A method, system and electronic equipment for monitoring the operation status of a power transformer and distribution device

Abstract

The present application relates to the technical field of monitoring the operation of power transformers and distribution equipment, and specifically to a method, system and electronic equipment for monitoring the operation status of power transformers and distribution equipment, which specifically includes: clustering the power data of all loads at each acquisition moment to obtain each clustering cluster at each acquisition moment, and calculating the load significance of each clustering cluster according to the intra-cluster difference and inter-cluster difference of the clustering cluster at the same acquisition moment; combining the similarity between the clustering clusters at any two adjacent acquisition moments to obtain the load fluctuation of each clustering cluster group; calculating the load change degree at any two adjacent acquisition moments based on the load significance of the redundant clusters and the load fluctuation degree of each clustering cluster group; and adjusting reactive power based on the load change degree. The voltage fluctuation caused by load changes is reduced, the overall stability of the power grid is improved, and the normal operation of the power transformers and distribution equipment is further guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of monitoring the operation status of power transformers and distribution equipment, and in particular to a method, system and electronic equipment for monitoring the operation status of power transformers and distribution equipment. Background Art

[0002] The transformer and distribution device is an indispensable key link in the power system, responsible for converting high-voltage electric energy into low-voltage electric energy suitable for end users, ensuring the effective distribution and use of electric energy. Power factor monitoring at the front end of the transformer and distribution device is the key to ensuring the efficient and stable operation of the power system, allowing real-time understanding of the demand and flow of reactive power, thereby effectively reducing energy loss during transmission and improving the transmission efficiency of the power grid.

[0003] In the power system, the transformer and distribution equipment can be divided into two categories, namely the front-end transformer and distribution equipment and the terminal transformer and distribution equipment. The front-end refers to the high-voltage substation, whose main function is to increase or decrease the voltage, control and protect the power system. The terminal refers to the distribution box, which is responsible for converting high-voltage electricity into low-voltage electricity suitable for home, commercial or industrial use.

[0004] By maintaining an appropriate power factor in a large industrial park, voltage fluctuations can be prevented, power supply quality can be ensured, and energy efficiency standards of various industrial chains in the park can be met. The power factor is affected by reactive power. However, changes in the load at the user terminal will have a direct impact on the monitoring of reactive power at the front end of the transformer and distribution device. When the user load increases, especially when inductive loads such as motors are started, more reactive power will be consumed, resulting in a decrease in the front end power factor. On the contrary, when the load decreases, the reactive power demand decreases, which may increase the power factor. When monitoring reactive power at the transformer and distribution device, the increase or decrease of local loads may cause significant changes in reactive power demand. When monitoring reactive power at the front end of the transformer and distribution device, the existing technology tends to capture the overall trend, but does not respond sufficiently to small-scale or specific regional load fluctuations, resulting in inaccurate estimates of reactive power demand, and thus the ideal power factor cannot be maintained in a short period of time, causing the transformer and distribution device to be overloaded and unable to operate normally.

[0005] In order to prevent the overload of the transformer and distribution device, one of the existing technologies is to try to quickly provide dynamic voltage support after the transformer and distribution device has problems, including installing dynamic reactive compensation devices such as static synchronous compensator STATCOM (STATic synchronous COMpensator) and static var compensator SVC (Static Var Compensator). For example, Huang Xiaoqing et al. proposed a dynamic reactive optimization configuration method considering the characteristics of the power grid (Electric Power Automation Equipment, 2016, Vol. 36, No. 9). On the basis of considering the topological structure and power distribution characteristics of the power grid, the importance index of node voltage stability is used to screen out important nodes for maintaining the transient voltage stability of the system; combined with the weakness ranking of node voltage stability based on the deviation, the candidate installation points of the compensation device are quickly determined; using the improved trajectory sensitivity index, the difference in the improvement effect of the limit removal time of multiple key faults and the improvement effect on the transient voltage stability of the whole system when the compensation device is configured at different locations is analyzed, and the optimal installation point of the dynamic reactive compensation device is selected. However, this method determines the installation node based on historical data, and it is difficult to feedback the changes in the load fluctuation of the power grid in real time, lacking timeliness and flexibility.

