queue of the k-th port of the j-th transmission node; It is an indicator value (0 or 1). When it is 1, it means that in the (t+1)%Nth CSQF cycle, the pth queue of the kth port of the jth transmission node is the currently active queue that is allowed to send data (according to the CSQF queue rotation rules).

[0120] Under the same network traffic, deterministic flow consumes network resources. Smaller, better load balancing across different links Smaller, more balanced load distribution across different windows on the link The smaller the value, the lower the probability of deterministic flow collision delay, and the less jitter. The dual load balancing design of the link and window can effectively reduce jitter caused by variations in transmission delay while maintaining latency limits.

[0121] Based on this, the objective function (also known as the tabu search objective function) is shown in the following formula (I): Formula (1) in, The objective function is... The sum of network resources occupied by all scheduled deterministic flows is used to optimize network resource utilization while satisfying the end-to-end delay constraints. The standard deviation of the load rate of all routing links in the power deterministic network is used to measure the load balance between links. This is the average of the variances of the load rates for each period window across all links, used to measure the load balance between period windows. and This is a preset constant coefficient. That is, the larger the objective function value is, the more deterministic flows in the network consume network resources or the worse the load balancing effect.

[0122] in, The calculation formula is shown in Formula (II): Formula (II) Where F represents the set of deterministic flows, For the i-th deterministic flow A flag indicating whether the scheduling was successful; For the i-th deterministic flow Path hop count; For the i-th deterministic flow The amount of data; For the i-th deterministic flow The actual end-to-end delay; The calculation formula is shown in formula (III): Formula (3) in, The total number of links in the power deterministic network. Let the load rate of the j-th link be . The average load rate across all links; The calculation formula is shown in formula (iv): Formula (IV) in, Number of ports; A set of periodic windows; Let be the load of the k-th port of the j-th transmission node in the t-th period window; Let be the average load of the k-th port of the j-th transmission node over all periodic windows.

[0123] To further illustrate the above embodiments, this application also provides a specific application example of a deterministic network routing and packet scheduling method for power systems. This primarily addresses the problem in current DetNets where network nodes are not strictly synchronized, link delays and processing delays are not negligible, and deterministic flows struggle to guarantee deterministic jitter. Based on CSQF, this application utilizes its redundant queues to appropriately delay the forwarding period, combining routing selection, injection time planning, and periodic setting for scheduling. This achieves dual load balancing through routing and periodic windows, reducing the impact of delayed deterministic flows on other deterministic flows. Under the condition of satisfying deterministic latency, it significantly reduces the jitter of deterministic flows and provides deterministic quality of service support. See also... Figure 4 This application example first uses depth-first traversal to generate a stream. All feasible routes from src to dst are identified and sorted according to their route weights. For deterministic flows, the route with the lowest weight is selected, and a scheduling scheme for the deterministic flows is solved using tabu-based routing and packet scheduling algorithms. Finally, the remaining network resources are allocated to the BE flows, and an overall scheduling scheme is generated. Figure 4 In this context, F_{sche} represents the set of scheduled flows; F_{unsche} represents the set of unscheduled flows (also known as the set of feasible routes).

[0124] This application example solves the scheduling scheme based on tabu search, and combines route selection, injection time planning, and periodic specified set. The specific algorithm is shown in Table 1 below: Table 1 As shown in Table 1, this algorithm solves the scheduling scheme based on tabu search, and combines route selection, injection time planning, and periodic set specification. Specifically, it divides the flow set into a set of scheduled flows. and unscheduled stream sets and will flow The scheduling scheme is defined as a triple of injected time, path, and period specified sets, i.e. In each loop, attempt to move a portion of the unscheduled flow into the loop. Or move some of the scheduled flows out to When scheduling or releasing a flow, calculate based on the current network state. The weights of combinations of time, path, and period specified in the injection are sorted, and selection is made based on probability according to the weights. The system generates feasible scheduling schemes and updates network resource usage. Based on the objective function value of the scheduling scheme generated in each iteration, it updates the tabu table or the current solution.

