A topology identification method and device for adapting to a multi-energy power supply network in a park

Abstract

The present invention relates to a topology identification method and device for adapting to a multi-energy power supply network in a park, and belongs to the field of communication technology. The present invention proposes a method for jointly optimizing the data packet transmission delay and packet loss rate of topology identification based on a confidence interval algorithm on reinforcement learning, and performs topology identification and joint optimization of the data packet transmission delay and packet loss rate of topology identification from two time scales. Topology identification is performed on a small time scale, and the data packet transmission delay and packet loss rate of topology identification are jointly optimized on a large time scale. The topology identification result affects the update of the preference value of the coding scheme, and the update result of the preference value will react on the selection of the coding strategy in the next period, thereby achieving continuous improvement of the topology identification performance.

Description

Technical Field

[0001] The invention relates to a topology identification method and a device for adapting a park multi-energy power supply network, and belongs to the technical field of communication. Background Art

[0002] Nowadays, the park needs to rationally plan multiple forms of energy such as wind, light, electricity, geothermal, etc. to form a multi-level energy ladder to provide users with the most cost-effective comprehensive energy services. In order to achieve the above goals, a large number of distributed energy sources are connected to the park power supply system to form a multi-energy power supply network in the park. Compared with the traditional energy network, the multi-energy power supply of the park needs to consider the distribution, conversion and coordination of various distributed energy sources in the park. By establishing a multi-energy coupling system, the complementary characteristics of multiple energy sources in the park are analyzed, and through the interaction between the park and users, load regulation based on demand response and distributed energy storage charging and discharging scheduling are carried out.

[0003] In order to comprehensively analyze the multi-energy coupling relationship in the park, the park dispatch center needs to obtain the geographical location, connection mode, operating status and other information of each node in the power supply network in the park in real time. Therefore, it is necessary to realize massive, high-reliability and low-latency data communication between the equipment terminals in the park and the park dispatch center. In addition, under the background of a large number of terminals and massive sensor access, the power supply network in the park has the characteristics of multiple points, wide areas and complex structure, and it is very difficult to select the optimal data transmission path. In order to support the park to establish a reasonable data transmission plan and ensure high-reliability and low-latency communication between the equipment terminals in the park, it is necessary to accurately identify the topology of the current multi-energy power supply network in the park. Topology identification refers to the identification and analysis of the connection relationship and operating status between power sources, loads and energy storage devices in the network during the flow of electric energy, with the aim of providing physical support for data communication of energy services in the park. Among them, power line carrier communication modulates data information into high-frequency signals and then couples them to power lines for transmission. It is a communication method that can achieve high-speed data transmission without the need to set up additional communication lines. Its cost is greatly reduced compared with communication technologies such as optical fiber and 5G, and it has become the preferred communication solution for topology identification in the current park.

[0004] However, the current topology identification technology of multi-energy power supply networks in campuses still faces the following challenges:

[0005] First, distributed energy has a certain degree of randomness and intermittency in output, which causes the voltage of the park's multi-energy power supply network to fluctuate greatly and the power quality to decline, causing serious attenuation of the power line carrier channel. The sole use of power line carrier communication is difficult to support the topology identification of the park's multi-energy power supply network.

[0006] Second, the large-scale access of distributed photovoltaics, energy storage, and charging piles to the campus power supply system will introduce a high proportion of power electronic devices, which will generate strong electromagnetic interference (EMI) noise to the campus power supply network, causing the accuracy of power data collection at branches and nodes of the campus power supply network to decrease, resulting in reduced reliability of the topology identification results of the campus power supply network.

[0007] Third, the park's multi-energy equipment is characterized by large number and frequent switching, which causes the topology of the park's power supply network to change rapidly. Therefore, the park has high requirements for the real-time performance of the identification results. In addition, the access of multi-energy equipment leads to an increase in the noise power of the power supply network and a decrease in the reliability of data communication. The traditional power supply network topology identification performance optimization method only targets a single communication indicator, resulting in the inability to take into account both the transmission delay and reliability of the topology identification data packet.

[0008] In summary, the current topology identification technology of multi-energy power supply network in the park has problems such as difficulty in transmitting power line carrier topology identification signals, reduced accuracy of power data collection of power supply network branches and nodes, and difficulty in simultaneously optimizing the data packet transmission delay and packet loss rate of topology identification for a single indicator. Summary of the invention

[0009] In order to solve the above technical problems, the purpose of the present invention is to provide a topology identification method for adapting to a campus multi-energy power supply network.

[0010] A topology identification method for adapting a multi-energy power supply network in a park of the present invention comprises the following specific steps:

[0011] A multi-time scale system model for topology identification of campus power supply network is constructed based on the topology identification method combined with power line industrial frequency carrier.

[0012] Refining the multi-time scale system model for topology identification of the campus power supply network into a delay model and a packet loss rate model;

[0013] The topology identification method based on the power line power frequency carrier and the above delay model and packet loss rate model describes the optimization problem of data packet transmission delay and packet loss rate of topology identification as minimizing the weighted and average value of topology identification delay and data packet loss rate under the specified number of periods;

[0014] A joint optimization method for data packet transmission delay and packet loss rate of topology identification based on confidence interval algorithm on reinforcement learning is proposed. The optimal solution of the optimization problem is optimized under the conditions of satisfying the number constraints of campus topology identification equipment and coding scheme selection constraints.