[0006] The popularity of new energy vehicles has brought new problems to the safety of electricity use in industrial parks. During the peak charging period, a large number of electric vehicles are connected to the power grid, which will cause a significant increase in load and a decrease in the output voltage of the transformer and distribution device. Some vulnerable nodes even have voltages below the lower limit, threatening the safe operation of the power grid. Gao Shuang et al. proposed a multi-period active and reactive hybrid control strategy for optimizing charging of electric vehicle clusters (Journal of Electric Power Systems and Automation, 2018, Vol. 30, No. 4); CHOIW et al. proposed a method to improve and control charging piles for reactive compensation from a hardware perspective (2016 IEEE Power and Energy Society General Meeting). Yao Lanni et al. proposed to control the electric vehicles connected to the power grid to perform bidirectional support for charging and discharging to strengthen the reactive optimization of the distribution system (Automation of Electric Power Systems, 2022, Vol. 46, No. 6). However, none of the above methods effectively improves the timeliness of monitoring the operating status of transformer and distribution devices and the accuracy of analyzing the impact of short-term load power changes on power grid stability. Summary of the invention

[0007] In order to solve the above technical problems, the purpose of this application is to provide a method, system and electronic equipment for monitoring the operating status of a transformer and distribution device. The technical solutions adopted are as follows:

[0008] In a first aspect, an embodiment of the present application provides a method for monitoring the operating status of a power transformer and a power distribution device, the method comprising the following steps:

[0009] Collect reactive power data of the front section of the transformer and distribution device and power data of each load at the terminal at each moment;

[0010] The power data of all loads at each collection moment are clustered to obtain each cluster at each collection moment, and the load significance of each cluster is calculated according to the intra-cluster difference and inter-cluster difference of the clusters at the same collection moment;

[0011] Based on the similarity between the clusters at any two adjacent collection moments, the cluster groups and redundant clusters at any two adjacent collection moments are obtained; based on the load significance and the differences between the clusters in each cluster group, the load fluctuation of each cluster group is obtained;

[0012] The load change degree between any two adjacent acquisition moments is calculated based on the load significance of redundant clusters and the load fluctuation degree of each cluster group;

[0013] Reactive power is adjusted based on the load change degree.

[0014] In one embodiment, the process of obtaining the load significance of each cluster is as follows:

[0015] Calculate the discreteness of all elements in each cluster at each acquisition moment and the discrimination of each cluster;

[0016] The load significance of each cluster at each collection moment is determined based on the discrete degree and the discrimination degree. The load significance of each cluster at each collection moment is negatively correlated with the discrete degree and positively correlated with the discrimination degree.

[0017] In one embodiment, the process of obtaining the discrimination degree of each cluster is as follows:

[0018] Calculate the mean of all elements in each cluster, and record it as the first mean; calculate the difference between any cluster and the first mean of each cluster remaining except any cluster at the same acquisition time, and record it as the first difference; take the mean of all the first differences of any cluster as the discrimination of any cluster.

[0019] In one embodiment, the process of obtaining each cluster group and redundant clusters is as follows:

[0020] The mean of all elements in each cluster is taken as the centroid of each cluster; at any two adjacent acquisition moments, the distance between any cluster at one acquisition moment and the centroid of each cluster at the other acquisition moment is calculated, and the cluster corresponding to the minimum value of all the distances of any cluster at the other acquisition moment is taken as the cluster that successfully matches the any cluster; wherein, each pair of clusters that have been successfully matched does not participate in the subsequent matching process; the any cluster and the successfully matched cluster are recorded as a cluster group;

[0021] When the number of clusters at two adjacent acquisition moments is inconsistent, the extra clusters that cannot be matched are recorded as redundant clusters.

[0022] In one embodiment, the process of obtaining the load fluctuation of each cluster group is as follows:

[0023] Calculate the difference between the load significance of the two clusters in each cluster group, recorded as the second difference;

[0024] Calculate the difference between the number of elements within the two clusters in each cluster group, and record it as the number difference;

[0025] All loads are numbered, and the set consisting of the load numbers corresponding to all elements in each cluster is recorded as the number set, and the distance measure between the number sets of two clusters in each cluster group is calculated;

[0026] The load fluctuation of each cluster group is determined based on the second difference, the number difference and the distance metric, and the load fluctuation of each cluster group is positively correlated with the second difference, the number difference and the distance metric respectively.

[0027] In one embodiment, the process of acquiring the load change degree at any two adjacent collection moments is as follows:

[0028] Calculate the mean of the load fluctuation of all cluster groups at any two adjacent acquisition moments, recorded as the second mean; for each redundant cluster at any two adjacent acquisition moments, calculate the product of the load significance of the redundant cluster and the number of elements in the cluster, recorded as the first product;

[0029] A fusion value of the first product of all redundant clusters at any two adjacent acquisition moments and the second mean value is used as the load change degree at any two adjacent acquisition moments.

[0030] In one embodiment, the reactive power adjustment based on the load change degree is specifically:

[0031] Based on the load change degree of any two adjacent collection moments, the reactive power change amount of each collection moment except the first collection moment is calculated;

[0032] Calculate the average value of the reactive power change at each collection moment except the first collection moment, record it as the third mean, set the abnormal change threshold, if the difference between the reactive power change monitored at the current moment and the reactive power change at the previous moment is greater than the abnormal change threshold, use reactive power compensation equipment for compensation; otherwise, no adjustment is made.