[0125] In other words, this application example considers the non-negligible link and processing delays in large-scale deterministic networks and the uncertainty of DetNet stream transmission under non-strict time synchronization. It proposes a periodic window mechanism, and the algorithm considers dual load balancing of links and periodic windows. While meeting the deterministic latency requirements of DetNet streams, it significantly reduces jitter caused by transmission delay variations, ensuring the quality of service (QoS) requirements of deterministic streams. Utilizing the CSQF multi-queue redundancy mechanism, the forwarding period is appropriately adjusted while meeting the deterministic latency requirements, eliminating the impact of deterministic packets arriving at unexpected periods due to transmission delay variations on other deterministic packets. The algorithm combines routing selection, injection time planning, and periodic specification sets for scheduling, expanding the solution space and increasing solvability. The scheduling scheme achieves low latency and low jitter for deterministic streams, meeting the scheduling requirements of DetNet streams, and considering reducing the network resources occupied by deterministic streams, thus improving the schedulability of BE streams.

[0126] Therefore, the beneficial effects of the power deterministic network routing and packet scheduling method provided in the application examples of this application are as follows: 1. Compared with existing large-scale deterministic network flow scheduling models, this application presents a power deterministic network routing and packet scheduling model based on CSQF dual load balancing. It utilizes the redundancy mechanism of the CSQF traffic shaper to appropriately adjust the forwarding period based on PIFO under the condition of satisfying deterministic delay, thereby eliminating the impact of deterministic packets arriving at unexpected periods on other deterministic packets.

[0127] 2. This application example significantly reduces jitter caused by transmission latency variations while meeting the deterministic latency requirements of DetNet flows through dual load balancing of routing and windowing, thus satisfying the quality of service requirements for deterministic flows. Simultaneously, it reserves available network resources for deterministic flows arriving at unexpected times, increasing the likelihood of their on-time arrival.

[0128] 3. Compared with existing flow scheduling algorithms, this application presents a DetNet flow routing and packet scheduling method based on CSQF and dual load balancing. This algorithm combines route selection, injection time planning, and periodic set specification for scheduling, expanding the solution space and increasing solvability. Furthermore, it reduces solution time and improves application efficiency by using metaheuristic algorithms and the DSK strategy. Ultimately, it provides deterministic quality of service support for time-sensitive data flows in the Industrial Internet.

[0129] From a software perspective, this application also provides a power deterministic network routing and packet scheduling apparatus for executing all or part of the power deterministic network routing and packet scheduling method, see [link to relevant documentation]. Figure 5 The power deterministic network routing and packet scheduling device specifically includes the following components: The feasible path acquisition module 10 is used to determine the feasible path from the source node to the destination node for each deterministic flow with end-to-end time delay constraints in the deterministic flow set in the power deterministic network based on the network topology model and the flow model, thereby forming a set of feasible paths for each deterministic flow.

[0130] The decision variable construction module 20 is used to construct three-dimensional scheduling decision variables for each of the deterministic flows; wherein, the three-dimensional scheduling decision variables include a routing path selected from the set of feasible paths, an injection time offset, and a CSQF periodic specification set used to indicate that the deterministic flow is scheduled to be forwarded at each hop of its routing path.

[0131] The joint scheduling solution module 30 is used to perform joint scheduling solution on the three-dimensional scheduling decision variables of each deterministic flow with the goal of minimizing the objective function, so as to obtain the routing and packet scheduling schemes that satisfy resource constraints, transmission constraints, end-to-end delay constraints and period specification constraints for each deterministic flow; wherein, the objective function is used to achieve dual load balancing of links and periodic windows based on continuous CSQF periods.

[0132] The remaining resource allocation module 40 is used to calculate the remaining network resources of the power deterministic network based on the routing and packet scheduling scheme, and to allocate transmission resources to the nondeterministic flow in the traffic model based on the remaining network resources.

[0133] The embodiments of the power deterministic network routing and packet scheduling apparatus provided in this application can be used to execute the processing flow of the embodiments of the power deterministic network routing and packet scheduling method described above. Its functions will not be repeated here, but can be referred to the detailed description of the embodiments of the power deterministic network routing and packet scheduling method described above.