[0015] Furthermore, the power line power frequency carrier combined topology identification method is a joint optimization of topology identification at a small time scale and topology identification at a large time scale;

[0016] Among them, the large time scale refers to a period, within which the topology identification of the multi-energy power supply network in the park is completed once; the small time scale refers to a round, within which the list update of the equipment identified in this round is completed; among them, there are K(e) rounds in the e-th period, in the k-th round, the k-th layer equipment is identified, and the equipment list is updated after the identification is completed, and the k-th round ends; the value of K(e) is determined by the current coding strategy and the actual communication situation of the park power supply network. When the topology identification device identifies the connection relationship of all equipment in the park, the e-th period ends; after a period ends, the preference value of the coding strategy of the current period is calculated by analyzing the delay and packet loss rate of the topology identification process in the current period, and the preference value list is updated, and then the coding strategy with the highest preference value is selected in the next period.

[0017] Furthermore, the process of topology identification is as follows:

[0018] S1.1: The central node broadcasts a power line power frequency topology identification data packet to all slave nodes; the slave node receiving the data packet returns a power line carrier request data packet containing the device ID; the central node estimates the SINR based on the received request data packet and records the slave nodes that meet the data transmission conditions in the first layer; after the central node broadcasts the topology identification data packet, it waits for a specified time, then updates the list of the first layer slave nodes, and establishes a communication connection between the central node of the campus multi-energy power supply network and the first layer slave nodes;

[0019] S1.2: When the kth layer (k>1) slave node is identified, the central node uses the k-1th layer slave node as a relay and sends a power line power frequency topology identification data packet; the slave node that receives the topology identification data packet sends a power line carrier request data packet to the k-1th layer slave node it monitors, and the k-1th layer slave node analyzes whether the SINR of the request data packet it receives meets the data transmission conditions. If so, it sends a power line carrier identification data packet containing the k-1th layer slave node and the corresponding slave node device ID that sends the request data packet to the central node;

[0020] S1.3: When the central node waits for a specified time after broadcasting the k-th layer power line power frequency topology identification data packet, the k-th layer slave node topology identification is completed; then the central node updates the slave node list and establishes a communication connection between the k-th layer slave node and the k-1-th layer slave node;

[0021] S1.4: Repeat S1.2 and S1.3 until all slave nodes have established communication connections with the central node through the upper-level slave nodes, and the topology identification equipment list is established; the topology map generation module obtains the equipment list, draws the planar topology image of the campus power supply network, reports to the scheduling master station through the external data network and stores it locally.

[0022] Furthermore, the multi-time scale system model for topology identification of the park power supply network is:

[0023] The central node performs the topology identification scenario of the multi-energy power supply network in the park. There are M+1 nodes in total, consisting of 1 central node and M slave nodes. The set is represented as d0 represents the central node, and the connection method between the slave nodes is unknown;

[0024] The joint optimization problem of packet transmission delay and reliability for topology identification in large time scales, E periods, is expressed as The topology recognition device completes a topology recognition in each period; in a small time scale, each period contains K(e) rounds, and the set is expressed as The value of K(e) is related to the current coding strategy and the actual communication situation of the power supply network. The central node updates the slave node list in each round and completes the identification of a new layer of slave nodes. k (e) represents the number of slave nodes identified by the topology identification device in the kth round in the eth period, K(e) and n k The relationship (e) is as follows:

[0025]

[0026] In the process of identifying a new layer of slave nodes in each round, I encoding schemes are uniformly set for the slave nodes, and the set is expressed as where c i Indicates the i-th coding scheme. In each round, the unidentified slave nodes select a unified coding scheme to send request packets to the upper slave nodes; set the coding scheme selection variable a i,k (e), if a i,k (e) = 1 means that when the central node completes the update of the kth round of slave node list in the eth period, it commands the unidentified slave node to use the i-th encoding scheme to send a request data packet to the upper-level slave node. i,k (e) = 0 means no selection; different encoding schemes are selected in different rounds to form the encoding strategy of the central node in a topology identification process. The set can be expressed as where s l Represents the lth encoding strategy.

[0027] Furthermore, the delay model refers to:

[0028] The central node broadcasts the topology identification data packet and waits for T m After a certain period of time, the receiving of identification packets is stopped and the identification of the slave nodes of a new layer is performed. The transmission delay of the uplink request packet and the identification packet in a period of time is the number of rounds of topology identification and the waiting time T of the central node. m The product of is expressed as:

[0029] τ(e)=T m K(e).

[0030] Furthermore, the packet loss rate model refers to:

[0031] use It represents the average packet loss rate of the data packets when the slave node selects the i-th coding scheme in the k-th round of the e-th period. It is initialized to 0 and updated at the end of each round. The update formula is as follows:

[0032]

[0033] Among them, s k (e) represents the number of unidentified nodes that listen to the topology identification data packet in the kth round of the eth period; P i,k (e) represents the packet loss rate when the slave node selects the i-th coding scheme in the k-th round of the e-th period, as shown below:

[0034]

[0035] in The central node selects the i-th encoding scheme in the k-th round of the e-th period to establish a slave node d m With slave node d n The packet loss indicator variable during the communication connection is as follows:

[0036]

[0037] in Used to indicate the slave node d m In the kth round of the eth period, the i-th encoding scheme is selected to send to the slave node d n Whether packet loss occurs when sending request packets, Indicates that the request data packet is lost; Used to indicate the slave node d n Whether packet loss occurs when the i-th encoding scheme is selected in the k-th round of the e-th period to send the identification data packet to the central node, Indicates that the identification data packet is lost; when one of the request data packet and the identification data packet is lost, it will cause the failure of establishing the communication path, and it is considered that packet loss occurs during the process of establishing the communication connection.