[0033] In one embodiment, the process of obtaining the reactive power variation is as follows:

[0034] The load change degree between any collection moment and the corresponding previous collection moment is used as a weight, and the product of the weight and the reactive power data at the front end of the transformer and distribution device at any collection moment is calculated as the reactive power change at any collection moment.

[0035] In a second aspect, an embodiment of the present application provides a system for monitoring the operating status of a power transformer and distribution device, including:

[0036] Data acquisition module: collects reactive power data of the front section of the transformer and distribution device and power data of each load at the terminal at each moment;

[0037] Load power analysis module: cluster the power data of all loads at each collection moment to obtain each cluster at each collection moment, and calculate the load significance of each cluster according to the intra-cluster difference and inter-cluster difference of the clusters at the same collection moment;

[0038] Based on the similarity between the clusters at any two adjacent collection moments, the cluster groups and redundant clusters at any two adjacent collection moments are obtained; based on the load significance and the differences between the clusters in each cluster group, the load fluctuation of each cluster group is obtained;

[0039] The load change degree between any two adjacent acquisition moments is calculated based on the load significance of redundant clusters and the load fluctuation degree of each cluster group;

[0040] Reactive power adjustment module: performs reactive power adjustment based on the load change degree.

[0041] In the third aspect, an embodiment of the present application also provides a device for monitoring the operating status of a power transformer and distribution device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of any one of the above-mentioned methods for monitoring the operating status of a power transformer and distribution device are implemented.

[0042] The embodiments of the present application have at least the following beneficial effects:

[0043] The present application proposes a method, system and electronic device for monitoring the operating status of a transformer and distribution device. The method clusters the power data of all loads at each collection moment, calculates the load significance of each clustering cluster according to the intra-cluster difference and inter-cluster difference of the clustering cluster at the same collection moment, reflects the volatility and consistency of load power consumption under the same power consumption characteristics, and evaluates the impact of load power consumption changes on the stability of the power grid; obtains each clustering cluster group and redundant clusters at any two adjacent collection moments based on the similarity between the clustering clusters at any two adjacent collection moments; obtains the load fluctuation of each clustering cluster group based on the load significance and the difference between the clustering clusters in each clustering cluster group, reflects the load power consumption at any two adjacent collection moments. The impact of dynamic changes in patterns on power grid stability; based on the load significance of redundant clusters and the load fluctuation of each cluster group, the load change degree between any two adjacent collection moments is calculated to reflect the short-term dynamic changes in the average power consumption of all terminal loads, thereby improving the accuracy of the analysis of the impact of short-term load power changes on power grid stability; reactive power adjustment is performed based on the load change degree; the timeliness and flexibility of reactive power adjustment caused by load fluctuations are improved, the timeliness of monitoring the operating status of power transformers and distribution devices is improved, and the accuracy of the analysis of the impact of short-term load power changes on power grid stability is improved, so that the normal operation of power transformers and distribution devices is further guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A flowchart of a method for monitoring the operating status of a power transformer and distribution device provided in one embodiment of the present application;

[0046] Figure 2 Schematic diagram of the process of obtaining load significance;

[0047] Figure 3 A schematic diagram of the structure of a system for monitoring the operating status of a power transformer and distribution device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a method, system and electronic device for monitoring the operating status of a transformer and distribution device proposed in the present application, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0049] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0050] The following is a detailed description of a method, system and electronic device for monitoring the operating status of a power transformer and distribution device provided by the present application in conjunction with the accompanying drawings.

[0051] See also Figure 1 , which shows a flowchart of a method for monitoring the operating status of a power transformer and distribution device provided by an embodiment of the present application, the method comprising the following steps:

[0052] Step S1, collecting reactive power data of the front end of the transformer and distribution device and power data of each load at the terminal at each moment, and performing preprocessing.

[0053] The power transformation and distribution device is applicable to various scenarios. This embodiment takes the power transformation and distribution device in an industrial park as an example. The power transformation and distribution device is responsible for distribution between different voltage levels.

[0054] A multifunctional online power quality monitoring device is connected to the front-end power transformer and distribution device, which is responsible for collecting reactive power at the front end of the power grid. This embodiment uses a PQ-F335 power quality monitoring device for reactive power collection. In other embodiments of this application, other types of online power quality monitoring devices can also be used for reactive power collection. A power transformer and distribution monitoring system is connected to the terminal power transformer and distribution device, which is responsible for collecting power data of each load. The power transformer and distribution monitoring system collects power data of the load, which is a known technology, and the specific process will not be repeated.

[0055] Preferably, in one embodiment of the present application, when performing reactive power and load power data collection, the data collection time interval in this embodiment is 1s, and the data collection time length is 24h. As other implementation methods, the implementer can determine the data collection time interval and data collection time length according to actual conditions.

