Device and method for evaluating power supply reliability of power grid with source, network, load, storage and power distribution

Abstract

A kind of source network load storage distribution network power supply reliability evaluation device and method, belong to power supply reliability evaluation technical field, it can achieve active improvement and accurate control during load end operation by multidimensional analysis of instant operation parameter value and past power supply reference value, improve the reliability and coordination of distribution network to load end power supply performance.Due to the automatic differential analysis operation, the change of power supply scene and the effect of load end reliability on the power supply process of distribution network to load end are efficiently introduced, the small abnormal value in the value is executed to explore, so that the power supply of distribution network to load end can execute feedback and configuration control mode to small abnormal value, guarantee the continuous improvement of power supply performance, according to the response learning and control mode improvement of past power supply value, match power supply change and continuously improve and improve the power supply process of distribution network to load end, improve the power supply performance of distribution network to load end.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power supply reliability evaluation, and particularly relates to a power supply reliability evaluation device and method for a source-grid-load-storage power distribution network. BACKGROUND

[0002] As an important public infrastructure, the power distribution network plays an important role in ensuring power supply, supporting economic and social development, and serving the improvement of people's livelihood. With the advancement of new power system construction, the power distribution network is gradually transforming from a simple power network that accepts and distributes power to users to a power network that aggregates and interacts with source-grid-load-storage and flexibly couples with the upper-level power grid, and its functions in promoting the local consumption of distributed power sources and carrying new types of loads are becoming increasingly prominent.

[0003] For the control of the power supply of the source-grid-load-storage power distribution network, a technical solution is often used to achieve the control of the power supply of the source-grid-load-storage power distribution network, which includes collecting power usage data at the load end and constructing a model to perform load power prediction Pload.

[0004] However, on the other hand, the current power distribution network generally relies on pre-set constant parameters for control of the load end, lacks the function of dynamic configuration and automatic improvement, and thus often has the defect that the power supply mode of the power distribution network does not match the load end during actual power supply. Moreover, under the influence of the use scenarios of the load end and the degradation or wear of the load end elements, the control mode relying on pre-set constant parameters cannot easily meet the high reliability requirements of power supply. SUMMARY

[0005] To solve the defects in the prior art, the application provides a power supply reliability evaluation device and method for a source-grid-load-storage power distribution network, which is suitable for performing abnormal detection on the target operation parameter value during the operation of the load end, can find hidden abnormalities in the value, and can perform abnormal detection on the current collected value in combination with the past power supply values. According to the target power supply value, a number of past power supply reference values with reference are selected from the information table, and the load end is subjected to power supply during the same target power supply value. The error between the specific parameters brought about by the various elements and the principle parameters sent in is often subjected to slight changes in the principle parameters sent in.

[0006] The application uses the following technical solutions.

[0007] A power supply reliability evaluation method for a source-grid-load-storage power distribution network, comprising:

[0008] The power utilization data of the load end is collected, and a model is constructed to perform load power prediction Pload;

[0009] The power supply reliability evaluation method of the source grid-load-storage-distribution grid further includes:

[0010] Step 1: obtaining a target power supply value of the load end, a target operation parameter value and a power supply scenario value 1, and constructing an obtained leading parameter vector 1 and a detected parameter vector 1 according to the operation parameter value;

[0011] Step 2: extracting a plurality of clusters of past power supply reference values from the power supply value for registering the power supply value of the distribution grid to the load end according to the target power supply value, and constructing an obtained power supply reference value group, and extracting corresponding leading parameter vector 2 and detected parameter vector 2 of each cluster of past power supply reference values;

[0012] Step 3: performing differential analysis 1 on the target operation parameter value and the plurality of clusters of past power supply reference values, extracting a parameter difference vector 1 of the target operation parameter value, and a parameter difference vector 2 of each cluster of past power supply reference values;

[0013] Step 4: performing differential analysis 2 on the plurality of clusters of past power supply reference values according to the target operation parameter value, and extracting a corresponding detection parameter difference vector between the target operation parameter value and the plurality of clusters of past power supply reference values;

[0014] Step 5: identifying a difference reference coefficient of each cluster of past power supply reference values according to the parameter difference vector 1 and the plurality of parameter difference vectors 2, forming an abnormal detection value of the target operation parameter value according to the plurality of detection parameter difference vectors and the difference reference coefficient of each cluster of past power supply reference values, and configuring a control mode of the load end according to the abnormal detection value.