[0038] Furthermore, the optimization problem is:

[0039] P1:

[0040] C1:

[0041] C2:

[0042] C3: Among them, λ is a weighted coefficient used to balance the impact of the packet loss rate of topology identification on the optimization problem. When λ increases, the reliability optimization effect of topology identification is enhanced and the delay optimization effect is weakened. C1 is the constraint on the number of slave nodes in the topology identification of the campus multi-energy power supply network, which means that at the end of topology identification within one period, the communication connection relationship of all M energy slave node terminals in the current campus power supply network must be identified. C2 and C3 are coding scheme selection constraints, where C3 means that within one round, the slave node can only select one coding scheme to transmit the request and identification data packet.

[0043] Furthermore, the specific process of the joint optimization method of data packet transmission delay and packet loss rate based on topology identification of confidence interval algorithm on reinforcement learning is as follows:

[0044] First, the central node selects a comparative coding scheme to determine the maximum identification round. The data packet transmission delay and packet loss rate using the comparative coding scheme are greater than those using For any data packet with any coding scheme in the central node, the central node sorts out the preference value list of the coding strategy based on the maximum identification round and initializes the preference value θ of each coding scheme. l (0);

[0045] Secondly, the central node traverses and selects different encoding strategies when performing the first L rounds of identification operations;

[0046] Again, the central node calculates the preference value θ of the current encoding strategy l (e) and average preference value

[0047]

[0048]

[0049] Among them, θ l (e) represents the preference value calculated by the central node when selecting the lth encoding strategy for topology identification in the eth period. represents the average preference value update of the central node for the lth encoding strategy after the eth topology identification is completed, y l (e) represents the total number of times the central node selects the lth encoding strategy after performing e topology identifications;

[0050] Then, after the traversal phase is over, in each period, the topology identification device selects the coding strategy with the largest preference estimation value among all coding strategies for topology identification. The preference estimation value of the coding strategy is calculated as follows:

[0051]

[0052] Among them, θ l (e) represents the preference value of the central node for the lth encoding strategy at the beginning of the eth topology identification;

[0053] Finally, repeat steps 2.3 and 2.4. After iterating to the set E periods, the central node selects the current average preference value The largest encoding strategy is used as the optimal encoding strategy for current campus topology identification.

[0054] Furthermore, the topology identification device used in the topology identification process includes a topology identification device and a node auxiliary device, wherein the topology identification device corresponds to the central node in the topology identification process, and the node auxiliary device corresponds to the slave node in the topology identification process;

[0055] Among them, the topology identification device, as a central node, uses power line industrial frequency communication to broadcast topology identification data packets, establish a list of equipment for topology identification of the power supply network of the equipment park, update the list of identified equipment within a round, formulate an adaptive coding strategy, and generate a topology plane image of the power supply network of the park;

[0056] The node auxiliary device, as a slave node, identifies the topology identification instruction issued by the topology identification device and performs corresponding operations, analyzes the node auxiliary device request data packet SINR, and uses power line carrier communication to upload the request data packet and the identification data packet to establish a communication connection with the topology identification device.

[0057] Furthermore, the topology identification device includes a central processing unit, a data storage module, a topology map generation module, and a power line communication module;

[0058] The central processor is responsible for receiving and issuing campus topology identification instructions, analyzing the SINR of node auxiliary device request packets, establishing a topology identification device list, and formulating an adaptive coding strategy. The device list is used to mark the level to which each device belongs and the device ID of the upper layer device that establishes a communication connection with the device.

[0059] The topology generation module has a simple graphics processing function, obtains the equipment list generated by the central processor, and forms a topological plane image of the power supply network in the park;

[0060] The data storage module is responsible for storing the list of equipment for topology identification of the campus power supply network or the local storage of the topology plane image;

[0061] The power line communication module has the communication capability of receiving and sending power line carrier signals and power line industrial frequency signals. The power line communication module of the topology identification device broadcasts the topology identification data packet to each node auxiliary device in the power supply network in the form of power line industrial frequency. The node auxiliary device that receives the topology identification data packet sends a power line carrier request data packet or an identification data packet to the topology identification device.

[0062] Further, the node auxiliary device includes a microprocessor and a power line communication module;

[0063] Among them, the microprocessor is responsible for identifying the topology identification instructions issued by the topology identification device and performing corresponding operations, analyzing the SINR of the node auxiliary device request data packet;

[0064] The power line communication module of the node auxiliary device is responsible for receiving the power line industrial frequency topology identification data packet sent by the topology identification device on the power line and the power line carrier request data packet sent by the unidentified node auxiliary device, and sending the power line carrier identification data packet to the topology identification device.

[0065] A topology identification device adapted to a multi-energy power supply network in a park, comprising:

[0066] Modeling module: used to construct a multi-time scale system model for topology identification of the campus power supply network based on the topology identification method combined with the power line frequency carrier;

[0067] Refinement module: used to refine the multi-time scale system model of the park power supply network topology identification into a delay model and a packet loss rate model;

[0068] Conversion module: used for describing the optimization problem of data packet transmission delay and packet loss rate of topology identification based on the topology identification method combined with the power line power frequency carrier and the above delay model and packet loss rate model as minimizing the weighted and average value of topology identification delay and data packet loss rate under the specified number of periods;

[0069] Optimization module: It is used to propose a joint optimization method for data packet transmission delay and packet loss rate of topology identification based on the confidence interval algorithm on reinforcement learning, and optimize the optimal solution of the optimization problem under the conditions of satisfying the number constraints of campus topology identification equipment and the selection constraints of coding schemes.