[0056] The reactive power and load power data collected at each moment are normalized to eliminate the influence of the dimension. This embodiment uses the maximum and minimum value normalization method for normalization, and other methods can also be used for normalization in other embodiments of the present application. This application does not limit this.

[0057] Each load connected to the distribution box is numbered, and at each collection moment, the power data of all loads are arranged in ascending order of their numbers to obtain the load power sequence at that collection moment; the reactive power data of all moments collected at the front end of the transformer and distribution device are arranged in ascending order of time, and the obtained sequence is recorded as the reactive power sequence.

[0058] It should be understood that arranging the power data of all loads at each collection moment in ascending order of number to form a load power sequence at each collection moment, and arranging the reactive power data of all collection moments in ascending order of time to form a reactive power sequence, is only one implementation method of the present application. As other implementation methods, the implementer does not need to sort them completely according to the load number or chronological order. The implementer decides according to the actual application situation, and this application does not impose any restrictions.

[0059] Step S2, clustering the power data of all loads at each collection moment to obtain each cluster at each collection moment, and calculating the load significance of each cluster according to the intra-cluster difference and inter-cluster difference of the clusters at the same collection moment.

[0060] In different industrial parks, the types and quantities of power loads vary, and therefore have different power consumption characteristics. For example, the power loads in accommodation areas such as employee dormitories usually have peak power consumption in the morning and evening, and the power size is relatively stable; while the power loads in commercial areas such as supermarkets and convenience stores and the power loads in factory areas usually have peak power consumption during the daytime on weekdays, and may be lower on weekends and holidays, with less stable power size, and both require a larger power supply to support their daily operations. During rush hour, as a large number of electric vehicles are connected to the grid, the power load increases, but the regulation and storage capacity of the distribution system with bidirectional support for charging and discharging also increases accordingly; during the rush hour after work, a large number of electric vehicles are disconnected from the grid, and the regulation and storage capacity of the distribution system with bidirectional support for charging and discharging is reduced accordingly. Loads with different power consumption characteristics have different effects on the stability of the grid.

[0061] (1) The power data of all loads at each collection moment are arranged in ascending order of power size to form a sequence which is recorded as the first power sequence;

[0062] The first power sequence at each acquisition moment is used as the input of the DBSCAN algorithm, where the parameter minPts is set to 2 and the maximum radius is set to the difference between the maximum and minimum values ​​in the first power sequence at the acquisition moment. , the output is the clusters.

[0063] It should be understood that, as an embodiment of the present application, the DBSCAN algorithm is used to cluster the power data in the first power sequence at each acquisition moment. The above only provides a method for clustering the power data in the first power sequence. As other implementation methods, on the premise of achieving the purpose of clustering the power data in the first power sequence, other clustering algorithms in the prior art can be used to cluster the power data in the first power sequence.

[0064] All loads in the same cluster are loads with similar power consumption characteristics at that moment.

[0065] (2) Calculate the degree of dispersion of all elements in each cluster at each collection time.

[0066] Preferably, as an embodiment of the present application, the discrete degree may be the standard deviation of all elements in each cluster at each acquisition moment;

[0067] In other embodiments of the present application, the degree of discreteness may be the variance of all elements in each cluster at each acquisition moment.

[0068] (3) Calculate the discrimination of each cluster at each collection time:

[0069] The mean of all elements in each cluster at each acquisition moment is calculated, recorded as the first mean; at each acquisition moment, the difference between the first mean of any cluster and each remaining cluster except the any cluster is calculated, recorded as the first difference; the mean of all the first differences of the any cluster is taken as the discrimination of the any cluster.

[0070] Preferably, as an embodiment of the present application, the first difference may be an absolute value of a difference between the first mean of the any cluster and each remaining cluster except the any cluster;

[0071] In other embodiments of the present application, the first difference may be the square of the difference between the first mean of the any cluster and each remaining cluster except the any cluster.

[0072] The discrimination degree of each cluster reflects the degree of difference between the cluster and the other clusters at the same time.

[0073] (4) Determine the load significance of each cluster at each collection time based on the discrete degree and the distinction of each cluster at each collection time:

[0074] The load significance of each cluster is negatively correlated with the degree of dispersion of each cluster, and positively correlated with the degree of discrimination of each cluster.

[0075] It should be noted that the positive correlation relationship described in this application means that the independent variable increases as the dependent variable increases, and the independent variable decreases as the dependent variable decreases; the negative correlation relationship means that the independent variable decreases as the dependent variable increases, and the independent variable increases as the dependent variable decreases. The specific positive correlation and negative correlation can be determined according to actual conditions during the application process, and this application does not impose any special restrictions.

[0076] Preferably, as an embodiment of the present application, the expression of the load significance of each cluster can be: , where is the load significance of cluster i, is the discrete degree of cluster i, is the discrimination degree of cluster i;

[0077] In other embodiments of the present application, the expression of the load significance of each cluster may be: , where is an exponential function with the natural number e as base.