[0134] The power deterministic network routing and packet scheduling component of the aforementioned device can be performed in either a server or a client device. The choice can be made based on the processing capabilities of the client device and the limitations of the user's usage scenario. This application does not impose any limitations in this regard. If all operations are performed in the client device, the client device may further include a processor for the specific processing of power deterministic network routing and packet scheduling.

[0135] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0136] The server and the client device can communicate using any suitable network protocol, including those not yet developed as of the date of this application. Such network protocols may include, for example, TCP / IP, UDP / IP, HTTP, HTTPS, etc. Furthermore, such network protocols may also include RPC (Remote Procedure Call Protocol) and REST (Representational State Transfer Protocol) protocols used on top of the aforementioned protocols.

[0137] This application also provides an electronic device, which may include a processor, a memory, a receiver, and a transmitter. The processor is used to execute the power deterministic network routing and packet scheduling method mentioned in the above embodiments. The processor and memory can be connected via a bus or other means, taking a bus connection as an example. The receiver can be connected to the processor and memory via wired or wireless means.

[0138] The processor can be a central processing unit (CPU). The processor can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0139] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the power deterministic network routing and packet scheduling method in the embodiments of this application. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and modules stored in the memory, thereby implementing the power deterministic network routing and packet scheduling method in the above method embodiments.

[0140] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor, etc. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0141] The one or more modules are stored in the memory, and when executed by the processor, they perform the power deterministic network routing and packet scheduling method in the embodiment.

[0142] In some embodiments of this application, the user equipment may include a processor, a memory, and a transceiver unit. The transceiver unit may include a receiver and a transmitter. The processor, memory, receiver, and transmitter may be connected via a bus system. The memory is used to store computer instructions, and the processor is used to execute the computer instructions stored in the memory to control the transceiver unit to send and receive signals.

[0143] As one implementation method, the functions of the receiver and transmitter in this application can be implemented by transceiver circuits or dedicated transceiver chips, and the processor can be implemented by dedicated processing chips, processing circuits or general-purpose chips.

[0144] As another implementation approach, the server provided in this application embodiment can be implemented using a general-purpose computer. That is, the program code implementing the processor, receiver, and transmitter functions is stored in memory, and the general-purpose processor implements the processor, receiver, and transmitter functions by executing the code in memory.

[0145] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned power deterministic network routing and packet scheduling method. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0146] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the aforementioned power deterministic network routing and packet scheduling method.

[0147] This application also provides a power deterministic network system, see [link to relevant documentation]. Figure 6 It includes multiple transmission nodes supporting the CSQF forwarding mechanism in a power deterministic network and electronic devices; the electronic devices include a processor and a memory, the memory storing a computer program, which, when executed by the processor, implements the power deterministic network routing and packet scheduling method.

[0148] In the research and development and implementation of the technical solutions involved in this application, all user personal information (if any) was processed in strict accordance with the principles of legality, legitimacy, necessity, and good faith. Specifically, the relevant data was obtained through one or more of the following compliant methods: 1) Before collecting users' personal information, the purpose, method, and scope of the collection have been clearly communicated to the users, and the users' individual and explicit authorization and consent have been obtained; 2) The personal data used comes from publicly available datasets permitted by laws and regulations, and the personal data has undergone necessary anonymization or de-identification processing during use to ensure that no specific individual can be identified and the information is irretrievable; 3) The use of personal data is limited to the technical research and development, model training and verification purposes described in this application, and strict technical and management measures have been taken to protect data security and prevent information leakage, abuse and unauthorized access.

[0149] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. The programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave.

[0150] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0151] In this application, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0152] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to the embodiments of this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for routing and packet scheduling in a deterministic power network, characterized in that, include: Based on the network topology model and flow model of the power deterministic network, for each deterministic flow with end-to-end delay constraints in the deterministic flow set in the flow model, a feasible path from the source node to the destination node is determined for each deterministic flow, forming a set of feasible paths for each deterministic flow. Each of the deterministic flows is constructed with its own three-dimensional scheduling decision variables; wherein, the three-dimensional scheduling decision variables include a route path selected from the set of feasible paths, an injection time offset, and a CSQF period specification set for indicating that the deterministic flow is scheduled to be forwarded at each hop of its route path; With minimizing the objective function as the optimization objective, the three-dimensional scheduling decision variables of each deterministic flow are jointly solved to obtain the routing and packet scheduling schemes that satisfy resource constraints, transmission constraints, end-to-end delay constraints, and period specification constraints for each deterministic flow; wherein, the objective function is used to achieve dual load balancing of links and periodic windows based on continuous CSQF periods; The remaining network resources of the power deterministic network are calculated based on the routing and packet scheduling scheme, and transmission resources are allocated to the nondeterministic flows in the traffic model based on the remaining network resources.