[0070] Furthermore, the topology identification method combining power line power frequency carrier in the modeling module is a joint optimization of topology identification at a small time scale and topology identification at a large time scale;

[0071] Among them, the large time scale refers to a period, within which the topology identification of the multi-energy power supply network in the park is completed once; the small time scale refers to a round, within which the list update of the equipment identified in this round is completed; among them, there are rounds in the th period, in the th round, the th layer of equipment is identified, and the equipment list is updated after the identification is completed, and the th round ends; the value of is determined by the current coding strategy and the actual communication situation of the park power supply network. When the topology identification device identifies the connection relationship of all equipment in the park, the th period ends; after one period ends, the preference value of the coding strategy of the current period is calculated by analyzing the delay and packet loss rate of the topology identification process in the current period, and the preference value list is updated, and then the coding strategy with the highest preference value is selected in the next period.

[0072] By means of the above scheme, the present invention has at least the following advantages:

[0073] (1) The present invention utilizes the complementary characteristics of power line carrier and power line industrial frequency, uses the power line industrial frequency to broadcast topology identification data packets, and uses the power line carrier to transmit request data packets and identification data packets on the power line industrial frequency communication path, which can solve the problem that a single power line carrier communication in a campus multi-energy power supply network is difficult to support the topology identification process.

[0074] (2) The present invention establishes a communication path by analyzing the SINR of the power line carrier request data packet, and identifies the topological structure of the campus multi-energy power supply network from the perspective of data transmission. There is no need to collect power data on the power supply network nodes or branches, which can avoid the problem of large errors in the identification results due to low accuracy of power data collection.

[0075] (3) The present invention proposes a joint optimization method for data packet transmission delay and packet loss rate of topology identification based on a confidence interval algorithm on reinforcement learning. By formulating an adaptive coding strategy, a weighted sum problem of minimizing the topology identification delay and data packet loss rate under a specified number of periods is constructed. The topology identification preference value list is updated on a large time scale, and the equipment list is updated on a small time scale. This solves the problem that traditional topology identification optimization methods target a single indicator and therefore cannot take into account the joint optimization problem of both data packet transmission delay and packet loss rate at the same time.

[0076] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show a certain embodiment of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0078] Figure 1 It is a schematic diagram of the topology identification architecture of the multi-energy power supply network in the park of the present invention;

[0079] Figure 2 It is a schematic diagram of the structure of the topology identification device of the present invention;

[0080] Figure 3 It is a schematic diagram of the structure of the node auxiliary device of the present invention;

[0081] Figure 4 It is the principle diagram of the method of the present invention;

[0082] Figure 5 It is a topology identification flow chart of the present invention;

[0083] Figure 6 It is a flow chart of joint optimization of data packet transmission delay and packet loss rate in topology identification of the present invention. DETAILED DESCRIPTION

[0084] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0085] 1. A topology identification device adapted to a multi-energy power supply network in a park

[0086] like Figure 1 As shown, the topology identification device is connected to the dispatching main station through an external data network, and a node auxiliary device is deployed at the input or output port of each energy equipment terminal in the park. The node auxiliary device shares the device ID with its corresponding energy equipment.

[0087] The topology identification device includes a central processing unit, a data storage module, a topology map generation module, and a power line communication module. Figure 2As shown. The central processing unit is responsible for receiving and issuing campus topology identification instructions, analyzing the SINR of node auxiliary device request data packets, establishing a topology identification equipment list, and formulating an adaptive coding strategy, wherein the equipment list is used to mark the level to which each device belongs and the device ID of the upper layer device that establishes a communication connection with the device. The topology map generation module has a simple graphics processing function, obtains the equipment list generated by the central processing unit, and forms a topology plane image of the power supply network in the campus. The data storage module is responsible for storing the park power supply network topology identification equipment list or the local storage of the topology plane image. The power line communication module has the communication capability of sending and receiving power line carrier signals and power line industrial frequency signals at the same time. The power line communication module of the topology identification device broadcasts the topology identification data packet to each node auxiliary device in the power supply network in the form of power line industrial frequency. The node auxiliary device that receives the topology identification data packet sends a power line carrier request data packet or an identification data packet to the topology identification device.

[0088] The node auxiliary device includes a microprocessor and a power line communication module, such as Figure 3 As shown. The microprocessor is responsible for identifying the topology identification instructions issued by the topology identification device and performing corresponding operations and analyzing the SINR of the node auxiliary device request data packet. The power line communication module of the node auxiliary device is responsible for receiving the power line power frequency topology identification data packet issued by the topology identification device on the power line and the power line carrier request data packet issued by the unidentified node auxiliary device, and sending the power line carrier identification data packet to the topology identification device.

[0089] 2. A topology identification method for multi-energy power supply networks in a park

[0090] Based on the above device, the present invention designs a topology identification method adapted to the multi-energy power supply network of the park, performs topology identification on a small time scale, and jointly optimizes the data packet transmission delay and packet loss rate of topology identification on a large time scale. The specific method principle is as follows Figure 4 shown.

[0091] Among them, the large time scale refers to a period, in which the topology identification device completes the topology identification of the multi-energy power supply network in the park; the small time scale refers to a round, in which the topology identification device completes the list update for the equipment identified in this round. Among them, there are K(e) rounds in the e-th period, in the k-th round, the k-th layer equipment is identified, and the equipment list is updated after the identification is completed, and the k-th round ends. The value of K(e) is determined by the current coding strategy and the actual communication situation of the park power supply network. When the topology identification device identifies the connection relationship of all equipment in the park, the e-th period ends. After a period ends, the topology identification device calculates the preference value of the current period coding strategy by analyzing the delay and packet loss rate of the topology identification process in the current period, and updates the preference value list, and then selects the coding strategy with the highest preference value in the next period.

