Communication power intelligent power distribution method based on power distribution control equipment

Abstract

The application relates to the technical field of power distribution control and discloses a communication power supply intelligent power distribution method based on a power distribution control device. The method comprises the following steps: acquiring the predicted standby power time length of each branch and the time length deviation of the predicted standby power time length relative to the preset promised time length in the main power interruption state; determining surplus branches and deficit branches according to the time length deviation; generating distribution weights in combination with historical performance records; constructing an allocation transfer relationship; when the allocation adjustment trigger condition is met, verifying the link state of the branches participating in the adjustment and generating an allocation adjustment instruction; updating the residual allocation of the corresponding branch by executing the allocation adjustment instruction, and performing power-off control when the actual power consumption generated after the allocation is updated reaches the corresponding residual allocation, and updating the historical performance records and reconstructing the allocation transfer relationship according to the actual transfer parameter. The communication power supply intelligent power distribution method can realize dynamic allocation adjustment and closed-loop control of shared energy storage power.

Description

Technical Field

[0001] This invention relates to the field of power distribution control technology, and more specifically to a smart power distribution method for communication power supply based on power distribution control equipment. Background Technology

[0002] In power consumption scenarios such as communication base stations and industrial park microgrids, multiple branch lines typically share the same power distribution and energy storage backup system. When the main power supply fails and is interrupted, the system needs to rely on energy storage devices to provide continuous power.

[0003] Existing power distribution control methods often employ static priority-based sequential power outage strategies during power distribution scheduling, lacking resource coordination strategies based on actual power demand and service agreements. When some branches have redundant backup power duration while other branches face backup power shortages, the system cannot establish an effective quota transfer relationship between surplus and deficit states, resulting in the overall idle waste of backup power resources and the failure of some branches to fulfill their obligations. Furthermore, the system's historical resource fulfillment status is not incorporated into the allocation strategy.

[0004] After a power distribution command is issued, existing power distribution control systems typically execute power supply control according to pre-determined quotas, lacking continuous monitoring of changes in the available power of energy storage devices and the actual power consumption of each household branch. When the available power of energy storage devices decays at a faster rate, or when the actual power load of a household branch deviates from the predicted power load, the pre-determined quotas may no longer be suitable for the operating status of the energy storage devices and household branches, easily causing the actual power consumption of some household branches to exceed the allocated quota.

[0005] Therefore, the present invention provides a method for intelligent power distribution of communication power supply based on power distribution control equipment. Summary of the Invention

[0006] The purpose of this invention is to provide a smart power distribution method for communication power supply based on power distribution control equipment, so as to solve the problems in existing power distribution control methods, such as the lack of dynamic adjustment of backup power quota allocation, the lack of quota transfer relationship between surplus and deficit branches, and the lack of power outage control and performance record update based on actual power consumption after quota adjustment.

[0007] The objective of this invention can be achieved through the following technical solutions: A method for intelligent power distribution of communication power supply based on power distribution control equipment includes the following steps: The power load and energy reserve of each branch circuit under the main power interruption are obtained. The power load and energy reserve are analyzed for backup power to obtain the expected backup power duration of each branch circuit. The backup power deviation is analyzed for the expected backup power duration to obtain the duration deviation of each branch circuit. Based on the duration deviation, the branches are classified into surplus branches and deficit branches. Historical performance records are obtained and allocation weights are generated by combining them with the duration deviation. Based on the allocation weights, the quota transfer relationship between surplus branches and deficit branches is constructed. Monitor the residual capacity decay status of energy storage equipment and dynamically set quota adjustment trigger conditions based on the residual capacity decay status; when the quota adjustment trigger conditions are met, parse the order of transfer in the quota transfer relationship to match the surplus branch and deficit branch participating in the adjustment, and generate quota adjustment instructions based on the quota transfer relationship. Execute quota adjustment instructions to update the remaining quota and duration deviation of the corresponding household branch; trigger power outage control when the actual electricity consumption of the household branch reaches the remaining quota; obtain actual transfer parameters to update historical performance records, and iteratively reconstruct quota transfer relationships based on the updated duration deviation.

[0008] Furthermore, the process of performing the backup power deviation analysis is as follows: Obtain the preset commitment duration for each household branch, and extract the duration deviation for each household branch based on the difference between the expected backup power duration and the preset commitment duration. Then, split the duration deviation into positive duration deviation and negative duration deviation.

[0009] Furthermore, the process of obtaining the estimated backup power duration is as follows: Extract historical power load data for each household branch and input it along with current power load data into a pre-trained temporal convolutional network to output the predicted power sequence for each household branch within a preset prediction time period. The predicted power sequence of each household branch is used to allocate electricity to obtain the theoretical allocated electricity for each household branch. The estimated backup power duration for each household branch is calculated based on the predicted power sequence and theoretical power allocation for each branch.

[0010] Furthermore, the process of establishing the quota transfer relationship is as follows: The order of acceptance of the deficit branches is determined according to the allocation weight, and a source-sink traffic transmission topology including surplus branch nodes and deficit branch nodes is established. The quota transfer path and transfer quota parameters in the source-sink traffic transmission topology are analyzed to generate a quota transfer relationship that includes transfer quota parameters and the order of transfer.

[0011] Furthermore, the process of establishing the source-sink traffic transmission topology is as follows: Based on the descending order of the allocation weights, the branch nodes in the set of deficit branches are sorted in descending order, and the sorted sequence of branch nodes is determined as the order of acceptance. Establish logical source nodes and logical sink nodes, and connect the logical source nodes to each surplus branch node whose theoretically transferable electricity is greater than zero. Each surplus branch node is connected to each deficit branch node with a power supply greater than zero, and each deficit branch node is connected to a logical sink node to obtain the source-sink traffic transmission topology.

[0012] Furthermore, the process of obtaining the allocation weights is as follows: Construct a fuzzy membership function based on duration deviation to convert the duration deviation of each household branch into a backup power status membership degree; Based on the membership degree of the backup power status, each household branch is classified into surplus branches and deficit branches. The theoretical transferable power is obtained based on the positive duration deviation, and the power to be supplemented is obtained based on the negative duration deviation. A set of surplus branches and a set of deficit branches are established. Extract the historical performance records of each branch in the set of deficient branches, use the time decay method to obtain the historical performance completion degree of each deficient branch, and standardize the historical performance completion degree and the amount of electricity to be made up to obtain a standardized decision matrix. The information entropy weight method is used to integrate the historical degree of insufficient replenishment and the amount of electricity to be replenished in the standardized decision matrix to generate the allocation weight of each deficit branch.