[0078] The discrete degree of the clustering cluster measures the volatility of power consumption in loads with similar power consumption characteristics. The smaller the value, the smaller the power difference between the loads in the power consumption characteristics, and the smaller the range of load power values ​​in the clustering cluster, the smaller the difference in power consumption patterns of the loads in the clustering cluster, the greater the consistency of the power consumption patterns of the loads in the clustering cluster, and the greater the load significance of the clustering cluster;

[0079] The discrimination degree of a cluster expresses the degree of discrimination between the cluster and other clusters at the same time. The larger the value, the more significant the difference between the load power in the cluster and other clusters at the same time. The more unique the power consumption pattern or demand of the load in the cluster, the greater the load significance of the cluster.

[0080] The load significance of a cluster is used to evaluate the uniqueness and consistency of the power consumption pattern represented by the load of the selected cluster compared with the power consumption patterns represented by the loads of other clusters. The larger the value, the more likely the loads in the cluster are to be loads with the same power consumption pattern.

[0081] Step S3, based on the load significance and the similarity between the clusters at any two adjacent collection moments, analyze the dynamic changes of the load power consumption mode at any two adjacent collection moments, and calculate the load fluctuation of each cluster group at any two adjacent collection moments.

[0082] In the power system, the stability of the power consumption pattern is crucial to the impact of the reactive power of the power distribution system. However, due to the dynamic changes of various external factors, such as weather conditions, time, special events, and economic activity levels, the power consumption pattern may show significant differences at different times, resulting in changes in the reactive power of the power grid, affecting the stability of the power grid and the power supply quality.

[0083] (1) The mean of all elements in each cluster is taken as the centroid of each cluster. It should be noted that in other embodiments of the present application, the centroid of each cluster may also be the median of all elements in each cluster, which is not specifically limited in this embodiment.

[0084] (2) Taking the j-1th collection time and the j-th collection time as examples, calculate the Euclidean distance between the centroids of each cluster at the j-1th collection time and the centroids of each cluster at the j-1th collection time. It should be noted that for the measurement of the distance between the centroids of clusters, this application only provides a distance measurement method. On the premise of achieving the purpose of measuring the distance between the centroids, the implementer may also select other distance measurement methods.

[0085] (3) Obtain each cluster group at any two adjacent acquisition moments based on the Euclidean distance between the centroids:

[0086] In any two adjacent acquisition moments, obtain any clustering cluster at any acquisition moment and the matching clustering cluster at another acquisition moment, specifically: obtain the Euclidean distance between any clustering cluster and the centroid of each clustering cluster at another acquisition moment, and take the clustering cluster corresponding to the minimum value of all the Euclidean distances of any clustering cluster at another acquisition moment as the clustering cluster that successfully matches the any clustering cluster; in the matching process, each pair of clustering clusters that have been successfully matched does not participate in the subsequent matching process, for example, clustering cluster k at the j-1th acquisition moment and clustering cluster q at the jth acquisition moment are first successfully matched, then clustering cluster k and clustering cluster q will no longer participate in the subsequent matching process of other clustering clusters. If the Euclidean distance between any clustering cluster and the centroid of multiple clustering clusters at another acquisition moment is the same, then select the clustering cluster with the closest number to any clustering cluster for matching. Each group of matched clustering clusters at two adjacent acquisition moments is recorded as each clustering cluster group. When the number of clusters at two adjacent collection moments is inconsistent, the extra clusters that cannot be matched are recorded as redundant clusters, and these clusters are regarded as newly added power consumption patterns and are not matched.

[0087] (4) Analyze the differences between the two clusters in each cluster group:

[0088] Calculate the difference between the load significance of the two clusters in each cluster group, recorded as the second difference;

[0089] Calculate the difference between the number of elements within the two clusters in each cluster group, and record it as the number difference;

[0090] The set consisting of the load numbers corresponding to all elements in each cluster is recorded as a number set, the number sets of the two clusters in each cluster group are used as the input of the Jaccard similarity algorithm, and the Jaccard distance between the number sets of the two clusters in each cluster group is output as the first difference degree between the number sets of the two clusters in each cluster group.

[0091] Preferably, as an embodiment of the present application, the second difference may be the absolute value of the difference between the load significance of two clusters in each cluster group, and the number difference may be the absolute value of the difference between the number of elements in two clusters in each cluster group;

[0092] In other embodiments of the present application, the second difference may be the square difference between the load significance of two clusters in each cluster group, and the number difference may be the square difference between the number of elements within the two clusters in each cluster group.

[0093] It should be noted that the present application only provides a method for calculating the degree of difference between the number sets of two cluster clusters in each cluster cluster group. There are many existing methods for calculating the degree of difference between sets. Implementers can also use other similarity algorithms to obtain the degree of difference between the number sets of two cluster clusters in each cluster cluster group. The present application does not make any specific restrictions.

