A power supply and distribution system performance health state dynamic monitoring system

Abstract

The present application relates to the technical field of power supply and distribution monitoring, in particular to a power supply and distribution system performance health state dynamic monitoring system, comprising: a state acquisition module, a power supply switching module, a power supply monitoring module, a switching matching module and a state evaluation module; the power supply equipment is combined as a power supply distribution network by taking the power supply equipment as a power supply node, the switching behavior characteristics of the current power supply node are viewed according to the running time and the switching number of the power supply node according to the running information of the power supply node; the similarity when the power supply mode is switched is checked with the voltage deviation and the load deviation of the running information, and the running information of each power supply node is updated; the switching mode of the current power supply node under the power supply mode switching is described with the switching behavior characteristics as the starting point, the switching sequence of the switching mode is used to describe the switching matching network of each power supply node, and finally the optimal configuration scheme corresponding to each power supply node is screened; the efficiency of abnormal detection and the reliability of switching of the power supply and distribution system are realized.

Description

Technical Field

[0001] This invention relates to the field of power supply and distribution monitoring technology, specifically a dynamic monitoring system for the performance and health status of power supply and distribution systems. Background Technology

[0002] A power supply and distribution system is a system for supplying and distributing electrical energy. Its main function is to obtain electrical energy from the power grid and, through reasonable line and equipment configuration, safely, reliably, economically, and efficiently deliver the electrical energy to various electrical loads to meet the electricity needs of different users. However, existing technologies largely rely on manual inspections and fixed threshold alarms, making power supply and distribution systems unable to adapt to the complex topologies of distributed power sources and microgrids.

[0003] For example, Chinese Patent Publication No. CN119253869A discloses an intelligent monitoring method and system based on power supply and distribution systems, relating to the field of power grid monitoring technology. The method includes acquiring detected power parameters; establishing detection intervals and determining interval change parameters; determining interval change combinations based on all interval change parameters; establishing historical intervals and randomly defining comparison intervals with a width equal to the detection duration within the historical intervals, and determining comparison change combinations within each comparison interval; determining combination similarity values ​​based on interval change combinations and comparison change combinations, and determining the combination similarity value with the largest value according to a sorting rule, and defining the comparison interval corresponding to the combination similarity value as a similar interval; establishing prediction intervals based on similar intervals, and defining the detected power parameters in the prediction intervals as predicted power parameters; and outputting an aging warning signal when there is a situation where the predicted power parameter is greater than a preset warning power parameter.

[0004] For example, Chinese Patent Publication No. CN117913970A discloses an adaptive switching system for power supply and distribution, including a status monitoring module, a power monitoring module, a voltage analysis module, a status analysis module, and a switching determination module. This invention monitors the voltage of power generation equipment in real time and performs matching and comparison judgments. When the supply voltage fluctuates significantly, it analyzes the power supply status of the equipment by correlating current data. By determining whether the power generation equipment matches the load, it analyzes whether there is a risk of failure in the power generation equipment, predicts the safe operating status of the power generation equipment, and determines whether it is necessary to control the power generation equipment to shut down in a timely manner to avoid failure. When load demand increases, it isolates the excess load and performs a gradual switching of the power load. When load demand decreases, it directly switches the power load to another power generation equipment.

[0005] Existing technologies describe using time intervals of distribution nodes to predict aging warning signals based on similarity intervals during matching, and using transformer rates to identify the current load requiring switching. However, existing technologies use fixed thresholds for power parameters and identify similarity intervals to describe the current power supply and distribution scenario. This leads to the identification of voltage and other parameters not adapting to load changes and ignoring voltage correlations before and after distribution switching. Consequently, it becomes difficult to identify system performance and voltage instability during distribution switching, reducing the safety and efficiency of the power supply and distribution system. Summary of the Invention

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a dynamic monitoring system for the performance and health status of a power supply and distribution system, comprising: a status acquisition module, used to collect the operating information of power supply equipment under various power supply modes, and to combine the power supply equipment as power supply nodes into a power distribution network, wherein the operating information includes voltage, frequency and load rate.

[0007] The power supply switching module is used to view the switching behavior characteristics of the current power supply node based on the power supply node's operating information, according to the power supply node's operating time and switching count.

[0008] The power supply monitoring module is used to view the operating information before and after the power supply mode switching. It checks the similarity of different power supply mode switching based on the voltage deviation and load deviation of the operating information, and updates the operating information of each power supply node based on the similarity of the power supply mode.

[0009] The switching matching module is used to describe the switching mode of the current power supply node under the power supply mode switching based on the switching behavior characteristics, and to describe the switching matching network of each power supply node based on the switching order of the switching mode.

[0010] The status assessment module is used to generate multiple power supply configuration schemes based on the updated operating information of each power supply node and the switching matching network combination, and to select the optimal configuration scheme for each power supply node based on the power supply method involved in the power supply configuration scheme.