[0013] Furthermore, the process of obtaining the aforementioned historical fulfillment level is as follows: Extract the historical performance records of each defaulting branch within a pre-configured historical time window; The historical performance record includes transfer quota parameters, actual transfer parameters, quota adjustment time, and performance results; The time interval between the quota adjustment time and the current time is divided into multiple time intervals, and a numerical time weight is pre-configured for each time interval; For historical performance records where the transfer quota parameter is greater than zero and the performance result is either normal completion or termination due to insufficient available power of the energy storage equipment, the ratio of the actual transfer parameter to the transfer quota parameter is calculated as the performance completion ratio of the corresponding historical performance record. The completion rate of each valid historical performance record is multiplied by its corresponding time weight, and the sum of the multiplication results is divided by the sum of the corresponding time weights to obtain the historical performance completion rate of the corresponding deficit branch.

[0014] Furthermore, the process of generating the quota adjustment instruction is as follows: Continuously monitor high-priority adjustment events and future timestamps. After the quota adjustment triggering conditions are met, perform link status handshake verification on the corresponding branch according to the order of transfer in the quota transfer relationship to determine the valid matching link. For valid matching links, the transfer quota parameters are encapsulated in a protocol to generate quota adjustment instructions for updating the corresponding quota data of surplus and deficit branches.

[0015] Furthermore, the process of setting the quota adjustment trigger condition is as follows: A real-time observation window for energy storage devices is constructed and the capacity time series is extracted. The capacity time series is then smoothed and first-order derivative is calculated to obtain the residual decay rate of the energy storage devices. Based on the residual decay rate, query the pre-configured trigger strategy table to determine the quota adjustment trigger method and monitoring period, and set the trigger method and monitoring period as the quota adjustment trigger conditions.

[0016] Furthermore, the process of iteratively reconstructing the quota transfer relationship is as follows: Monitor the actual electricity consumption of each household branch after the most recent quota update, and execute power outage control of the corresponding household branch based on the comparison result of the actual electricity consumption and the corresponding remaining quota. When the corresponding branch circuit is de-energized, the main power supply is restored, or the quota transfer relationship is reconstructed, the actual transfer parameters corresponding to the indebted branch circuit are calculated, and the historical performance record is updated based on the actual transfer parameters. After updating the historical performance record, the current duration deviation is re-entered into the process of constructing the quota transfer relationship between the surplus branch and the deficit branch, thereby realizing the re-establishment of the quota transfer relationship.

[0017] The beneficial effects of this invention are as follows: 1. After a main power outage, the current power load of each household branch and the current available power reserve of the energy storage device are obtained, and the expected backup power duration of each household branch is calculated based on the predicted power sequence. Based on the time deviation between the expected backup power duration and the preset commitment duration, the household branches are divided into surplus branches and deficit branches. The historical performance records of deficit branches are processed according to preset time intervals and time weights, and allocation weights are generated using the information entropy weight method in conjunction with the amount of power to be replenished. A source-sink flow transmission topology is established based on the theoretically transferable power and the amount of power to be replenished, and the quota transfer path and transfer quota parameters are determined using the network simplex algorithm, thereby realizing the logical power quota adjustment of shared energy storage devices.

[0018] 2. By continuously collecting the current available power reserve of energy storage devices, a capacity time series is established, and the reserve decay rate is obtained through smoothing filtering and first-order differentiation. Based on the preset interval to which the reserve decay rate belongs, the corresponding quota adjustment trigger mode and monitoring cycle are read, thereby adjusting the detection frequency of quota adjustment when the available power of energy storage devices changes. After the quota adjustment trigger condition is met, the link status handshake verification is performed on the branch nodes in the quota transfer relationship according to the transfer order, and quota adjustment instructions are generated only for branch nodes that return valid responses, thereby reducing the occurrence of sending quota adjustment instructions to disconnected branch nodes.

[0019] 3. After executing the quota adjustment instruction, continuously acquire the actual electricity consumption of each branch after the most recent quota update, and compare the actual electricity consumption with the corresponding remaining quota; when the actual electricity consumption reaches or exceeds the remaining quota, send a disconnection control signal to the switch actuator of the corresponding branch. When the quota replenishment event ends, record the performance result based on the comparison result of the actual transfer parameter and the transferred quota parameter, and write the transferred quota parameter, actual transfer parameter, quota adjustment time and performance result into the historical performance record; reconstruct the quota transfer relationship based on the updated duration deviation. Attached Figure Description

[0020] The invention will now be further described with reference to the accompanying drawings.

[0021] Figure 1 This is a flowchart of a smart power distribution method for communication power supply based on power distribution control equipment according to the present invention; Figure 2 This is a flowchart illustrating the establishment of quota transfer relationships in this invention; Figure 3 This is a diagram showing the source-sink flow transmission topology and quota transfer relationship between surplus and deficit branches in this invention. Detailed Implementation