[0092] (1) Topology Identification Process

[0093] By taking advantage of the complementary advantages of power line carrier and power line frequency, the topology identification of the multi-energy power supply network in the park is carried out. By taking advantage of the strong anti-interference ability of power line frequency communication, the power line frequency topology identification data packet is broadcasted, and the communication connection is established by calculating the SINR of the power line carrier request data packet. This can solve the problem of difficulty in transmitting topology identification data packets caused by poor quality of power line carrier communication channels. The specific process of topology identification is as follows: Figure 5 shown.

[0094] Step 1.1: The topology identification device broadcasts the power line power frequency topology identification data packet to all node auxiliary devices. The node auxiliary device that receives the data packet returns the power line carrier request data packet containing the device ID to the topology identification device. The topology identification device estimates the SINR based on the received request data packet and records the devices that meet the data transmission conditions as layer 1 devices. After broadcasting the topology identification data packet, the topology identification device waits for a specified time, updates the layer 1 device list, and establishes a communication connection between the topology identification device and the layer 1 devices.

[0095] Step 1.2: When the kth (k>1) layer device is identified, the topology identification device uses the k-1th layer node auxiliary device as a relay to send a power line industrial frequency topology identification data packet; the node auxiliary device that receives the topology identification data packet sends a power line carrier request data packet to the k-1th layer node auxiliary device it monitors, and the k-1th layer node auxiliary device analyzes whether the SINR of the request data packet it receives meets the data transmission conditions. If so, it sends a power line carrier identification data packet containing the k-1th layer node auxiliary device and the device ID of the node auxiliary device that sends the request data packet to the topology identification device.

[0096] Step 1.3: When the topology identification device waits for a specified time after broadcasting the k-th layer power line power frequency topology identification data packet, the topology identification of the k-th layer device is completed. Then the topology identification device updates the device list and establishes a communication connection between the k-th layer device and the k-1-th layer device.

[0097] Step 1.4: Repeat steps 1.2 and 1.3 until all devices have established communication connections with the topology identification device through the upper-layer device, and the topology identification device list is established. The topology map generation module obtains the device list, draws the topology image of the campus power supply network, reports it to the dispatching master station through the external data network, and stores it locally.

[0098] (2) Joint optimization of packet transmission delay and packet loss rate based on topology identification

[0099] Furthermore, the coding scheme adopted by the node auxiliary device when sending the request data packet and the identification data packet affects both the delay and packet loss rate of the data packet transmission. The use of a highly complex coding scheme can improve the signal-to-interference-noise ratio, increase the reliability of data communication, and reduce the packet loss rate of data packet transmission; but on the other hand, the complex coding scheme increases the redundancy of data packet information, increases the data packet transmission delay, and reduces the number of identification data packets that can be received by the topology identification device in one round, resulting in an increase in topology identification rounds. Therefore, the present invention proposes a method for jointly optimizing the data packet transmission delay and packet loss rate of topology identification data communication based on reinforcement learning. By formulating an adaptive coding strategy, the data packet transmission delay and packet loss rate of topology identification are jointly optimized to achieve high-reliability and low-latency topology identification of the multi-energy power supply network in the park. The specific implementation plan is as follows.

[0100] 1) System Model

[0101] Considering the topology identification scenario of the multi-energy power supply network in the park, there are M+1 devices, consisting of 1 topology identification device and M node auxiliary devices. The set is expressed as d0 represents a topology identification device, and the connection mode between various devices is unknown.

[0102] The present invention adopts a discrete time slot model to describe the multi-scale topology identification process and the joint optimization of the packet transmission delay and packet loss rate of topology identification. On a large time scale, considering the joint optimization problem of the packet transmission delay and packet loss rate of topology identification in E periods, the set is expressed as The topology recognition device completes a topology recognition in each period; in a small time scale, each period contains K(e) rounds, and the set is expressed as The value of K(e) is related to the current coding strategy and the actual communication situation of the power supply network. The topology identification device updates the equipment list in each round and completes the identification of the new layer of equipment. k(e) represents the number of devices identified by the topology identification device in the kth round of the eth period, then K(e) is related to n k The relationship (e) is as follows:

[0103]

[0104] The present invention uniformly sets I encoding schemes for the node auxiliary device in each round of new layer device identification process, and the set is expressed as where c i represents the i-th coding scheme. In each round, the unidentified node auxiliary device selects a unified coding scheme to send a request data packet to the upper node auxiliary device; set the coding scheme selection variable a i,k (e), if a i,k (e) = 1 means that the topology identification device instructs the unidentified node auxiliary device to use the i-th encoding scheme to send a request data packet to the upper node auxiliary device when completing the update of the k-th round of equipment list in the e-th period. i,k (e) = 0 means no selection; different encoding schemes are selected in different rounds to form the encoding strategy of the topology identification device in a topology identification process. The set can be expressed as where s l Represents the lth encoding strategy.

[0105] 2) Delay Model

[0106] The total delay of data packet transmission in a round can be divided into two parts: the transmission delay of the downlink topology identification data packet and the sum of the transmission delay of the uplink request data packet and the identification data packet. Since the downlink topology identification data packet is sent in the form of broadcast, its transmission delay in each round is much smaller than the total transmission delay of the request data packet and the identification data packet. Therefore, the present invention ignores the influence of the downlink data packet transmission delay and only considers the sum of the transmission delay of the uplink request data packet and the identification data packet.