[0011] The beneficial effects of this invention are as follows: First, this invention constructs a power distribution network based on the topology of the equipment, calculates the voltage deviation amplitude, frequency deviation amplitude, and load distribution ratio as feature values, and uses the feature difference between adjacent time points and the relative number of power supply nodes to mark the current power distribution network to indicate abnormal equipment existing in the power supply and distribution switching operation state; then, it statistically analyzes the switching frequency and average switching time to identify the existing abnormal patterns, and models the switching behavior feature difference between adjacent nodes and conditional probability. After identifying combinations with similar switching behaviors, the output data is used as its switching behavior features to identify the behavioral characteristics of power supply and distribution switching when anomalies occur, thereby realizing adaptive prediction and anomaly early warning of switching behavior.

[0012] Second, this invention constructs a similarity matrix based on the time series distribution of voltage deviation and load deviation. After labeling the deviation type corresponding to the current power supply node with the similarity matrix, the deviation type is labeled in the form of time points, thereby improving the temporal correlation during handover tracking.

[0013] Third, this invention categorizes switching methods based on parameters such as voltage interruption time and power type. It then constructs a switching matching network to describe the switching behavior under corresponding parameters, thus describing the multi-path switching sequence and device dependencies. Furthermore, by utilizing the switching matching network and updated operational information, combined with the least squares method, it solves for the optimal target mode of each power supply node under the current power supply configuration scheme, thereby improving the parameter configuration efficiency and processing effect for multi-parameter optimization scenarios. Attached Figure Description

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] Figure 1 This is a system framework diagram of a dynamic monitoring system for the performance and health status of a power supply and distribution system.

[0016] Figure 2 This is a flowchart illustrating the status acquisition module of a dynamic monitoring system for the performance and health status of a power supply and distribution system.

[0017] Figure 3 This is a flowchart of a power supply monitoring module in a dynamic monitoring system for the performance and health status of a power supply and distribution system.

[0018] Figure 4 This is a schematic diagram of the switching mode in the switching matching module of a dynamic monitoring system for the performance health status of a power supply and distribution system.

[0019] Figure 5 This is a flowchart illustrating the status assessment module of a dynamic monitoring system for the performance health status of a power supply and distribution system. Detailed Implementation

[0020] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. Where specific techniques or conditions are not specified in the embodiments, they shall be performed in accordance with the techniques or conditions described in the literature in the art or in accordance with the product manual.

[0021] See Figure 1 A dynamic monitoring system for the performance and health status of a power supply and distribution system includes: a status acquisition module, a power supply switching module, a power supply monitoring module, a switching matching module, and a status evaluation module; the output of the status acquisition module is connected to the power supply switching module, the output of the power supply switching module is connected to the power supply monitoring module, the output of the power supply monitoring module is connected to the switching matching module, and the output of the switching matching module is connected to the status evaluation module.

[0022] The status acquisition module is used to collect the operating information of power supply equipment under various power supply methods, and combine the power supply equipment as power supply nodes into a power distribution network. The operating information includes voltage, frequency and load rate.

[0023] The power supply switching module is used to view the switching behavior characteristics of the current power supply node based on the power supply node's operating information, according to the power supply node's operating time and switching count.

[0024] The power supply monitoring module is used to view the operating information before and after the power supply mode switching. It checks the similarity of different power supply mode switching based on the voltage deviation and load deviation of the operating information, and updates the operating information of each power supply node based on the similarity of the power supply mode.

[0025] The switching matching module is used to describe the switching mode of the current power supply node under the power supply mode switching based on the switching behavior characteristics, and to describe the switching matching network of each power supply node based on the switching order of the switching mode.

[0026] The status assessment module is used to generate multiple power supply configuration schemes based on the updated operating information of each power supply node and the switching matching network combination, and to select the optimal configuration scheme for each power supply node based on the power supply method involved in the power supply configuration scheme.

[0027] It should be noted that multiple ammeters and voltmeters are used at the location of the power supply equipment to identify the operating status of multiple power supply devices within the entire power distribution network.

[0028] The status parameter module first collects the current power supply method, such as the actual power supply equipment, whether backup equipment is used for power supply, and the voltage, frequency and load rate of the power supply equipment during operation. These relevant data are used as the operating information collected at this time.

[0029] like Figure 2As shown, the status acquisition module also includes: treating power supply equipment as power supply nodes and forming a power distribution network based on the connection relationship of each power supply equipment.

[0030] Using the voltage, frequency, and load rate of each power supply node in the power distribution network, the voltage deviation amplitude, frequency deviation amplitude, and load distribution ratio of each power supply node are calculated and used as characteristic values ​​of the operation information.

[0031] Determine the change characteristic value of the operation information at two adjacent time points, and determine the degree of change of the power distribution network under the corresponding power supply mode based on the number of nodes corresponding to each change characteristic value. Mark the power supply equipment based on the difference of the degree of change between two adjacent time points.