[0022] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0023] Example 1: like Figure 1 As shown, a smart power distribution method for communication power supply based on power distribution control equipment includes the following steps: S10. Obtain the power load and energy reserve of each branch circuit under the main power interruption state, perform backup power analysis on the power load and energy reserve to obtain the expected backup power duration of each branch circuit; perform backup power deviation analysis on the expected backup power duration to obtain the duration deviation of each branch circuit. The process of obtaining the electrical load of each branch circuit and the remaining power capacity of the energy storage device under the condition of main power interruption is as follows: The main power supply voltage is collected by the incoming line voltage transformer, and the status of the main power supply circuit breaker is read. When the main power circuit breaker is in the open state, the main power supply is determined to be interrupted; when the main power circuit breaker is not in the open state, the main power supply voltage is compared with the voltage determination threshold. If the main power supply voltage is lower than or equal to the voltage judgment threshold and continues to reach the confirmation delay, the main power supply is determined to be interrupted. The voltage judgment threshold is determined based on the rated voltage and the preset drop ratio; For example, when the preset drop ratio is 30%, the voltage judgment threshold is set to 70% of the rated voltage, and the confirmation delay is set to 20ms; If the main power supply voltage is higher than the voltage judgment threshold, or if the main power supply voltage is lower than or equal to the voltage judgment threshold for a duration that does not reach the confirmation delay, maintain normal power supply status and continue to collect the main power supply voltage. When the main power supply is interrupted, the device enters the backup power distribution state and wakes up the battery management system (BMS) inside the energy storage device. Read the current available power balance output by the battery management system (BMS) and use the current available power balance as the available power data of the energy storage device for subsequent theoretical power allocation calculation; Start the power acquisition function of the smart energy meter at the output end of the branch circuit, and acquire the sampling power sequence of each branch circuit within the sampling time window determined by the number of sampling points in the pre-configured sampling period. The sampled power sequence is smoothed using a pre-configured exponential smoothing coefficient; The initial smoothing value is determined using the initial sampling power value, and subsequent smoothing values ​​are determined based on the current sampling power value and the smoothing value at the previous sampling time. The smoothed power values ​​within the sampling time window are averaged to obtain the current power load of each household branch. The process of analyzing the power load and power reserve to obtain the estimated backup power duration for each household branch is as follows: S101. Input the historical power load data and current power load data of each household branch into the pre-trained temporal convolutional network, and output the predicted power sequence of each household branch in the future preset prediction time period. Specifically, the process of obtaining the predicted power sequence is as follows: The sampling time and corresponding sampling power of each household branch are retrieved in chronological order within multiple pre-set historical power supply cycles to form the historical power load data of the corresponding household branch; Historical electricity load data are arranged according to the same sampling period, and the current electricity load is used as the latest power data at the current sampling time to form the input time series of the corresponding household branch; During pre-training, the historical input time series of each household branch is used as the training input of the temporal convolutional network, and the actual power sequence that belongs to the same household branch as the historical input time series and is continuously collected after the historical input time series is used as the training target. The length of the actual power sequence is the same as the number of prediction sampling points within the preset prediction time period; The input time series of each household branch is fed into a pre-trained temporal convolutional network, which outputs the predicted power series of the corresponding household branch within a preset prediction time period in the future; The input time series length, prediction sampling period, and prediction time period length are pre-configured by the power distribution control equipment. S102. Perform power allocation on the predicted power sequence of each household branch to obtain the theoretical power allocation for each household branch. Specifically, the power allocation method is as follows: Based on the prediction sampling period corresponding to the prediction power sequence, the prediction power sequence of each household branch is accumulated and converted into power and time to obtain the expected power consumption of each household branch in the future preset prediction time period. Calculate the proportion of the expected power consumption of each branch to the total expected power consumption of all branches, and allocate the current available power reserve of the energy storage equipment according to the proportion to obtain the theoretical power allocation of each branch. The theoretically allocated power is used to represent the logical power quota allocated by the shared energy storage device to the corresponding household branch, and does not represent the physical division of the power within the energy storage device; When the total expected power consumption of all household branches is zero, proportional allocation will not be performed, all household branches will be marked as no-load branches, and the power load data of each household branch will be reacquired in the next sampling period. S103. Calculate the expected backup power duration for each household branch based on the predicted power sequence and theoretical power allocation of each branch. Specifically, the method for obtaining the estimated backup power duration is as follows: When the temporal convolutional network outputs a valid predicted power sequence, the predicted power values ​​of the corresponding household branch within a preset prediction time period are averaged to obtain the predicted average power. When the predicted average power is greater than zero, the theoretical power allocation is divided by the predicted average power to obtain the expected backup power duration. Among them, the theoretical power allocation is in the unit of electricity, the predicted average power is in the unit of power, and the expected backup power duration is in the unit of time. When the predicted average power is less than or equal to zero, the ratio calculation is not performed, the corresponding household branch is marked as a branch with no effective load, and the branch is temporarily not included in the quota transfer calculation for surplus or deficit branches. When the temporal convolutional network fails to output a valid predicted power sequence, the current power load of the corresponding household branch is used to replace the predicted average power, and the expected backup power duration is calculated in the same way. When the current electrical load is less than or equal to zero, perform the same treatment as when the predicted average power is less than or equal to zero. The process of performing backup power deviation analysis on the expected backup power duration to obtain the duration deviation of each household branch is as follows: S111. Obtain the preset commitment duration corresponding to each household branch, and extract the duration deviation of each household branch based on the difference between the expected backup power duration and the preset commitment duration, and split the duration deviation into positive duration deviation and negative duration deviation. Specifically, the method for obtaining the duration deviation is as follows: Access the service agreement database for local maintenance of power distribution control equipment and extract the preset commitment duration for each branch circuit, calculated from the moment the main power supply is interrupted. Subtract the corresponding preset commitment duration from the estimated backup power duration of each household branch to obtain the duration deviation of each household branch. When the duration deviation is greater than or equal to zero, it is determined that the expected backup power duration of the corresponding branch circuit has reached the preset commitment duration, and the corresponding duration deviation is determined as a positive duration deviation. When the duration deviation is less than zero, it is determined that the expected backup power duration of the corresponding branch circuit has not reached the preset promised duration, and the corresponding duration deviation is determined as a negative duration deviation. A duration deviation dataset is created based on the node identifiers and corresponding duration deviations of the branch roads, and the duration deviation dataset is output to the branch road classification step.