[0094] The second difference and the number difference characterize the change characteristics between the power consumption patterns of the loads of the two clusters in the cluster group, and reflect the dynamic change of the power consumption pattern over time. The larger the difference, the more significant the change in the power consumption pattern.

[0095] The first difference degree indicates the degree of change in the power consumption mode, that is, the difference in the power consumption mode of the load at different times. The larger the value of the first difference degree of the cluster group, the lower the similarity between the clusters and the greater the change in the power consumption mode.

[0096] (5) Based on the second difference, the number difference and the first difference degree, calculate the load fluctuation of each cluster group:

[0097] The load fluctuation degree of each cluster group is positively correlated with the second difference, the number difference and the first difference degree of each cluster group.

[0098] Preferably, as an embodiment of the present application, the load fluctuation degree of each cluster group may be the product of the second difference, the number difference and the first difference degree of each cluster group;

[0099] In other embodiments of the present application, the load fluctuation degree of each cluster group may be the sum of the second difference, the number difference and the first difference degree of each cluster group.

[0100] The greater the second difference and the number difference of the cluster groups, the more significant the change in the power consumption pattern, the greater the contribution to the load fluctuation between two adjacent collection moments, and the greater the instability of the power consumption pattern.

[0101] The greater the first difference degree of the cluster groups, the greater the contribution to the load fluctuation of each cluster group at two adjacent collection moments.

[0102] The load fluctuation of each cluster group at two adjacent collection moments describes the dynamic change of a power consumption pattern in the power grid between two adjacent moments. The larger the value, the greater the short-term dynamic change of the power consumption pattern.

[0103] Step S4: Calculate the load change degree of the overall load power at any two adjacent acquisition moments based on the load fluctuation degree of each cluster group.

[0104] Calculate the mean of the load fluctuation of all cluster groups at any two adjacent acquisition moments, recorded as the second mean;

[0105] For each redundant cluster at any two adjacent acquisition moments, calculate the product of the redundant cluster's load significance and the number of elements in the cluster, and record it as the first product;

[0106] A fusion value of the first product of all redundant clusters at any two adjacent acquisition moments and the second mean value is used as the load change degree of the overall load power at any two adjacent acquisition moments.

[0107] It should be understood that the fusion is to combine multiple variables, which may be specifically addition, multiplication, etc.; the specific fusion method can be determined according to actual conditions during the application process, and this application does not impose any special restrictions.

[0108] Preferably, as an embodiment of the present application, the load change degree of the overall load power at any two adjacent acquisition moments may be the sum of the first products of all the redundant clusters and the sum of the second mean values;

[0109] In other embodiments of the present application, the load change degree of the overall load power at any two adjacent acquisition moments may be the product of the mean of the first products of all the redundant clusters and the second mean.

[0110] The first product of the redundant clusters represents a quantitative measure of the impact of the redundant clusters on the stability of the power grid, reflecting the potential impact of the new power consumption mode on the stability of the power grid. The larger the value, the more significant the characteristics of the new power consumption mode, and the greater the contribution to the load change of the overall power of the load between two adjacent moments. The larger the sum of the first product and the second mean, the greater the change in the average power consumption of all terminal loads.

[0111] Step S5: adjusting the reactive power based on the load change degree of the overall load power at any two adjacent data collection moments.

[0112] The load change degree of the overall load power at any two adjacent acquisition moments is normalized so that the load change degree of the overall load power under all time changes has the same scale, which is convenient for comparing the relative sizes between different time changes.

[0113] Changes in the power consumption patterns of all terminal loads connected to the transformer and distribution device directly affect the load characteristics of the power grid, and thus affect the demand and distribution of reactive power. Reactive power is a key factor in maintaining voltage stability and transmission efficiency in the power system, and its changes are closely related to changes in load power.

[0114] In the power system, the increase or decrease of load will cause the change of current phase, thus affecting the generation of reactive power. For example, when the load increases, it may cause the current to lag behind the voltage, increasing the demand for reactive power; conversely, when the load decreases, the demand for reactive power may decrease. Therefore, by analyzing the changes in the power consumption pattern, the reactive power generated by the change of terminal load can be monitored, and the corresponding grid adjustment can be carried out accordingly.

[0115] Based on the load change of the overall power between any two adjacent data collection moments, the reactive power change at each collection moment except the first collection moment is calculated:

[0116] Taking the j-1th collection moment and the jth collection moment as an example, the load change degree of the overall power between the j-1th collection moment and the jth collection moment is used as the weight, and multiplied by the reactive power monitored at the front end of the transformer and distribution device at the jth collection moment. The weighted result obtained is the change in reactive power caused by the change in terminal load. The same method is used to calculate the weighted results of the reactive power at all collection moments, and the calculated weighted result is used as the reactive power change at the jth collection moment.