[0032] Preferably, the power supply equipment can be represented as transformers, switch cabinets, generator sets, UPS uninterruptible power supplies, and photovoltaic inverters, etc., to describe the current continuous status under power supply. The identified voltage and frequency can help check the stability of the current passing through each device. As for the load rate, it is mainly for the generator sets, UPS power supplies, and other devices directly related to power generation in the power supply equipment, to check whether there is an excessive load on the nodes of these devices that causes the devices to malfunction.

[0033] The voltage deviation amplitude and frequency deviation amplitude mentioned above are used to describe the proportional values ​​when deviations occur, while the load distribution ratio is the proportion of each node relative to the total number of nodes calculated based on the extracted load rate.

[0034] Preferably, the aforementioned degree of change is used to describe the weighted sum of voltage deviation amplitude, frequency deviation amplitude, and load distribution ratio to indicate whether the current is stable at two adjacent moments. If the degree of change has a large value, it can be considered that there may be an unstable abnormality under the current power supply method. If the value is small, it indicates that it is relatively stable, indicating that the operating state is relatively stable.

[0035] That is, the degree of change Represented as: ;in, This represents the degree of change at time t; It is represented as the characteristic value of the change in voltage deviation amplitude of the i-th power supply node at time t; This is represented by the characteristic value of the change in the frequency deviation amplitude of the i-th power supply node at time t; This is represented by the characteristic value of the change in the load distribution ratio of the i-th power supply node at time t; , and These represent the weights corresponding to the voltage deviation amplitude, frequency deviation amplitude, and load distribution ratio, respectively. The weight values ​​can be set to 0.4, 0.3, and 0.3, respectively, or other values ​​can be used, and their weight sums are set to 1. This represents the number of power supply nodes, with i ranging from 1 to N. The weighted sum for the degree of change is calculated based on the number of nodes involved, examining each power supply node, and then using the average value and standard deviation obtainable from the degree of change as its threshold. The average value plus three times the standard deviation is used as the standard threshold for judging whether the current is stable between two adjacent time points. Then, the corresponding power supply equipment is marked based on the difference in the degree of change to indicate whether there are related problems under the current power supply method.

[0036] Preferably, when combined into a power distribution network, the implementation method further includes: based on the topological relationship of the power supply equipment, detecting the continuous operating time of the power supply equipment under the corresponding current supply demand, and using the operating information of any power supply equipment when it is turned on, calculating the proportion of the operating time of each power supply equipment to all power supply equipment in the power distribution network, and connecting the power supply equipment with the same operating time proportion as adjacent power supply nodes.

[0037] At this point, by describing the connections between power supply equipment and physical devices, and connecting parts of these devices with the same operating duration ratio, the load can be evenly distributed among the devices when comparing their operating information, preventing single-point overload. Since devices with the same proportion usually have similar operating states, connecting them forms redundant paths, improving the network's tolerance to device failures. When identifying adjacent nodes, the power distribution network not only checks for physically connected power supply equipment but also checks for devices with the same operating duration ratio, further improving the network's tolerance to device anomalies. In the event of a failure in some devices, the redundantly connected devices can be quickly switched to maintain power continuity.

[0038] It should be noted that when power supply equipment related to runtime ratio is described as an additional adjacent node, the topology between related power supply equipment also needs to be traversed. If the circuits or other parts corresponding to the topology cannot be connected, the corresponding power supply equipment will not be connected. Only some power supply equipment connected in the same topology will be selected to prevent the situation where some power supply nodes are not related when connecting. At the same time, the purpose of considering adjacent power supply nodes is to associate non-directly adjacent power supply nodes that can be under the same power demand to illustrate whether the relevant content of the operating information of multiple power supply nodes under the same power demand is similar or highly correlated, so as to help identify whether the voltage, frequency and load rate of each power supply device in the overall power supply mode are operating normally.

[0039] In one embodiment of the present invention, when viewing relevant information of a power supply node during switching, the main focus is on the timing of each switching, the type of power supply connected before and after the switching, whether the switching was successful, and the voltage interruption time or frequency offset during the switching. These parameters, which are directly related to the current during switching, are used to identify whether the current power supply switching process is in a normal state.

[0040] The power type before and after the switch directly indicates the power supply node currently in use. A successful switch reflects whether the work was completed normally under the number of switches. The voltage interruption time and frequency offset indicate the relevant status of the operating information. At this time, the frequency and average switching time corresponding to the switching time point are mainly used to check whether the current power supply equipment can complete the power supply and distribution process normally. If there is an abnormality, the corresponding content will be marked and output as switching behavior characteristics. Based on the switching behavior characteristics in the power supply and distribution scenario, the operating status of the entire system is judged.