[0024] S20. Classify branch banks by household based on time deviation to obtain surplus branch banks and deficit branch banks. Obtain historical performance records and generate allocation weights by combining them with time deviations. Construct quota transfer relationships between surplus branch banks and deficit branch banks based on allocation weights. The process of classifying branch banks based on time deviation to obtain surplus branches and deficit branches, and obtaining historical performance records and generating allocation weights by combining them with time deviation is as follows: S201. Construct a fuzzy membership function based on duration deviation to convert the duration deviation of each household branch into a backup power status membership degree. Specifically, the process of converting to a backup power state membership is as follows: Input the duration deviation of each household branch into the pre-set Sigmoid fuzzy membership function, and output the backup power status membership with a value range of 0 to 1; The Sigmoid fuzzy membership function takes the duration deviation as input, divides the duration deviation by the scale parameter and takes the opposite number, performs an exponential operation on the result, adds 1 to the exponential operation result and takes the reciprocal to obtain the membership degree of the backup power status. The scale parameter uses a time unit to determine the impact of the duration deviation on the rate of change of the membership degree of the standby power status, and the scale parameter is a preset time value that is greater than zero. When the duration deviation is zero, the membership degree of the backup power status is 0.5; The scale parameters are pre-configured by the power distribution control equipment and remain unchanged within the same backup power allocation cycle; S202. Classify each household branch into surplus branches and deficit branches according to the membership degree of the backup power status. Obtain the theoretical transferable power according to the positive duration deviation and the power to be supplemented according to the negative duration deviation. Establish a set of surplus branches and a set of deficit branches. Specifically, the method for classifying surplus branches and deficit branches is as follows: For branch lines of households that have obtained effective estimated backup power duration, they are classified according to the comparison results of backup power status membership degree and benchmark fuzzy median. The benchmark fuzzy median is fixed at 0.5, which is used to represent the classification boundary when the expected backup power duration is equal to the preset commitment duration; When the membership degree of the backup power status is greater than or equal to 0.5, the node identifier of the corresponding household branch is written into the surplus branch set, and the positive duration deviation is multiplied by the predicted average power to obtain the theoretical transferable power. When the membership degree of the backup power status is less than 0.5, the node identifier of the corresponding household branch is written into the set of the deficient branches, and the absolute value of the negative duration deviation is multiplied by the predicted average power to obtain the power to be supplemented. When the temporal convolutional network fails to output a valid predicted power sequence, the current electricity load is used to replace the predicted average power. Branches with no effective load are not included in the surplus branch set or the deficit branch set. It should be noted that the theoretically transferable electricity, the electricity to be replenished, the transfer quota parameter, and the remaining quota all use the same electricity unit; S203. Extract the historical performance records of each branch in the set of deficient branches, use the time decay method to obtain the historical performance completion degree of each deficient branch, and standardize the historical performance completion degree and the amount of electricity to be replenished to obtain the standardized decision matrix. Specifically, the standardized decision matrix is ​​obtained as follows: Call the log storage module of the power distribution control equipment to extract the historical performance records of each indebted branch within the pre-configured historical time window; The historical performance record shall include at least the transfer quota parameters, actual transfer parameters, quota adjustment time and performance results; The time interval between the quota adjustment time and the current time is divided into multiple non-overlapping and continuous time intervals, and a numerical time weight is pre-configured for each time interval. Based on the time interval to which each historical performance record belongs, read the time weight corresponding to that time interval; The time weights corresponding to each time interval decrease sequentially in order of increasing time interval, and remain unchanged during the same weight allocation calculation. For historical performance records where the transfer quota parameter is greater than zero and the performance result is either normal completion or termination due to insufficient available power of the energy storage equipment, the ratio of the actual transfer parameter to the transfer quota parameter is calculated as the performance completion ratio of that record. When the actual transfer parameter is greater than the transfer quota parameter, the fulfillment ratio of this record will be treated as 1. The completion rate of each valid historical performance record is multiplied by its corresponding time weight, and the sum of the multiplication results is divided by the sum of the corresponding time weights to obtain the historical performance completion rate of the corresponding deficit branch. The historical fulfillment level is a dimensionless value between 0 and 1; Subtracting the historical fulfillment rate from 1 yields the historical shortfall rate. For deficit branches that lack valid historical performance records, the degree of historical insufficiency will be directly set to 0. The degree of historical shortfall and the amount of electricity to be supplemented are used as two positive evaluation indicators to establish an original decision matrix; The two positive evaluation indicators were standardized by range processing to obtain the standardized decision matrix; It should be noted that when any positive evaluation index has the same value in all deficit branches, the corresponding positive evaluation index will be set to the equivalent standardized result to reduce the possibility of the range being zero, which would prevent the standardization process from being completed. S204. The information entropy weight method is used to integrate the historical degree of insufficient replenishment and the amount of electricity to be replenished in the standardized decision matrix to generate the allocation weight of each deficit branch. Specifically, the method for generating and allocating weights is as follows: The historical insufficiency and the amount of electricity to be supplemented were both processed using range standardization: For each indicator, process it according to the minimum and maximum values ​​of that indicator among all the deficit branches; When the maximum and minimum values ​​of this indicator are the same, the standardized result of all deficit branches on this indicator is set to 1. Divide the standardized result of each deficit branch in each indicator by the sum of the standardized results of that indicator in all deficit branches to obtain the indicator weight of each deficit branch in that indicator. When the weight of an indicator is 0, the contribution of that data item to the information entropy is set to 0, and no logarithmic operation is performed. Calculate the corresponding indicator information entropy based on the weight of each indicator, and subtract the indicator information entropy from 1 to obtain the information utility of the corresponding indicator. The information utility is normalized by summing the information utility of each indicator to obtain the objective weights corresponding to the degree of historical shortfall and the amount of electricity to be supplemented. The standardized results of each deficit branch on the two indicators are multiplied by their corresponding objective weights and then summed to obtain the comprehensive evaluation score of each