[0117] Calculate the average value of the reactive power change at all remaining collection moments except the first collection moment, record it as the third mean value, set the abnormal change threshold value to one fifth of the third mean value, it should be noted that the implementer of the abnormal change threshold value can set it by himself, and this application does not make specific restrictions. The abnormal change threshold value is used to detect significant changes in reactive power. If the difference between the reactive power change amount monitored at the current moment and the reactive power change amount at the previous moment is greater than the abnormal change threshold value, it indicates that the reactive power demand at the front end of the transformer and distribution has changed significantly due to a substantial increase or decrease in the load. According to the analysis results, adjust the reactive power to return it to the normal range. Reactive compensation equipment (such as SVC, STATCOM, etc.) can be used for compensation, and electric vehicles connected to the power grid can also be used to compensate for the discharge of the power grid. After the compensation is implemented, continue to monitor the changes in reactive power to ensure the compensation effect and maintain the normal operation of the transformer and distribution device. Otherwise, no adjustment is required. Among them, using reactive compensation equipment for compensation is a prior art, and the specific method will not be repeated.

[0118] The schematic diagram of the load saliency acquisition process is as follows: Figure 2 shown.

[0119] See also Figure 3 , Figure 3 Schematic diagram of a system for monitoring the operation status of a power transformer and distribution device provided in an embodiment of the present application. Figure 3 , a transformer and distribution device operation status monitoring system comprises:

[0120] Data acquisition module: collects reactive power data of the front section of the transformer and distribution device and power data of each load at the terminal at each moment;

[0121] Load power analysis module: cluster the power data of all loads at each collection moment to obtain each cluster at each collection moment, and calculate the load significance of each cluster according to the intra-cluster difference and inter-cluster difference of the clusters at the same collection moment;

[0122] Based on the similarity between the clusters at any two adjacent collection moments, the cluster groups and redundant clusters at any two adjacent collection moments are obtained; based on the load significance and the differences between the clusters in each cluster group, the load fluctuation of each cluster group is obtained;

[0123] The load change degree between any two adjacent acquisition moments is calculated based on the load significance of redundant clusters and the load fluctuation degree of each cluster group;

[0124] Reactive power adjustment module: Reactive power adjustment is performed based on the load change degree. In one embodiment of the present application, the reactive power adjustment module has a bidirectional support function for electric vehicle charging and discharging, and can use electric vehicles connected to the power grid to discharge the power grid to achieve reactive power adjustment. In another embodiment of the present application, the reactive power adjustment module includes a mobile reactive compensation device (such as a mobile SVC, STATCOM, etc.), and the reactive power adjustment module can issue instructions to dispatch the mobile reactive compensation device to move and connect to the power transformer and distribution device that needs to be compensated to achieve reactive power adjustment.

[0125] Based on the same inventive concept as the above method, an embodiment of the present application also provides a device for monitoring the operating status of a power transformer and distribution device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above methods for monitoring the operating status of a power transformer and distribution device when executing the computer program.

[0126] In summary, the embodiment of the present application provides a method for monitoring the operating status of a transformer and distribution device, which clusters the power data of all loads at each collection moment, calculates the load significance of each clustering cluster according to the intra-cluster difference and inter-cluster difference of the clustering cluster at the same collection moment, reflects the volatility and consistency of load power consumption under the same power consumption characteristics, and evaluates the impact of load power consumption changes on the stability of the power grid; based on the similarity between the clustering clusters at any two adjacent collection moments, obtains each clustering cluster group and redundant clusters at any two adjacent collection moments; based on the load significance and the difference between the clustering clusters in each clustering cluster group, obtains the load fluctuation of each clustering cluster group, reflecting the load consumption at any two adjacent collection moments. The impact of dynamic changes in power patterns on grid stability; based on the load significance of redundant clusters and the load fluctuation of each cluster group, the load change degree between any two adjacent collection moments is calculated to reflect the short-term dynamic changes in the average power consumption of all terminal loads, thereby improving the accuracy of the analysis of the impact of short-term load power changes on grid stability; reactive power adjustment is performed based on the load change degree; the timeliness and flexibility of reactive power adjustment caused by load fluctuations are improved, the timeliness of monitoring the operating status of power transformers and distribution devices is improved, and the accuracy of the analysis of the impact of short-term load power changes on grid stability is improved, so that the normal operation of power transformers and distribution devices is further guaranteed.