[0107] After the topology identification device broadcasts the topology identification data packet, it waits for T m After a certain period of time, the device stops receiving identification packets and identifies the new layer of devices. The delay between the transmission of uplink request packets and identification packets in a period of time is defined as the number of rounds of topology identification and the waiting time T of the topology identification device. m The product of is expressed as:

[0108] τ(e)=T m K(e) (2)

[0109] 3) Packet loss rate model

[0110] In terms of topology identification packet loss rate model construction, It represents the average packet loss rate of the data packets when the node auxiliary device selects the i-th coding scheme in the k-th round of the e-th period. It is initialized to 0 and updated at the end of each round. The update formula is as follows:

[0111]

[0112] Among them, s k (e) represents the number of unidentified nodes that listen to the topology identification data packet in the kth round of the eth period; P i,k (e) represents the packet loss rate when the node auxiliary device selects the i-th coding scheme in the k-th round of the e-th period, as shown below,

[0113]

[0114] in The topology identification device selects the i-th encoding scheme in the k-th round of the e-th period to establish device d m With device n The packet loss indicator variable during the communication connection is as follows:

[0115]

[0116] in Used to indicate node auxiliary device d m In the kth round of the eth period, the i-th encoding scheme is selected to send to the node auxiliary device d n Whether packet loss occurs when sending request packets, Indicates that the request data packet is lost; Used to indicate node auxiliary device d n Whether packet loss occurs when selecting the i-th encoding scheme in the k-th round of the e-th period to send the identification data packet to the topology identification device, Indicates that the identification data packet is lost; when one of the request data packet and the identification data packet is lost, it will cause the failure of establishing the communication path, and it is considered that packet loss occurs during the process of establishing the communication connection.

[0117] 4) Problem Construction

[0118] Based on the above topology identification method and system model, the present invention constructs the optimization problem as minimizing the weighted sum average of the topology identification delay and the packet loss rate under a specified number of periods, as follows:

[0119] P1:

[0120] C1:

[0121] C2:

[0122] C3:

[0123] Among them, λ is a weighted coefficient used to balance the impact of the packet loss rate of topology identification on the optimization problem. When λ increases, the reliability optimization effect of topology identification is enhanced, and the delay optimization effect is weakened. C1 is the device quantity constraint for topology identification of the multi-energy power supply network in the park, indicating that at the end of topology identification within one period, the communication connection relationship of all M energy equipment terminals in the current park power supply network must be identified. C2 and C3 are coding scheme selection constraints, among which C3 means that within one round, the node auxiliary device can only select one coding scheme to transmit request and identification data packets.

[0124] 5) Optimization steps

[0125] The specific process of joint optimization of packet transmission delay and packet loss rate based on topology identification is as follows: Figure 6 As shown,

[0126] The above is only a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that in step 2.1, the topology identification device first selects a comparison coding scheme to determine the maximum identification round, and the data packet transmission delay and packet loss rate using the comparison coding scheme are greater than those using For any data packet with any coding scheme in the topology identification device, the topology identification device sorts out the preference value list of the coding strategy based on the maximum identification round, and initializes the preference value θ of each coding scheme l (0);

[0127] Step 2.2: The topology identification device traverses and selects different encoding strategies when performing the first L rounds of identification operations;

[0128] Step 2.3: The topology identification device calculates the preference value θ of the current encoding strategy l (e) and average preference value

[0129]

[0130]

[0131] Among them, θ l (e) represents the preference value calculated by the topology identification device when selecting the lth encoding strategy for topology identification in the eth period, represents the average preference value update of the topology identification device for the lth encoding strategy after the eth topology identification is completed, y l (e) represents the total number of times the topology identification device selects the lth encoding strategy after performing e topology identifications.

[0132] Step 2.4: After the traversal phase is completed, in each period, the topology identification device selects the coding strategy with the largest preference estimation value from all coding strategies for topology identification. The preference estimation value of the coding strategy is calculated as follows:

[0133]

[0134] Among them, θ l (e) represents the preference value estimate of the topology identification device for the lth encoding strategy at the beginning of the eth topology identification. The present invention uses the upper confidence interval algorithm in the reinforcement learning algorithm to achieve a compromise between empirical selection and exploratory selection, thereby avoiding the topology identification device from falling into a local optimal situation due to failure to explore all encoding strategies.

[0135] Step 2.5: Repeat steps 2.3 and 2.4, and after iterating to the set E periods, the topology identification device selects the current average preference value The largest encoding strategy is used as the optimal encoding strategy for current campus topology identification.

[0136] For those of ordinary skill in the art, without departing from the technical principles of the present invention, several improvements and variations can be made, and these improvements and variations should also be regarded as the protection scope of the present invention; those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The schemes in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0138] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0140] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0141] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A topology identification method for adapting a multi-energy power supply network in a park, characterized in that The specific steps are: A multi-time scale system model for topology identification of campus power supply network is constructed based on the topology identification method combined with power line industrial frequency carrier. Refining the multi-time scale system model for topology identification of the campus power supply network into a delay model and a packet loss rate model; The topology identification method based on the power line power frequency carrier and the above delay model and packet loss rate model describes the optimization problem of data packet transmission delay and packet loss rate of topology identification as minimizing the weighted and average value of topology identification delay and data packet loss rate under the specified number of periods; A joint optimization method for packet transmission delay and packet loss rate of topology identification based on the confidence interval algorithm on reinforcement learning is proposed. The optimal solution of the optimization problem is optimized under the conditions of satisfying the number constraints of campus topology identification equipment and the selection constraints of coding schemes. The multi-time scale system model for topology identification of the park power supply network is: The central node performs topology identification of the multi-energy power supply network in the park. There are M+1 nodes in total, consisting of 1 central node and M slave nodes. The set is represented as d0 represents the central node, and the connection method between the slave nodes is unknown; In the large time scale, the packet transmission delay and reliability joint optimization problem of topology identification in E periods is represented by ε={1,...,e,...,E}, and the topology identification device completes a topology identification in each period. In the small time scale, each period contains K(e) rounds, and the set is represented by The value of K(e) is related to the current coding strategy and the actual communication situation of the power supply network. The central node updates the slave node list in each round and completes the identification of a new layer of slave nodes. k (e) represents the number of slave nodes identified by the topology identification device in the kth round in the eth period, K(e) and n k The relationship (e) is as follows: In each round of identification of a new layer of slave nodes, a uniform encoding scheme is set for the slave nodes, and the set is represented as In each round, the unidentified slave nodes select a unified coding scheme to send request packets to the upper slave nodes; Set the encoding scheme to select variable a i,k (e), if a i,k (e) = 1 means that when the central node completes the update of the kth round of slave node list in the eth period, it commands the unidentified slave node to use the i-th encoding scheme to send a request data packet to the upper-level slave node. i,k (e) = 0 means no selection; different encoding schemes are selected in different rounds to form the encoding strategy of the central node in a topology identification process. The set can be expressed as where s l represents the lth encoding strategy; The packet loss rate model refers to: use It represents the average packet loss rate of the data packets when the slave node selects the i-th coding scheme in the k-th round of the e-th period. It is initialized to 0 and updated at the end of each round. The update formula is as follows: Among them, s k (e) represents the number of unidentified nodes that listen to the topology identification data packet in the kth round of the eth period; P i,k (e) represents the packet loss rate when the slave node selects the i-th coding scheme in the k-th round of the e-th period.