deficit branch. Then divide the comprehensive evaluation score of each deficit branch by the sum of the comprehensive evaluation scores of all deficit branches to obtain the allocation weight, and the sum of all allocation weights is 1. If there is only one deficient branch, set the allocation weight of that branch to 1; When the sum of the information utility of the two indicators is 0 or the overall evaluation score is 0, all deficit branches are assigned the same weight. Among them, such as Figure 2 As shown, the process of constructing the quota transfer relationship between surplus branches and deficit branches based on the allocation weights is as follows: S211. Determine the order of acceptance of the deficit branches according to the allocation weight, and establish a source-sink traffic transmission topology that includes surplus branch nodes and deficit branch nodes. Specifically, the method for establishing a traffic transmission topology is as follows: Based on the descending order of the allocation weights, the branch nodes in the set of deficit branches are sorted in descending order, and the sorted sequence of branch nodes is determined as the order of acceptance. Establish logical source nodes and logical sink nodes. Connect the logical source nodes to each surplus branch node with theoretically transferable electricity greater than zero. Connect each surplus branch node to each deficit branch node with electricity to be supplemented greater than zero. Connect each deficit branch node to the logical sink node to obtain the source-sink traffic transmission topology. It should be noted that the source-sink flow transmission topology only represents the logical power quota allocation relationship under the same shared energy storage device, and does not mean that the power is directly transmitted from the physical power supply circuit of the surplus branch to the physical power supply circuit of the deficit branch. It should be noted that the source-sink flow transmission topology is used to represent the logical power quota allocation relationship within the shared energy storage device, and does not represent the direct transmission of power from the physical power supply circuit of the surplus branch to the physical power supply circuit of the deficit branch. S212. Analyze the quota transfer path and transfer quota parameters in the source-sink traffic transmission topology, and generate a quota transfer relationship that includes transfer quota parameters and the order of transfer. Specifically, the method for establishing the quota transfer relationship between surplus branches and deficit branches is as follows: Set the theoretical transferable power of each surplus branch as the upper limit of the capacity of the connection arc between the logical source node and the corresponding surplus branch node. Set the amount of electricity to be replenished for each deficit branch as the upper limit of the capacity of the connection arc between the corresponding deficit branch node and the logical sink node; The upper limit of the connection arc capacity between the surplus branch node and the deficit branch node is set to the smaller value between the theoretical transferable power of the corresponding surplus branch and the power to be supplemented of the corresponding deficit branch. The smaller value between the total theoretically transferable electricity and the total electricity to be supplemented is determined as the total transferred electricity in this quota adjustment. According to the order of transfer of the deficit branches, a sequence cost is set for the connection arc between each surplus branch node and the deficit branch node. The deficit branch with the earlier transfer order corresponds to a lower sequence cost. When the transfer order is the same, the order is determined according to the preset order of the branch node identifiers. The network simplex algorithm is run to obtain the quota transfer paths and the transfer volume, and the transfer volume corresponding to each quota transfer path is determined as the transfer quota parameter, with the conditions of satisfying the total transferred volume, each connecting arc not exceeding the corresponding capacity limit, and the total path cost being the lowest. When the total theoretically transferable electricity is less than the total electricity to be supplemented, the branch that has not obtained the transfer quota parameter is marked as a branch with insufficient quota; the branch with insufficient quota retains the remaining quota before adjustment, and power outage control is executed when the actual electricity consumption reaches the remaining quota before adjustment. The surplus branch node identifier, deficit branch node identifier, transfer quota parameters and transfer order are encapsulated into structured data to obtain the quota transfer relationship; It should be noted that, as Figure 3 As shown, after the main power supply is interrupted and the backup power distribution state is entered, the power distribution control equipment classifies each household branch according to the time deviation between the expected backup power duration and the corresponding preset promised duration. Figure 3 The logical source node 101 is used to represent the logical supply end that can be used for quota adjustment on the side of shared energy storage equipment; the surplus branch nodes 102a and 102b represent the surplus branches obtained after classification; the deficit branch nodes 103a and 103b represent the deficit branches obtained after classification; the logical sink node 104 is used to represent the logical demand aggregation end on the side of the deficit branch. For surplus branches, the distribution control equipment calculates the theoretically transferable power based on the corresponding positive duration deviation and predicted average power, and uses the theoretically transferable power as the upper limit of the capacity of the connection arc between the logical source node 101 and the corresponding surplus branch node; for deficit branches, the distribution control equipment calculates the power to be supplemented based on the absolute value of the negative duration deviation and predicted average power, and uses the power to be supplemented as the upper limit of the capacity of the connection arc between the corresponding deficit branch node and the logical sink node 104. by Figure 3Taking the set of quota allocation parameters shown as an example, the theoretical maximum transferable electricity for surplus branch node 102a is 18kWh, and the theoretical maximum transferable electricity for surplus branch node 102b is 10kWh; the maximum amount of electricity to be supplemented for deficit branch node 103a is 16kWh, and the maximum amount of electricity to be supplemented for deficit branch node 103b is 12kWh; the order of acceptance is determined according to the allocation weight of the deficit branches, wherein the order of acceptance for deficit branch node 103a is before that for deficit branch node 103b. After the quota adjustment triggering conditions are met and the status verification of the effective matching link is completed, the power distribution control equipment determines the quota transfer path and transfer quota parameters based on the source-sink flow transmission topology; such as Figure 3 As shown, surplus branch node 102a transfers 12 kWh and 6 kWh to deficit branch nodes 103a and 103b, respectively, and surplus branch node 102b transfers 4 kWh and 6 kWh to deficit branch nodes 103a and 103b, respectively. The total transfer quota is 28 kWh, which is equal to the total theoretical transferable electricity of surplus branches and the total electricity to be supplemented of deficit branches. The distribution control equipment encapsulates the surplus branch node identifier, deficit branch node identifier, transfer quota parameters, and transfer order in a structured manner to generate quota transfer relationships, and generates quota adjustment instructions accordingly. It should be noted that, Figure 3 The source-sink flow transmission topology in the diagram is used to represent the logical power quota allocation relationship within the shared energy storage device. The transfer quota parameter on the connection arc is used to represent the adjustment amount of the logical quota, but does not represent the direct transmission of power from the physical power supply circuit of the surplus branch to the physical power supply circuit of the deficit branch.