[0041] The implementation of the power supply switching module also includes: calculating the switching frequency and average switching time of the current power supply node based on its operating time and switching frequency. Knowing the operating time and switching frequency of the power supply node, the current switching point and the average switching time within the operating time can be identified to describe the overall switching process. Simultaneously, using the switching frequency to describe the current number of switching operations allows the power supply node to be described within a relatively fixed time period in the form of a time interval or unit time when switching power supply modes, preventing abnormal over-switching and the problem of insufficient switching frequency leading to increased load on other power supply nodes.

[0042] Threshold detection is performed on switching frequency and average switching time to identify the switching behavior characteristics of the current power supply node. These characteristics are represented as normal switching, frequent switching, abnormal switching, and switching time delay, describing whether the current power supply node is in at least one of the following states when performing power supply mode switching: normal, warning, abnormal, or awaiting maintenance. The threshold detection method is based on the switching frequency and average switching time of the power supply node in historical data, forming a confidence interval corresponding to these values. The upper and lower limits of the confidence interval at the 95% level are extracted as thresholds. If the switching frequency and average switching time are within the confidence interval, it is considered normal. If the switching frequency and average switching time are close to the upper or lower limits of the confidence interval, and the difference from the upper or lower limits is less than 5%, it is considered to enter a warning state. If they are greater than the upper limit of the confidence interval or less than the lower limit, it is considered abnormal. If multiple consecutive time periods are considered abnormal, maintenance is required.

[0043] At this point, the test needs to check whether the currently selected power supply node can switch power supply modes normally in the power supply and distribution scenario, and whether the running time can be stable.

[0044] As for identifying the switching behavior characteristics of the current power supply node, the implementation method also includes: based on the threshold detection results of the same power supply node, performing a combined judgment on adjacent power supply nodes, and determining the adjacent change combination based on the difference in switching behavior characteristics of adjacent power supply nodes.

[0045] Based on the conditional probability of adjacent changes in the value of switching behavior features on the time axis, a preset time interval corresponding to the current time point is established. Multiple comparison intervals of equal size are defined in the preset time interval to identify the change probability of the conditional probability of switching behavior feature values, and these are used as comparison behavior feature combinations.

[0046] The switching behavior feature is output based on the part with the highest similarity among the combinations of comparative behavioral features.

[0047] The aforementioned handover behavior characteristics are composed of a feature vector based on the handover frequency and average handover time. Power supply nodes adjacent to the current power supply node in the power distribution network are selected to check whether there are differences in their handover behavior characteristics. After these power supply nodes are grouped into multiple adjacent change combinations, the correlation between the combined power supply nodes is described by the conditional probability of the handover behavior characteristic difference in multiple time periods on the time axis.

[0048] For example, suppose there are three adjacent power supply nodes. This adjacent relationship means that after adjacent nodes exist on the power distribution network, the four states of normal, warning, abnormal, and pending maintenance are respectively labeled with four forms: S0, S1, S2, and S3. The probability values ​​of the current power supply node's switching frequency and average switching time in these four states are calculated. Then, using the formula for conditional probability, the probability values ​​of the three power supply nodes are expressed using the formula for conditional probability. That is, given that power supply node A is in any of the four states of normal, warning, abnormal, or pending maintenance, the conditional probability values ​​of power supply nodes B and C in the corresponding states are calculated. Then, each conditional probability is... To compare multiple time periods and determine if changes exist, similarity is calculated using multiple change probabilities, such as the Pearson correlation coefficient. After inputting multiple change probabilities, a similarity value is obtained. Then, the time period corresponding to the largest similarity value and the related power supply node are identified and output as the switching behavior features to be processed later. The resulting switching behavior features will show data combinations where there is a high correlation between multiple adjacent power supply nodes, and they may be in abnormal or normal states simultaneously, as well as the correlation between multiple power supply nodes during abnormal time periods. This facilitates subsequent processing of the correlation between power supplies to identify whether the current power supply and distribution system is operating normally.

[0049] Preferably, the above-mentioned adjacent change combination represents a group of adjacent power supply nodes with a specific switching behavior characteristic difference. At this time, the adjacent change combination will be divided into multiple adjacent change combinations in pairs based on the different values ​​of their switching behavior differences.

[0050] Preferably, the aforementioned preset time interval represents the time interval extracted for the current time point. This time interval is extracted at half the size of the time window for collecting the operating information of the power supply equipment, in order to extract the state changes of different power supply nodes under power supply mode switching within a local time period, and to use the content of the power supply node in any state of normal, warning, abnormal and pending maintenance as the state identifier of the corresponding switching behavior characteristics.

[0051] Preferably, the above-mentioned combination of comparative behavioral features describes the progress process after the power supply mode switching under relative time by describing the change probability of conditional probability.