2. The method for topology identification of a multi-energy power supply network adapted to a campus according to claim 1, characterized in that: The power line power frequency carrier combined topology identification method is to perform topology identification on a small time scale and perform joint optimization of topology identification on a large time scale; Among them, the large time scale refers to a period, and within a period, the topology identification of the multi-energy power supply network in the park is completed once; the small time scale refers to a round, and within a round, the list update for the equipment identified in the round is completed; wherein, there are K(e) rounds in the e-th period, and in the k-th round, the k-th layer equipment is identified, and the equipment list is updated after the identification is completed, and the k-th round ends; the value of K(e) is determined by the current coding strategy and the actual communication situation of the park power supply network. When the topology identification device identifies the connection relationship of all equipment in the park, the e-th period ends; after a period ends, by analyzing the delay and packet loss rate of the topology identification process in the current period, the preference value of the coding strategy of the current period is calculated, and the preference value list is updated, and then the coding strategy with the highest preference value is selected in the next period; The process of topology identification is as follows: S1.1: The central node broadcasts a power line power frequency topology identification data packet to all slave nodes; the slave node receiving the data packet returns a power line carrier request data packet containing the device ID; the central node estimates the SINR based on the received request data packet and records the slave nodes that meet the data transmission conditions in the first layer; after the central node broadcasts the topology identification data packet, it waits for a specified time, then updates the list of the first layer slave nodes, and establishes a communication connection between the central node of the campus multi-energy power supply network and the first layer slave nodes; S1.2: When the kth layer (k>1) slave node is identified, the central node uses the k-1th layer slave node as a relay and sends a power line power frequency topology identification data packet; the slave node that receives the topology identification data packet sends a power line carrier request data packet to the k-1th layer slave node it monitors, and the k-1th layer slave node analyzes whether the SINR of the request data packet it receives meets the data transmission conditions. If so, it sends a power line carrier identification data packet containing the k-1th layer slave node and the corresponding slave node device ID that sends the request data packet to the central node; S1.3: When the central node waits for a specified time after broadcasting the k-th layer power line power frequency topology identification data packet, the k-th layer slave node topology identification is completed; then the central node updates the slave node list and establishes a communication connection between the k-th layer slave node and the k-1-th layer slave node; S1.4: Repeat S1.2 and S1.3 until all slave nodes have established communication connections with the central node through the upper-layer slave nodes, and the topology identification equipment list is established; the topology map generation module obtains the equipment list, draws the planar topology image of the campus power supply network, reports it to the dispatching master station through the external data network and stores it locally; The delay model refers to: The central node broadcasts the topology identification data packet and waits for T m After a certain period of time, the receiving of identification packets is stopped and the identification of the slave nodes of a new layer is performed. The transmission delay of the uplink request packet and the identification packet in a period of time is the number of rounds of topology identification and the waiting time T of the central node. m The product of is expressed as: τ(e)=T m K(e); in The central node selects the i-th encoding scheme in the k-th round of the e-th period to establish a slave node d m With slave node d n The packet loss indicator variable during the communication connection is as follows: in Used to indicate the slave node d m In the kth round of the eth period, the i-th encoding scheme is selected to send to the slave node d n Whether packet loss occurs when sending request packets, Indicates that the request data packet is lost; Used to indicate the slave node d n Whether packet loss occurs when the i-th encoding scheme is selected in the k-th round of the e-th period to send the identification data packet to the central node, Indicates that the identification data packet is lost; if one of the request data packet and the identification data packet is lost, it will cause the communication path to fail to be established, and it is considered that packet loss occurs during the communication connection establishment process; The optimization problem is: The λ weighting coefficient is used to balance the impact of the packet loss rate of topology identification on the optimization problem. When λ increases, the reliability optimization effect of topology identification is enhanced, and the delay optimization effect is weakened. C1 is the constraint on the number of slave nodes in the topology identification of the campus multi-energy power supply network, which means that at the end of the topology identification within one period, the communication connection relationship of all M energy slave node terminals in the current campus power supply network must be identified. C2 and C3 are coding scheme selection constraints, where C3 means that the slave node can only select one coding scheme to transmit request and identification data packets within one round.