2. The power deterministic network routing and packet scheduling method according to claim 1, characterized in that, Before determining the feasible path from the source node to the destination node for each of the deterministic flows, the method further includes: Construct a network topology model for the power deterministic network; wherein the network topology model includes: a directed graph, wherein the points in the directed graph represent transmission nodes in the power deterministic network, and the directed edges in the directed graph represent a unidirectional communication link between two transmission nodes; the transmission nodes include source nodes, relay nodes, and destination nodes.

3. The power deterministic network routing and packet scheduling method according to claim 1, characterized in that, Before determining the feasible path from the source node to the destination node for each of the deterministic flows, the method further includes: Construct a flow model for the power deterministic network; wherein the flow model includes: a set of deterministic flows for storing deterministic flows and a set of nondeterministic flows for storing nondeterministic flows; The deterministic flow is defined by a first tuple, which includes: the source node, destination node, end-to-end latency limit, transmission period, maximum tolerable jitter, total amount of data transmitted in one period, set of feasible paths, routing and packet scheduling scheme, binary variables, and priority; the binary variable takes a value of 1 to indicate that the flow has been successfully scheduled, otherwise it takes a value of 0. The deterministic flow has a higher priority than the non-deterministic flow; The nondeterministic flow is defined by a second tuple, which includes: the source node of the flow, the destination node, the total amount of data sent in a period, the set of feasible routes, and the priority.

4. The power deterministic network routing and packet scheduling method according to claim 1, characterized in that, Before determining the feasible path from the source node to the destination node for each of the deterministic flows, the method further includes: Construct a transmission node model for the power deterministic network; wherein the transmission node model represents that: each output port is configured with multiple priority queues, wherein the highest priority queues are dedicated to scheduling the deterministic flow and perform periodic round-robin transmission following the CSQF mechanism or CSQF extended rules; the remaining queues are used to schedule the nondeterministic flow; The resource constraint is used to indicate that, within any CSQF cycle, the total amount of deterministic stream data scheduled in any queue of any transmission node port does not exceed the capacity of that queue. The sending constraint is used to indicate that, in any CSQF cycle, the total amount of deterministic stream data scheduled to be sent to any queue does not exceed the maximum amount of data that the queue can send based on the port rate in one CSQF cycle; The CSQF extension rules include: when a data packet of a deterministic flow arrives at the transmission node in an unexpected CSQF period due to transmission delay, and the transmission node supports PIFO queue management, the transmission node makes a determination based on the period specification set of the deterministic flow to which the data packet belongs and the current period; if it is determined that there exists a deterministic queue that satisfies the end-to-end delay constraint of the deterministic flow and is the next to be converted to the sending state, and the sum of the total amount of data to be sent in the current deterministic queue and the data amount of the data packet satisfies the sending constraint, then the operation of inserting the data packet into the tail of the deterministic queue is performed; otherwise, the data packet is discarded.

5. The power deterministic network routing and packet scheduling method according to claim 1, characterized in that, Based on the network topology model and flow model of the power deterministic network, for each deterministic flow with end-to-end delay constraints within the deterministic flow set in the flow model, a feasible path from the source node to the destination node is determined for each deterministic flow, forming a set of feasible paths for each deterministic flow, including: For each deterministic flow within the deterministic flow set in the flow model of the power deterministic network, the K-shortest path algorithm is used to calculate the first K shortest paths from the source node to the destination node of the deterministic flow in the network topology model of the power deterministic network, forming an initial path set; Based on the end-to-end delay constraint of each deterministic flow and the preset single-hop transmission delay, all paths that satisfy the end-to-end delay constraint are selected from the initial path set as feasible paths, forming a set of feasible paths for each deterministic flow; wherein, a path satisfies the end-to-end delay constraint if the product of the number of hops of the path and the single-hop transmission delay is less than or equal to the end-to-end delay constraint.