[0052] In one embodiment of the present invention, the power supply monitoring module focuses on checking voltage deviation and load deviation, and uses the deviation under different power supply methods to illustrate the overall stability of the power supply. Then, the similarity value is used to update the content marked on each power supply node to indicate whether each power supply device can operate normally under multiple switching. If it cannot, the similarity value is used to mark it, and the switching status of the relevant power supply device under the power supply method is checked and described again.

[0053] At this point, if there is a voltage deviation before and after the switch, it generally manifests as voltage drop, voltage offset, and voltage imbalance. For example, a voltage drop will cause the voltage to suddenly drop at the moment of switching, which will lead to a surge in current and reduce the stability of the overall power supply output current. Voltage offset indicates that the voltage deviates from the rated value after switching. The similarity identification can be based on long-term operation phenomena to identify whether there is a long-term reduction problem, and whether this phenomenon will lead to an increase in steady-state current, ultimately affecting the stability of the power supply system. As for voltage imbalance, it will directly show phenomena such as unbalanced three-phase current, causing motor heating. At this time, it is necessary to identify the type of voltage deviation by using the sequence length corresponding to the voltage deviation and the corresponding similarity, and then describe the operating information that needs to be updated and processed.

[0054] Preferably, the load deviation and voltage deviation represent similar situations. For example, load deviation generally indicates phenomena such as sudden load increase, sudden load decrease, and uneven load distribution. A sudden load increase indicates that a new high-power device may be added during the power supply mode switch, leading to a sudden increase in current and current exceeding the limit. A sudden load decrease indicates that after the power supply mode switch, there are some branch disconnections, resulting in a decrease in the overall switching current and impact on the power supply and distribution system. Finally, uneven load distribution indicates that there are excessive differences in the load of each node during the power supply mode switch, resulting in current imbalance and local overload in some locations. At this time, the load deviation is used to verify whether each power supply node can maintain stable operation before and after the power supply mode switch. The load rate corresponds to some types to illustrate whether the operation of the distribution node meets expectations under voltage and load deviation.

[0055] like Figure 3As shown, the implementation of the power supply monitoring module includes: taking the distribution position of voltage deviation and load deviation of each power supply node before and after the switch in the time series, and taking the distribution position as the target data point, the similarity is calculated based on the instantaneous value of the target data point and the time series in which the target data point is located, to form a similarity matrix corresponding to the target data point; this similarity matrix is ​​used to describe the similarity of the instantaneous value and long-sequence value of voltage deviation and load deviation, respectively. After these similarity values ​​are combined into a similarity matrix, the working status of each power supply node in the corresponding time period before and after the power supply mode switch is described. Then, the similarity matrix is ​​used to complete the problem identification to obtain whether there is an abnormal situation corresponding to the voltage deviation and load deviation. After marking the problem, the operation information of each power supply node before and after the power supply mode switch is updated.

[0056] Preferably, when calculating the similarity in the similarity matrix, the values ​​of the voltage deviation and load deviation in the time series are extracted at their respective positions. Then, the similarity value of each power supply node at the corresponding time period in the time series is obtained with the corresponding voltage deviation or load deviation in the historical data using the Pearson correlation coefficient. Then, a similarity matrix containing all similarity values ​​is formed in the order of each power supply node's label. If it is necessary to calculate using instantaneous values, the similarity is calculated with the value at the corresponding time point and the historical data, and then the similarity matrix is ​​obtained. At the same time, in the obtained similarity matrix, the similarity matrices corresponding to voltage deviation and load deviation will be separated or merged into a single similarity matrix. When merged, the similarity matrix stores each element in the form of data pairs.

[0057] Based on the similarity matrix corresponding to the target data point, the deviation type of the target data point in the predefined deviation type is determined, and the operation information of each power supply node is updated according to the time interval of the deviation type of the target data point in the time series.

[0058] The implementation of the deviation type also includes: based on the similarity matrix corresponding to the target data point, clustering each power supply node according to the similarity corresponding to voltage deviation and load deviation respectively, and identifying the cluster where each power supply node belongs. At this time, hierarchical clustering or DBSCAN algorithm can be used to group the power supply nodes; for example, first clustering a portion of the data corresponding to the similarity matrix of voltage deviation, and then clustering a portion of the data corresponding to load deviation. At this time, the current similarity matrix is ​​used as input, and the clusters set for different deviation types in historical data are input into the corresponding clusters. Then, after identifying the deviation type that the current target data point can correspond to, the corresponding deviation type is labeled to each power supply node in the form of time points.

[0059] When defining the deviation type corresponding to each power supply node, the deviation type includes, but is not limited to, load surge, load drop, uneven load distribution, voltage drop, voltage deviation, and voltage imbalance. This is used to describe the deviation type that exists in each power supply node. The interval time obtained afterwards indicates whether there will be a certain interval time when the relevant deviation type occurs. If there is, the interval time is updated to the operation information; otherwise, only the deviation type is updated.