3. The method for topology identification of a multi-energy power supply network adapted to a campus according to claim 1, characterized in that: The specific process of the joint optimization method of data packet transmission delay and packet loss rate based on topology identification of confidence interval algorithm on reinforcement learning is as follows: First, the central node selects a comparative coding scheme to determine the maximum identification round. The data packet transmission delay and packet loss rate using the comparative coding scheme are greater than those using For any data packet with any coding scheme in the central node, the central node sorts out the preference value list of the coding strategy based on the maximum identification round and initializes the preference value θ of each coding scheme. l (0); Secondly, the central node traverses and selects different encoding strategies when performing the first L rounds of identification operations; Then the central node calculates the preference value θ of the current encoding strategy l (e) and average preference value Among them, θ l (e) represents the preference value calculated by the central node when selecting the lth encoding strategy for topology identification in the eth period. represents the average preference value update of the central node for the lth encoding strategy after the eth topology identification is completed, y l (e) represents the total number of times the central node selects the lth encoding strategy after performing e topology identifications; Then, after the traversal phase is over, in each period, the topology identification device selects the coding strategy with the largest preference estimation value among all coding strategies for topology identification. The preference estimation value of the coding strategy is calculated as follows: Among them, θ l (e) represents the preference value of the central node for the lth encoding strategy at the beginning of the eth topology identification; Finally, repeat the above process and iterate to the set E period, and the central node selects the current average preference value The largest encoding strategy is used as the optimal encoding strategy for current campus topology identification.

4. The method for topology identification of a multi-energy power supply network adapted to a campus according to claim 1, characterized in that: The topology identification device used in the topology identification process includes a topology identification device and a node auxiliary device, wherein the topology identification device corresponds to the central node in the topology identification process, and the node auxiliary device corresponds to the slave node in the topology identification process; Among them, the topology identification device, as a central node, uses power line industrial frequency communication to broadcast topology identification data packets, establish a list of equipment for topology identification of the power supply network of the equipment park, update the list of identified equipment within a round, formulate an adaptive coding strategy, and generate a topology plane image of the power supply network of the park; The node auxiliary device, as a slave node, identifies the topology identification instruction issued by the topology identification device and performs corresponding operations, analyzes the node auxiliary device request data packet SINR, and uses power line carrier communication to upload the request data packet and the identification data packet to establish a communication connection with the topology identification device.

5. The method for topology identification of a multi-energy power supply network adapted to a campus according to claim 3, characterized in that: The topology identification device includes a central processing unit, a data storage module, a topology map generation module, and a power line communication module; The central processor is responsible for receiving and issuing campus topology identification instructions, analyzing the SINR of node auxiliary device request packets, establishing a topology identification device list, and formulating an adaptive coding strategy. The device list is used to mark the level to which each device belongs and the device ID of the upper layer device that establishes a communication connection with the device. The topology map generation module has a simple graphics processing function, obtains the equipment list generated by the central processor, and forms a topological plane image of the power supply network in the park; The data storage module is responsible for storing the list of equipment for topology identification of the campus power supply network or the local storage of the topology plane image; The power line communication module has the communication capability of receiving and sending power line carrier signals and power line industrial frequency signals. The power line communication module of the topology identification device broadcasts the topology identification data packet to each node auxiliary device in the power supply network in the form of power line industrial frequency. The node auxiliary device that receives the topology identification data packet sends a power line carrier request data packet or an identification data packet to the topology identification device.

6. The method for topology identification of a multi-energy power supply network adapted to a campus according to claim 3, characterized in that: The node auxiliary device includes a microprocessor and a power line communication module; Among them, the microprocessor is responsible for identifying the topology identification instructions issued by the topology identification device and performing corresponding operations, and analyzing the SINR of the node auxiliary device request data packet; The power line communication module of the node auxiliary device is responsible for receiving the power line industrial frequency topology identification data packet sent by the topology identification device on the power line and the power line carrier request data packet sent by the unidentified node auxiliary device, and sending the power line carrier identification data packet to the topology identification device.

7. A topology identification device adapted to a multi-energy power supply network in a park, characterized in that: include: Modeling module: used to construct a multi-time scale system model for topology identification of the campus power supply network based on the topology identification method combined with the power line frequency carrier; Refinement module: used to refine the multi-time scale system model of the park power supply network topology identification into a delay model and a packet loss rate model; Conversion module: used for describing the optimization problem of data packet transmission delay and packet loss rate of topology identification based on the topology identification method combined with the power line power frequency carrier and the above delay model and packet loss rate model as minimizing the weighted and average value of topology identification delay and data packet loss rate under the specified number of periods; Optimization module: It is used to propose a joint optimization method for data packet transmission delay and packet loss rate of topology identification based on the confidence interval algorithm on reinforcement learning, and optimize the optimal solution of the optimization problem under the conditions of satisfying the number constraints of campus topology identification equipment and the selection constraints of coding schemes.

8. The topology identification device for adapting a campus multi-energy power supply network according to claim 7, characterized in that: The topology identification method combined with the power line power frequency carrier in the modeling module is to perform topology identification at a small time scale and perform joint optimization of topology identification at a large time scale; Among them, the large time scale refers to a period, within which the topology identification of the multi-energy power supply network in the park is completed once; the small time scale refers to a round, within which the list update of the equipment identified in this round is completed; wherein, there are K(e) rounds in the e-th period, in the K(e)th round, the K(e)th layer equipment is identified, and the equipment list is updated after the identification is completed, and the e-th round ends; the value of K(e) is determined by the current coding strategy and the actual communication situation of the park power supply network. When the topology identification device identifies the connection relationship of all equipment in the park, the e-th period ends; after a period ends, the preference value of the coding strategy of the current period is calculated by analyzing the delay and packet loss rate of the topology identification process in the current period, and the preference value list is updated, and then the coding strategy with the highest preference value is selected in the next period.

Images