[0025] Example 2: Please see Figure 1 As shown, a smart power distribution method for communication power supply based on power distribution control equipment includes the following steps: S30. Monitor the residual capacity decay status of energy storage equipment and dynamically set quota adjustment trigger conditions based on the residual capacity decay status; when the quota adjustment trigger conditions are met, parse the order of transfer in the quota transfer relationship to match the surplus branch and deficit branch participating in the adjustment, and generate quota adjustment instructions based on the quota transfer relationship. The process of monitoring the residual capacity decay status of energy storage devices and dynamically setting quota triggering conditions based on the residual capacity decay status is as follows: S301. Construct a real-time observation window for the energy storage device and extract the capacity time series. Smooth and first-order derivative of the capacity time series to obtain the remaining capacity decay rate of the energy storage device. Specifically, the method for obtaining the margin decay rate is as follows: A real-time observation window for energy storage devices is established. The real-time observation window is a set of continuous sampling points consisting of multiple current available power balance sampling points collected continuously according to a pre-configured fixed sampling period, and a capacity time series is established according to the sampling time order. The capacity time series is fitted with a local polynomial using a pre-defined Savitzky-Gore filter algorithm to reduce the impact of measurement noise on the capacity time series. The first derivative of the fitted capacity time series is then calculated to obtain the rate of change of charge at each sampling time. The number of sampling points, sampling period, and polynomial order included in the real-time observation window are pre-configured by the power distribution control equipment. The number of sampling points contained in the real-time observation window is an odd number greater than the order of the polynomial; When the number of sampling points in the real-time observation window has not reached the configured number of sampling points, the current available power balance will continue to be collected, and the balance decay rate will not be calculated based on the incomplete real-time observation window. Once the number of sampling points reaches the configured value, the rate of change of electricity consumption at the latest sampling time is extracted. When the rate of change of power is less than zero, its absolute value is determined as the margin decay rate; when the rate of change of power is greater than or equal to zero, the margin decay rate is set to zero. The rate of decrease in charge is expressed as the change in charge per unit time. S302. Query the pre-configured trigger strategy table according to the margin decay rate, determine the quota adjustment trigger method and monitoring period, and set the trigger method and monitoring period as the quota adjustment trigger conditions. Specifically, the method for setting quota adjustment trigger conditions is as follows: Pre-store trigger strategy tables in power distribution control equipment; The triggering strategy table includes multiple non-overlapping and continuous margin decay rate intervals arranged from low to high according to the margin decay rate, as well as the triggering method and monitoring period corresponding to each margin decay rate interval. Except for the last interval, each interval adopts the boundary rule of the lower limit being included and the upper limit being excluded, and the last interval includes its upper limit; The multiple intervals cover all non-negative values ​​of the margin decay rate. Based on the interval to which the current margin decay rate belongs, the corresponding triggering method and monitoring period are read. When the triggering method is immediate adjustment, a high-priority adjustment event is generated immediately; when the triggering method is timed adjustment, the current time is added to the corresponding monitoring period to obtain a future timestamp, and the future timestamp is set as the triggering condition for timed quota adjustment. The monitoring cycle and triggering method for each section are pre-configured by the power distribution control equipment according to the operating parameters of the energy storage equipment, and remain unchanged during the same monitoring process; The process of parsing the order of transferees in the quota transfer relationship to match the surplus and deficit branches participating in the adjustment, and generating a quota adjustment instruction based on the quota transfer relationship when the quota adjustment trigger condition is met, is as follows: S303. Continuously monitor high-priority adjustment events and future timestamps. After the quota adjustment triggering conditions are met, perform link status handshake verification on the corresponding branch according to the order of transfer in the quota transfer relationship to determine the valid matching link. Specifically, the method for matching the surplus and deficit branches involved in the adjustment is as follows: The power distribution control equipment compares the current timestamp with the future timestamp and continuously monitors high-priority adjustment events. When a high-priority adjustment event is detected, or when the current timestamp reaches a future timestamp, the quota adjustment triggering condition is determined to be met, an execution permission signal is sent to the control center, and the quota transfer relationship constructed in S20 is retrieved. If no high-priority adjustment event is detected and the current timestamp has not reached the future timestamp, it is determined that the quota adjustment triggering condition is not met, the power distribution control equipment remains in a waiting state, and does not send an execution permission signal to the control center; After receiving the execution permission signal, the control center extracts the communication addresses of the corresponding surplus branch nodes and deficit branch nodes in sequence according to the order of transfer in the quota transfer relationship, and sends status query messages to the surplus branch nodes and deficit branch nodes respectively. When both the surplus branch node and the deficit branch node return a valid response message that conforms to the preset communication protocol and whose node identifier matches the request node identifier within the pre-configured response time limit, it is determined that both the corresponding surplus branch node and deficit branch node are in a communicable state and the link is valid, and the surplus branch node, deficit branch node and transfer quota parameter are determined as a valid matching link. If any branch node fails to return a valid response message within the response time limit, the branch node will be marked as a lost node and removed from the current quota transfer relationship and source-sink traffic transmission topology; Release the transfer quota parameters corresponding to the quota transfer relationship of the lost node, and do not send quota adjustment instructions to the lost node; For the surplus and deficit branch nodes after deleting the disconnected nodes, S211 and S212 are re-executed to regenerate the quota transfer relationship with the theoretical transferable electricity and the electricity to be supplemented as capacity constraints. S304. For valid matching links, the transfer quota parameters are encapsulated in a protocol to generate quota adjustment instructions for updating the corresponding quota data of surplus and deficit branches. Specifically, the method for generating quota adjustment instructions is as follows: Extract the surplus branch node identifier, deficit branch node identifier, and transferred quota parameter from the quota transfer relationship corresponding to the effective matching link; Based on the communication protocol pre-agreed and stored between the power distribution control equipment and the branch node, a data frame is constructed, which includes a frame start field, a control identifier field, a data load field, and a cyclic redundancy check field. The communication protocol predefines the field length, byte order, electricity unit, numerical scaling rules, and cyclic redundancy check type for each field. According to the byte order specified in the communication protocol, the communication addresses of surplus branch nodes, deficit branch nodes, and transfer quota parameters processed according to the numerical scaling rules are written into the data payload field, and the data frame is internally represented and transmitted in hexadecimal form. According to the cyclic redundancy check type specified in the communication protocol, the checksum is calculated for the frame start field, control identifier field, and data payload field, and the checksum is written into the cyclic redundancy check field to complete the data frame encapsulation and output the quota adjustment instruction.