[0060] In one embodiment of the present invention, when using switching behavior features, the data viewed includes voltage interruption time, power supply type connected during switching, and voltage values ​​before and after the voltage interruption time; the switching behavior features corresponding to these data are used as the identification subject, and the specific power supply node that has switched after the switching behavior features have completed the power supply mode switching at each power supply node is used as its switching mode, and then the switching sequence is displayed in the form of a network diagram.

[0061] Therefore, the implementation of the switching matching module also includes: using the status flag of the switching behavior feature as the basic condition of each power supply node, using the voltage interruption time, power type and voltage value before and after the voltage interruption time of each power supply node to segment the switching behavior feature and define the switching mode corresponding to the switching behavior feature; and forming the switching matching network of each power supply node by the transition of the status flag of the switching mode before and after the switching and by the switching order of the switching mode in the time series.

[0062] The aforementioned switching method is used to further describe whether each power supply node is in any of the states corresponding to normal, abnormal, warning, and pending maintenance when the power supply method is switched, and whether the voltage interruption time, power type, and voltage value before and after the voltage interruption time of the power supply node are within normal ranges. Although the voltage value of the power supply node is determined in the switching behavior characteristics, it is also necessary to explain whether the switching is due to a problem or a problem occurs after the switch, under the different power type and voltage value before and after the switch. These existing states are then used to form a switching matching network with a state transition pointing diagram, and connected to multiple power supply nodes when the switching behavior characteristics are implemented. These power supply nodes are used as the output at this time to illustrate the real-time working status of multiple devices configured under the current power supply and distribution system.

[0063] The switching order described above is used to explain the order of the switching behavior characteristics of the current input in the time series. This order will represent the order of abnormal situations with similar patterns in the power supply mode switching scenario, in order to describe the abnormal state changes that occur at different power supply nodes.

[0064] like Figure 4As shown, when defining switching modes based on switching behavior characteristics, we can categorize them based on power type, voltage, and timing-related issues. The figure illustrates some potential problems using these three aspects. This categorization of switching modes can then explain the switching behavior characteristics under different circumstances. When these switching modes are combined into a switching matching network according to the identified situations before and after the switching, the time of power supply mode switching indicates whether there are any related anomalies among the multiple power supply nodes in the current switching matching network. After displaying the potential problems of the power supply nodes, we gradually connect them according to the switching time points of each power supply node under the timing sequence to form an image identifying whether there are any anomalies and related anomaly problems among all relevant power supply nodes under the corresponding power supply mode.

[0065] In one embodiment of the present invention, the status assessment module needs to integrate the operation and switching status of a single power supply node in the above four modules to describe the power supply configuration scheme of multiple nodes under mutual influence, and the contribution of each node to the overall power supply and distribution system fault under the power supply configuration scheme, so as to illustrate the influence between multiple nodes. For example, in a parallel time sequence manner, the data obtained by each power supply node is comprehensively judged upstream and downstream, and the main reasons for the abnormality of the corresponding power supply node are returned to complete the health monitoring of the power supply and distribution system.

[0066] like Figure 5 As shown, the implementation of the status assessment module includes synchronizing the updated operating information of each power supply node to the switching matching network according to the switching order of the corresponding time series.

[0067] Identify the connection paths of the switching matching network, and identify and process the upstream and downstream nodes of each power supply node according to the connection order within each connection path. At this time, there will be multiple paths in the switching matching network. For example, there is a connection path from the main node to the static switch to the UPS power supply to the generator to the power distribution unit to the cabinet. In addition to the power supply node represented by the initial main node, there can be multiple groups of subsequent devices. At the same time, there are also connection paths in some scenarios where no generator is set. In this case, it is necessary to process the status of each device in each connection path and the corresponding order during switching in a parallel timing manner based on the causal relationship represented by the upstream and downstream nodes to describe the relative causal relationship between each power supply node and the part of the power supply configuration scheme that needs to be processed.

[0068] The implementation method for identifying and processing upstream and downstream nodes of each power supply node includes: responding to the upstream node of the current power supply node, configuring the power supply configuration scheme of each upstream node according to the number of state changes when pointing to the current power supply node, based on the state identifier of the upstream node.

[0069] When configuring upstream nodes, the main focus is on checking whether the upstream node has a normal, warning, abnormal, or maintenance-pending status under corresponding switching behaviors. If an upstream node is abnormal, it is highly likely to cause a corresponding abnormal situation in the current node. This process is repeated for each node to obtain its power supply configuration scheme in response to the upstream node. For upstream nodes, the focus is on understanding the causal relationship between the current power supply node and the upstream node after an anomaly occurs, in order to resolve potential related anomalies.