[0127] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. The above-mentioned specific embodiments of the present application are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0128] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0129] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for monitoring the operating status of a power transformer and distribution device, characterized in that: The method comprises the following steps: Collect reactive power at the front end of the transformer and distribution device and power data of each load at the terminal at each moment; The power data of all loads at each collection moment are clustered to obtain each cluster at each collection moment, and the load significance of each cluster is calculated according to the intra-cluster difference and inter-cluster difference of the clusters at the same collection moment; Based on the similarity between the clusters at any two adjacent collection moments, the cluster groups and redundant clusters at any two adjacent collection moments are obtained; based on the load significance and the differences between the clusters in each cluster group, the load fluctuation of each cluster group is obtained; The load change degree between any two adjacent acquisition moments is calculated based on the load significance of redundant clusters and the load fluctuation degree of each cluster group; Perform reactive power adjustment based on the load change degree; The process of obtaining the load significance of each cluster is as follows: Calculate the discreteness of all elements in each cluster at each acquisition moment and the discrimination of each cluster; Determine the load significance of each cluster at each collection moment based on the discreteness and the discrimination, wherein the load significance of each cluster at each collection moment is negatively correlated with the discreteness and positively correlated with the discrimination; The process of obtaining each cluster group and redundant clusters is as follows: The mean of all elements in each cluster is taken as the centroid of each cluster; at any two adjacent acquisition moments, the distance between any cluster at one acquisition moment and the centroid of each cluster at the other acquisition moment is calculated, and the cluster corresponding to the minimum value of all the distances of any cluster at the other acquisition moment is taken as the cluster that successfully matches the any cluster; wherein, each pair of clusters that have been successfully matched does not participate in the subsequent matching process; the any cluster and the successfully matched cluster are recorded as a cluster group; When the number of clusters at two adjacent acquisition moments is inconsistent, the extra clusters that cannot be matched are recorded as redundant clusters; The process of obtaining the load fluctuation of each cluster group is as follows: Calculate the difference between the load significance of the two clusters in each cluster group, recorded as the second difference; Calculate the difference between the number of elements within the two clusters in each cluster group, and record it as the number difference; All loads are numbered, and the set consisting of the load numbers corresponding to all elements in each cluster is recorded as the number set, and the distance measure between the number sets of two clusters in each cluster group is calculated; Determine the load fluctuation of each cluster group based on the second difference, the number difference and the distance metric, wherein the load fluctuation of each cluster group is positively correlated with the second difference, the number difference and the distance metric respectively; The process of obtaining the load change degree at any two adjacent acquisition moments is as follows: Calculate the mean of the load fluctuation of all cluster groups at any two adjacent acquisition moments, recorded as the second mean; for each redundant cluster at any two adjacent acquisition moments, calculate the product of the load significance of the redundant cluster and the number of elements in the cluster, recorded as the first product; Taking the fusion value of the first product of all redundant clusters at any two adjacent acquisition moments and the second mean as the load change degree at any two adjacent acquisition moments; The reactive power adjustment based on the load change degree is specifically: Based on the load change degree of any two adjacent collection moments, the reactive power change amount of each collection moment except the first collection moment is calculated; Set an abnormal change threshold. If the difference between the reactive power change monitored at the current moment and the reactive power change at the previous moment is greater than the abnormal change threshold, use reactive power compensation equipment for compensation; otherwise, no adjustment is made.

2. A method for monitoring the operating status of a power transformer and distribution device according to claim 1, characterized in that: The process of obtaining the discrimination degree of each cluster is as follows: Calculate the mean of all elements in each cluster, and record it as the first mean; calculate the difference between any cluster and the first mean of each cluster remaining except any cluster at the same acquisition time, and record it as the first difference; take the mean of all the first differences of any cluster as the discrimination of any cluster.

3. A method for monitoring the operating status of a power transformer and distribution device according to claim 1, characterized in that: The process of obtaining the reactive power variation is as follows: The load change degree between any collection moment and the corresponding previous collection moment is used as a weight, and the product of the weight and the reactive power data at the front end of the transformer and distribution device at any collection moment is calculated as the reactive power change at any collection moment.

4. A system for monitoring the operation status of a transformer and distribution device, implementing the method as claimed in claim 1, characterized in that: The system comprises: Data acquisition module: collects reactive power at the front end of the transformer and distribution device and power data of each load at the terminal at each moment; Load power analysis module: cluster the power data of all loads at each collection moment to obtain each cluster at each collection moment, and calculate the load significance of each cluster according to the intra-cluster difference and inter-cluster difference of the clusters at the same collection moment; Based on the similarity between the clusters at any two adjacent collection moments, the cluster groups and redundant clusters at any two adjacent collection moments are obtained; based on the load significance and the differences between the clusters in each cluster group, the load fluctuation of each cluster group is obtained; The load change degree between any two adjacent acquisition moments is calculated based on the load significance of redundant clusters and the load fluctuation degree of each cluster group; Reactive power adjustment module: performs reactive power adjustment based on the load change degree.

5. A device for monitoring the operating status of a transformer and distribution device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method for monitoring the operating status of a power transformer and distribution device as described in any one of claims 1 to 3 are implemented.

Images