[0026] S40. Execute quota adjustment instructions to update the remaining quota and duration deviation of the corresponding household branch; trigger power outage control when the actual electricity consumption of the household branch reaches the remaining quota; obtain actual transfer parameters to update historical performance records, and iteratively reconstruct quota transfer relationships based on the updated duration deviation. The process of executing the quota adjustment instruction to update the remaining quota and duration deviation of the corresponding sub-branch is as follows: S401. Parse the quota adjustment instruction and update the remaining quota in the quota register. Combine the current power load and the remaining committed duration to obtain the updated duration deviation. Specifically, the method for executing quota adjustment instructions to update the remaining quota and duration deviation of the corresponding sub-branch is as follows: After receiving the quota adjustment instruction issued by S30, the received data frame is checked according to the cyclic redundancy check type specified in the preset communication protocol. If the verification fails, the quota adjustment instruction is discarded and a retransmission request is sent to the control center; if the verification succeeds, the surplus branch node identifier, the deficit branch node identifier, and the transfer quota parameter are parsed from the data load field. When entering the backup power distribution state, the current cumulative power consumption reading of each household branch is read, and this reading is used as the reference reading of the smart energy meter corresponding to the first quota update. The remaining quota record of each household branch is established according to the initial remaining quota allocated by the power distribution control equipment. When adjusting the quota in the future, read the benchmark reading of the smart energy meter recorded by the surplus branch and the deficit branch at the most recent quota update, and obtain the actual electricity consumption generated after the most recent quota update based on the difference between the current cumulative electricity consumption reading and the corresponding benchmark reading. Subtract the actual electricity consumption from the remaining quota saved during the most recent quota update to obtain the remaining quota before this quota adjustment; When the transferred quota parameter is less than or equal to the remaining quota of the surplus branch before the current quota adjustment, the transferred quota parameter is deducted from the remaining quota of the surplus branch before the current quota adjustment and added to the remaining quota of the deficit branch before the current quota adjustment to obtain the updated remaining quota. If the transferred quota parameter is greater than the remaining quota of the surplus branch before this quota adjustment, the quota adjustment instruction will not be executed, and the quota transfer relationship will be reconstructed. After the quota adjustment is completed, record the current cumulative electricity readings of the surplus and deficit branches and use them as the benchmark readings of the smart meters corresponding to the most recent quota update; save the remaining quota of the deficit branches before this quota adjustment for use in calculating the actual transfer parameters. Read the current power load of each household branch in the current sampling period; when the current power load is greater than zero, divide the updated remaining quota by the current power load to obtain the expected backup power duration calculated from the current quota update time; When the current electrical load is less than or equal to zero, the division operation is not performed, and the corresponding branch is marked as a branch with no effective load. The remaining commitment duration is obtained by calculating the time difference between the committed power supply end time recorded in the service agreement database and the current quota update time. When the time difference is less than or equal to zero, the remaining committed duration is set to zero. The updated duration deviation is obtained by subtracting the remaining committed duration from the expected backup power duration. The remaining quota and transferred quota parameters use the same electricity unit; The process of triggering a power outage when the actual electricity consumption of a branch circuit reaches the remaining quota, and then iteratively reconstructing the quota transfer relationship based on the updated duration deviation, involves obtaining the actual transfer parameters, updating historical performance records, and then reconstructing the quota transfer relationship accordingly. S402. Monitor the actual electricity consumption of each household branch after the most recent quota update, and execute the power outage control of the corresponding household branch based on the comparison result of the actual electricity consumption and the corresponding remaining quota. Specifically, the method for triggering the power-off control is as follows: Read the current cumulative electricity consumption reading of the smart energy meter deployed at the output end of each household branch, and calculate the difference between the current cumulative electricity consumption reading and the benchmark reading of the smart energy meter corresponding to the most recent quota update to obtain the actual electricity consumption generated by the corresponding household branch after the most recent quota update. When the smart energy meter does not directly output the cumulative power consumption reading, the actual power consumption is obtained by discretely accumulating the power sampling value after the most recent quota update according to the sampling period of the smart energy meter; If the current cumulative electricity consumption reading is less than the benchmark reading corresponding to the most recent quota update, the reading is judged to be abnormal, the actual electricity consumption will not be updated for the time being, and the electricity meter data will be read again. Compare the actual electricity consumption with the corresponding remaining quota saved after the most recent quota update; When the actual electricity consumption is greater than or equal to the corresponding remaining quota, it is determined that the remaining quota of the corresponding branch has been exhausted. Before the corresponding branch has executed the power outage control, a disconnection control signal is sent to its switch actuator to cut off the power supply circuit of the corresponding branch and record the power outage status. When the actual electricity consumption is less than the corresponding remaining quota, the power supply circuit of the corresponding household branch is kept in the conducting state. The process of obtaining the actual transfer parameters to update historical performance records and iteratively reconstructing the quota transfer relationship based on the updated duration deviation is as follows: S403. When the corresponding branch circuit is de-energized, the main power supply is restored, or the quota transfer relationship is reconstructed, calculate the actual transfer parameters corresponding to the indebted branch circuit and update the historical performance record based on the actual transfer parameters. Specifically, the method for updating historical performance records is as follows: When the corresponding branch circuit loses power, the main power supply is restored, or the quota transfer relationship is reconstructed, the actual electricity consumption generated by the deficit branch after the quota adjustment instruction is executed is read, and the remaining quota of the deficit branch before this quota adjustment is read. When the actual electricity consumption is less than or equal to the remaining quota before this quota adjustment, the actual transfer parameter will be set to zero. When the actual electricity consumption exceeds the remaining quota before this quota adjustment, the portion of the actual electricity consumption exceeding the remaining quota before this quota adjustment will be determined as the actual transfer parameter. When the actual transfer parameter is greater than the transfer quota parameter in the quota adjustment instruction, the transfer quota parameter in the quota adjustment instruction will be determined as the actual transfer parameter. The actual transfer parameter is compared with the transfer quota parameter multiplied by a preset tolerance coefficient. The preset tolerance coefficient is a dimensionless ratio between 0 and 1, which is used to allow for the metering error of the smart energy meter and the quota control error. When the actual transfer parameter is greater than or equal to the comparison result, the quota replenishment event will be recorded as completed. When the actual transfer parameter is less than the comparison result, the performance result is recorded according to the cause of event termination. When the reason for termination is insufficient available power of the energy storage device, this quota replenishment event will be recorded as incomplete and counted in the number of incomplete events caused by insufficient available power of the energy storage device. When the termination is due to the main power supply being restored, the power supply to the branch line being cut off, or the communication link being interrupted, the corresponding termination reason should be recorded, but this event should not be counted as an incomplete event due to insufficient available power of the energy storage device. Each historical performance record should at least save the transferred quota parameters, actual transferred parameters, quota adjustment time, and performance result; The quota replenishment completion rate is calculated by dividing the number of completed tasks by the sum of the number of completed tasks and the number of uncompleted tasks due to insufficient available power of the energy storage equipment; When the sum of the number of times is zero, historical performance evaluation is not performed based on the completion rate, and the process is handled in accordance with the method for no valid historical performance record in S203. S404. After completing the update of historical performance records, re-enter the current duration deviation into S20, and execute S201 to S212 in sequence to rebuild the quota transfer relationship.