[0070] In response to downstream nodes of the current power supply node, the power supply configuration scheme of the downstream node is configured based on the deviation type of the current power supply node and the number of times the deviation type is repeated when the current node points to the downstream node. The deviation type mainly describes the voltage, load, and frequency-related deviations, as well as the updated voltage and load deviations. Then, the deviation type of the current node is examined to explain its main characteristics. Finally, it explains how the downstream node adjusts the current-related voltage, load, and frequency components in response to the deviation type of the current node to achieve overall power supply and distribution stability.

[0071] The aforementioned method of separating upstream and downstream nodes is mainly to prevent anomalies in parallel scenarios from affecting the current node. In such cases, the current power supply node may correspond to multiple parallel upstream and downstream nodes. Parallel timing and other methods are needed to separate the power supply nodes in parallel to prevent anomalies in parallel scenarios from affecting the current node.

[0072] Preferably, when using the parallel timing processing method, the main purpose is to solve the timing-related problems that exist after the switching matching network is connected, and to reduce the problems of time recognition misalignment and time recognition delay caused by transmission delay during partial data acquisition. These problems are verified by upstream and downstream of multiple nodes and the types of deviations that may be caused by voltage, frequency and load during the update are collected. The abnormal connection paths of the current power supply node under parallel timing are displayed one by one to illustrate how the power supply configuration scheme after different power supply node combinations should be set. Then, the set power supply configuration scheme needs to be adjusted to select the optimal configuration scheme that can complete the processing of the current power supply node.

[0073] The implementation of the state assessment module further includes: statistically analyzing power supply configuration schemes on each connection path, and filtering these schemes based on upstream and downstream nodes of each connection path; during filtering, iterating through the power supply configuration schemes of upstream and downstream nodes for the target being processed, such as the power supply equipment to be adjusted, the voltage amplitude to be adjusted, etc., using each power supply configuration scheme as the starting point for filtering, and obtaining global and local optimal solutions for the optimization objectives in the configuration schemes. Based on the power supply configuration schemes corresponding to the local and global optimal solutions, the current power supply node is adjusted, and the optimal configuration scheme is output. Finally, using the filtered optimal multi-objective solution, the corresponding power supply configuration scheme is output as the optimal configuration scheme.

[0074] Preferably, when upstream and downstream nodes screen power supply configuration schemes, the available power supply configuration schemes for each power supply node are extracted. For example, in the configuration scheme of upstream nodes, the equipment and parameters that need to be adjusted, such as input voltage thresholds, are extracted. If an upstream node is marked as abnormal, the impact of the current node, such as voltage drop, is marked. Then, the parameters that need to be adjusted, such as voltage deviation, are extracted from the configuration scheme of downstream nodes. After identifying the deviations marked on the current node, their impact on downstream distribution units is calculated, and it is determined whether there is an overload risk or other markings and deviation types in the downstream node. The parameters that need to be adjusted for upstream and downstream nodes are then extracted. At this time, frequency and load-related parameters are also viewed, and the viewing method is the same as that for voltage-related parameters. Then, the currently extracted parameters that need to be adjusted are used as the optimization targets of the power supply configuration scheme. When one or more optimization targets are to solve local problems, all targets are processed simultaneously to describe the global target.

[0075] After understanding the optimization objectives of the power supply configuration scheme, the local objectives include minimizing the switching delay after adjusting a single path and minimizing the local load balancing deviation. The global objective aims to minimize the total number of deviation types and the number of status indicators across all nodes. Then, using the voltage, frequency, load rate, and switching time configured in the power supply configuration scheme for each node as input values, the least squares method is applied to the local and global objectives sequentially as proportional values. For example, after calculating the deviations of these four values, the four deviations are weighted and summed to combine the objectives corresponding to the four deviations. The weighted values ​​can be determined by assigning weights to the number of power supply configuration schemes used for each of these four values ​​relative to the total number of schemes in historical data. This yields the current relative least squares solution. When the sum of the least squares solutions for each pairwise combination is minimized, it indicates that the local and global objectives are balanced, and the corresponding power supply configuration scheme is the optimal combination of multiple schemes. The corresponding power supply configuration scheme is then output to obtain the data content that needs to be adjusted for the current power supply node.

[0076] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention, which are still covered within the protection scope of the present invention.