6. The power deterministic network routing and packet scheduling method according to claim 1, characterized in that, The construction of the three-dimensional scheduling decision variables for each of the deterministic flows includes: For each of the aforementioned deterministic flows, perform the following operations respectively: Select a feasible path from the set of feasible paths of the deterministic flow as the routing path in the three-dimensional scheduling decision variables; An integer value is determined for the deterministic flow within the interval [0, P) as the injection time offset in the three-dimensional scheduling decision variable, where P is the transmission period of the deterministic flow; Generate an ordered set of integers of the same length as the number of hops in the routing path, as the period specification set in the three-dimensional scheduling decision variables; the period specification set satisfies the period specification constraint; Wherein, the period specification constraint is used to indicate that: the i-th integer in the period specification set represents the CSQF period number offset of the deterministic flow being scheduled and forwarded at the i-th hop of the routing path, and for any two adjacent period offsets in the period specification set... and ,satisfy , This represents the total number of CSQF cycles. This indicates the modulo operation.

7. The power deterministic network routing and packet scheduling method according to claim 1, characterized in that, The process involves jointly solving the three-dimensional scheduling decision variables of each deterministic flow, with the goal of minimizing the objective function and satisfying the end-to-end delay constraint, to obtain the routing and grouping scheduling schemes for each deterministic flow. This includes: Initialize the set of scheduled flows to an empty set, the set of unscheduled flows to contain all the aforementioned deterministic flows, and set the maximum number of iterations; Repeat the preset taboo search iteration steps until the maximum number of iterations is reached or the unscheduled flow set is empty, and output the historical best routing and group scheduling scheme as the routing and group scheduling scheme; The tabu search iteration steps include: In the current iteration, if the set of scheduled flows is not empty, a perturbation is performed based on a preset release probability; if performed, at least one deterministic flow is selected from the set of scheduled flows, its network resources are released, and it is moved to the set of unscheduled flows. For each deterministic flow in the set of unscheduled flows, a routing and packet scheduling scheme is determined for the deterministic flow from the selectable range of the three-dimensional scheduling decision variables, which simultaneously satisfies resource constraints, transmission constraints, end-to-end delay constraints, and period specification constraints; Calculate the value of the objective function based on the routing and packet scheduling schemes of all deterministic flows in the current scheduled flow set; update the tabu table based on the value of the objective function, and update the historically optimal routing and packet scheduling scheme.

8. The power deterministic network routing and packet scheduling method according to any one of claims 1 to 7, characterized in that, The objective function is shown in Formula (I) below: Formula (1) in, The objective function is... The sum of network resources occupied by all scheduled deterministic flows is used to optimize network resource utilization while satisfying the end-to-end delay constraints. The standard deviation of the load rate of all routing links in the power deterministic network is used to measure the load balance between links. This is the average of the variances of the load rates for each period window across all links, used to measure the load balance between period windows. and These are preset constant coefficients.

9. The power deterministic network routing and packet scheduling method according to claim 8, characterized in that, The calculation formula is shown in Formula (II): Formula (II) Where F represents the set of deterministic flows, For the i-th deterministic flow A flag indicating whether the scheduling was successful; For the i-th deterministic flow Path hop count; For the i-th deterministic flow The amount of data; For the i-th deterministic flow The actual end-to-end delay; The calculation formula is shown in formula (III): Formula (3) in, The total number of links in the power deterministic network. Let j be the load rate of the j-th link. The average load rate across all links; The calculation formula is shown in formula (iv): Formula (IV) in, Number of ports; A set of periodic windows; Let be the load of the k-th port of the j-th transmission node in the t-th period window; Let be the average load of the k-th port of the j-th transmission node over all periodic windows.

10. A power deterministic network system, characterized in that, include: Multiple transmission nodes and electronic devices supporting the CSQF forwarding mechanism in a power deterministic network; The electronic device includes a processor and a memory, the memory storing a computer program that, when executed by the processor, implements the power deterministic network routing and packet scheduling method as described in any one of claims 1 to 9.

Images