[0027] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A method for intelligent power distribution of communication power supply based on power distribution control equipment, characterized in that, Includes the following steps: The power load and energy reserve of each branch circuit under the main power interruption are obtained. The power load and energy reserve are analyzed for backup power to obtain the expected backup power duration of each branch circuit. The backup power deviation is analyzed for the expected backup power duration to obtain the duration deviation of each branch circuit. Based on the time deviation, the branch is classified into surplus branch and deficit branch. Historical performance records are obtained and the allocation weight is generated by combining the time deviation. Construct a quota transfer relationship between surplus and deficit branches based on allocation weights; Monitor the residual capacity decay status of energy storage equipment and dynamically set quota adjustment trigger conditions based on the residual capacity decay status; when the quota adjustment trigger conditions are met, parse the order of transfer in the quota transfer relationship to match the surplus branch and deficit branch participating in the adjustment, and generate quota adjustment instructions based on the quota transfer relationship. Execute quota adjustment instructions to update the remaining quota and duration deviation of the corresponding household branch; trigger power outage control when the actual electricity consumption of the household branch reaches the remaining quota; obtain actual transfer parameters to update historical performance records, and iteratively reconstruct quota transfer relationships based on the updated duration deviation.

2. The intelligent power distribution method for communication power supply based on power distribution control equipment according to claim 1, characterized in that: The process of performing the backup power deviation analysis is as follows: Obtain the preset commitment duration for each household branch, and extract the duration deviation for each household branch based on the difference between the expected backup power duration and the preset commitment duration. Then, split the duration deviation into positive duration deviation and negative duration deviation.

3. The intelligent power distribution method for communication power supply based on power distribution control equipment according to claim 2, characterized in that: The process of obtaining the estimated backup power duration is as follows: Extract historical power load data for each household branch and input it along with current power load data into a pre-trained temporal convolutional network to output the predicted power sequence for each household branch within a preset prediction time period. The predicted power sequence of each household branch is used to allocate electricity to obtain the theoretical allocated electricity for each household branch. The estimated backup power duration for each household branch is calculated based on the predicted power sequence and theoretical power allocation for each branch.

4. The intelligent power distribution method for communication power supply based on power distribution control equipment according to claim 1, characterized in that: The process of establishing the quota transfer relationship is as follows: The order of acceptance of the deficit branches is determined according to the allocation weight, and a source-sink traffic transmission topology including surplus branch nodes and deficit branch nodes is established. The quota transfer path and transfer quota parameters in the source-sink traffic transmission topology are analyzed to generate a quota transfer relationship that includes transfer quota parameters and the order of transfer.

5. The intelligent power distribution method for communication power supply based on power distribution control equipment according to claim 4, characterized in that: The process of establishing the source-sink traffic transmission topology is as follows: Based on the descending order of the allocation weights, the branch nodes in the set of deficit branches are sorted in descending order, and the sorted sequence of branch nodes is determined as the order of acceptance. Establish logical source nodes and logical sink nodes, and connect the logical source nodes to each surplus branch node whose theoretically transferable electricity is greater than zero. Each surplus branch node is connected to each deficit branch node with a power supply greater than zero, and each deficit branch node is connected to a logical sink node to obtain the source-sink traffic transmission topology.

6. The intelligent power distribution method for communication power supply based on power distribution control equipment according to claim 4, characterized in that: The process of obtaining the assigned weights is as follows: Construct a fuzzy membership function based on duration deviation to convert the duration deviation of each household branch into a backup power status membership degree; Based on the membership degree of the backup power status, each household branch is classified into surplus branches and deficit branches. The theoretical transferable power is obtained based on the positive duration deviation, and the power to be supplemented is obtained based on the negative duration deviation. A set of surplus branches and a set of deficit branches are established. Extract the historical performance records of each branch in the set of deficient branches, use the time decay method to obtain the historical performance completion degree of each deficient branch, and standardize the historical performance completion degree and the amount of electricity to be made up to obtain a standardized decision matrix. The information entropy weight method is used to integrate the historical degree of insufficient replenishment and the amount of electricity to be replenished in the standardized decision matrix to generate the allocation weight of each deficit branch.

7. The intelligent power distribution method for communication power supply based on power distribution control equipment according to claim 6, characterized in that: The process of obtaining the aforementioned historical performance completion level is as follows: Extract the historical performance records of each defaulting branch within a pre-configured historical time window; The historical performance record includes transfer quota parameters, actual transfer parameters, quota adjustment time, and performance results; The time interval between the quota adjustment time and the current time is divided into multiple time intervals, and a numerical time weight is pre-configured for each time interval; For historical performance records where the transfer quota parameter is greater than zero and the performance result is either normal completion or termination due to insufficient available power of the energy storage equipment, the ratio of the actual transfer parameter to the transfer quota parameter is calculated as the performance completion ratio of the corresponding historical performance record. The completion rate of each valid historical performance record is multiplied by its corresponding time weight, and the sum of the multiplication results is divided by the sum of the corresponding time weights to obtain the historical performance completion rate of the corresponding deficit branch.

8. The intelligent power distribution method for communication power supply based on power distribution control equipment according to claim 1, characterized in that: The process of generating the quota adjustment instruction is as follows: Continuously monitor high-priority adjustment events and future timestamps. After the quota adjustment triggering conditions are met, perform link status handshake verification on the corresponding branch according to the order of transfer in the quota transfer relationship to determine the valid matching link. For valid matching links, the transfer quota parameters are encapsulated in a protocol to generate quota adjustment instructions for updating the corresponding quota data of surplus and deficit branches.

9. A method for intelligent power distribution of communication power supply based on power distribution control equipment according to claim 8, characterized in that: The process of setting the quota adjustment trigger condition is as follows: A real-time observation window for energy storage devices is constructed and the capacity time series is extracted. The capacity time series is then smoothed and first-order derivative is calculated to obtain the residual decay rate of the energy storage devices. Based on the residual decay rate, query the pre-configured trigger strategy table to determine the quota adjustment trigger method and monitoring period, and set the trigger method and monitoring period as the quota adjustment trigger conditions.

10. A method for intelligent power distribution of communication power supply based on power distribution control equipment according to claim 1, characterized in that: The process of iteratively reconstructing the quota transfer relationship is as follows: Monitor the actual electricity consumption of each household branch after the most recent quota update, and execute power outage control of the corresponding household branch based on the comparison result of the actual electricity consumption and the corresponding remaining quota. When the corresponding branch circuit is de-energized, the main power supply is restored, or the quota transfer relationship is reconstructed, the actual transfer parameters corresponding to the indebted branch circuit are calculated, and the historical performance record is updated based on the actual transfer parameters. After updating the historical performance record, the current duration deviation is re-entered into the process of constructing the quota transfer relationship between the surplus branch and the deficit branch, thereby realizing the re-establishment of the quota transfer relationship.

Images