Claims

1. A power distribution system performance health state dynamic monitoring system, characterized in that, The method comprises the following steps: The state acquisition module is used to acquire the running information of the power supply equipment under each power supply mode, and the power supply equipment is combined as a power supply distribution network, and the running information comprises voltage, frequency and load rate; The power supply switching module is used to view the switching behavior characteristics of the current power supply node configuration according to the running information of the power supply node, the running time and the switching times of the power supply node; The power supply monitoring module is used to view the running information before and after the power supply mode switching, check the similarity of the power supply mode switching according to the voltage deviation and the load deviation of the running information, and update the running information of each power supply node according to the similarity of the corresponding power supply mode; The switching matching module is used to describe the switching mode of the current power supply node under the power supply mode switching according to the switching behavior characteristics, and describe the switching matching network of each power supply node according to the switching sequence of the switching mode; The state evaluation module is used to generate a plurality of power supply configuration schemes according to the updated running information of each power supply node and the switching matching network, and select the optimal configuration scheme corresponding to each power supply node according to the power supply mode involved in the power supply configuration scheme; The implementation mode of the power supply switching module further comprises: according to the running time and the switching times of the power supply node, the switching frequency and the average switching time of the current power supply node are counted; the switching frequency and the average switching time are subjected to threshold detection to identify the switching behavior characteristics of the current power supply node; When identifying the switching behavior characteristics of the current power supply node, the implementation mode further comprises: based on the threshold detection result of the same power supply node, the adjacent power supply nodes are combined for judgment, the switching behavior characteristics difference value of the adjacent power supply nodes is used to determine the adjacent change combination; the conditional probability of the switching behavior characteristics value of the adjacent change combination on the time axis is used to establish a preset time interval corresponding to the current time point, a plurality of equal-sized comparison intervals are marked in the preset time interval, the change probability of the conditional probability of the switching behavior characteristics value is identified, and the change probability is used as the comparison behavior characteristics combination; based on the part with the maximum similarity in the comparison behavior characteristics combination, the switching behavior characteristics is output.

2. The system according to claim 1, wherein, The state acquisition module further comprises: The power supply equipment is regarded as a power supply node, and the power supply distribution network is formed based on the connection relationship of each power supply equipment; The voltage deviation amplitude, the frequency deviation amplitude and the load distribution proportion of each power supply node are calculated respectively based on the voltage, the frequency and the load rate of each power supply node in the power supply distribution network, and the characteristic values of the running information are obtained; The change characteristic values of the running information at two adjacent time points are judged, and the number of nodes corresponding to each change characteristic value is used to judge the alternation degree of the power supply distribution network under the corresponding power supply mode, and the difference value of the alternation degree at each two adjacent time points is used to mark the power supply equipment.

3. The system of claim 1, wherein the system is configured to monitor the health of the power distribution system. When combined as a power supply distribution network, the implementation mode further comprises: Based on the topological relationship of the power supply equipment, the continuous running time of the power supply equipment under the corresponding current allocation demand is detected, and the running information of any power supply equipment at the start time is used to calculate the running time proportion of each power supply equipment in all power supply equipment in the power supply distribution network, and the power supply equipment with the same running time proportion is connected as adjacent power supply nodes.

4. The system of claim 1, wherein the system is configured to monitor the health of the power distribution system. The implementation mode of the power supply monitoring module comprises: The distribution positions of voltage deviation and load deviation of each power supply node before and after switching in time sequence are taken as target data points, similarity calculation is performed on the instantaneous value of the target data points and the time sequence where the target data points are located, and a similarity matrix corresponding to the target data points is formed; Based on the similarity matrix corresponding to the target data points, the belonging deviation type of the target data points in the predefined deviation type is judged, and the running information of each power supply node is updated according to the interval time of the deviation type of the target data points in the time sequence.

5. The power supply and distribution system performance health condition dynamic monitoring system according to claim 4, wherein, The implementation mode of the belonging deviation type further includes: Based on the similarity matrix corresponding to the target data points, the power supply nodes are clustered according to the similarity of voltage deviation and load deviation, and the corresponding deviation type is labeled to each power supply node in the form of time point according to the clustering cluster set in the historical data for different deviation types.

6. The system of claim 1, wherein the system is configured to monitor the health of the power distribution system. The implementation mode of the switching matching module further includes: The state identifier of the switching behavior feature is taken as the basic condition of each power supply node, the voltage interruption time, the power supply type and the voltage value before and after the voltage interruption time of each power supply node are used to segment the switching behavior feature, and the switching mode corresponding to the switching behavior feature is determined.

7. The system of claim 1, wherein the system is configured to monitor the performance of the power distribution system. The implementation mode of the state evaluation module includes: The updated running information of each power supply node is synchronized to the switching matching network according to the switching order of the corresponding time sequence; The connection path of the switching matching network is identified, and the upstream and downstream nodes of each power supply node are identified according to the connection order in each connection path; The power supply configuration scheme on each connection path is counted, and the power supply configuration scheme is screened according to the upstream and downstream nodes of each connection path; the optimal multi-objective solution after screening is taken as the output optimal configuration scheme.

8. The power supply and distribution system performance health state dynamic monitoring system according to claim 7, wherein, The implementation mode of identifying the upstream and downstream nodes of each power supply node includes: In response to the upstream nodes of the current power supply node, the power supply configuration scheme of each upstream node is configured according to the state change number of the upstream node when pointing to the current power supply node according to the state identifier of the upstream node; In response to the downstream nodes of the current power supply node, the power supply configuration scheme of each downstream node is configured according to the state change number of the downstream node when pointing to the current power supply node according to the state identifier of the downstream node.

Images