[0015] Further, in step 1, the target power supply value is used to represent the power supply specification value of the distribution grid for the present time power supply of the load end, the target power supply value includes the type of device of the load end, the power capacity of the load end, the rated parameter of the load end, and the control specification value of the distribution grid to the load end, the target operation parameter value includes an operation leading parameter 1 and an operation detection parameter 1, which respectively represent the parameter sent to the load end by the actual distribution grid during power supply and the working value of the load end obtained by detection, the power supply scenario value 1 includes scenario parameters of the load end during use, and value extraction is performed on the operation leading parameter 1 and the operation detection parameter 1 to obtain the leading parameter vector 1 and the detection parameter vector 1, and each value in the parameter vector represents a parameter.

[0016] Further, in step 2, the past power supply values of the power supply mode of the power distribution network to the load end are registered in the power supply value registration unit for registering the power supply mode of the power distribution network to the load end, and the past power supply values of the same purpose power supply value are selected from the past power supply values, and the past power supply values without abnormality during the power supply of the power distribution network to the load end are selected, so as to obtain a plurality of past power supply reference values and form a power supply reference value group.

[0017] Further, in step 2, the past power supply values of the power supply mode of the power distribution network to the load end are registered in the power supply value registration unit for registering the power supply mode of the power distribution network to the load end, and the past power supply values of the same purpose power supply value are selected from the past power supply values, and the past power supply values without abnormality during the power supply of the power distribution network to the load end are selected, so as to obtain a plurality of past power supply reference values and form a power supply reference value group.

[0018] Further, in step 3, the difference between the analysis lead-in parameter vector and the detection parameter vector is analyzed, and the difference between each parameter of the operation lead-in parameter vector and the detection parameter vector is calculated, and the difference is combined to form a parameter difference vector of the purpose operation parameter value. The difference between each parameter of the lead-in parameter vector and the detection parameter vector of each past power supply reference value is calculated, and the difference is combined to form a parameter difference vector of each past power supply reference value.

[0019] Further, in step 4, the difference between the detection parameter vector corresponding to the plurality of past power supply reference values and the detection parameter vector corresponding to the plurality of past power supply reference values is analyzed, and the difference between the lead-in parameter vector corresponding to the plurality of past power supply reference values and the lead-in parameter vector corresponding to the plurality of past power supply reference values is taken as a calibration reference, and the detection parameter difference vector corresponding to the plurality of past power supply reference values is obtained.

[0020] Further, in step 5, it is determined whether the abnormal detection value is lower than the difference threshold set in advance, and if so, the configuration of the control mode of the load end is not performed.

[0021] On the contrary, the purpose power supply value and the power supply scene value of the load end are sent to the control mode forming mode to form the operation parameter reference value corresponding to the load end, and the control mode of the load end is configured according to the operation parameter reference value.

[0022] Further, in step 4, a method for performing differential analysis on a plurality of clusters of past power supply reference values according to the target operation parameter value, extracting a respective detection parameter differential vector between the target operation parameter value and the plurality of clusters of past power supply reference values, includes:

[0023] respectively calculating the difference between each parameter of the detection parameter vector one and the plurality of detection parameter vectors two, and combining the differences to obtain a plurality of detection differential vectors, respectively calculating the difference between each parameter of the guide-in parameter vector one and the plurality of guide-in parameter vectors two, and combining the differences to obtain a plurality of guide-in differential vectors.

[0024] Further, in step 5, a method for identifying the differential reference coefficient of each cluster of past power supply reference values according to the parameter differential vector one and the plurality of parameter differential vectors two, includes:

[0025] identifying the scene action coefficient of each cluster of past power supply reference values according to the power supply scene value one of the load end and the corresponding power supply scene value two of each cluster of past power supply reference values;

[0026] performing approximation analysis on the plurality of parameter differential vectors two according to the parameter differential vector one, and identifying the differential correlation coefficient of each cluster of past power supply reference values according to the approximation degree between the parameter differential vector one and the plurality of parameter differential vectors two;

[0027] After calculating the scene action coefficient and the differential correlation coefficient of each cluster of past power supply reference values, identifying the differential reference coefficient of each cluster of past power supply reference values according to the scene action coefficient and the differential correlation coefficient, as shown in the following equation:

[0028]

[0029] In the equation, Esf j is the differential reference coefficient of the jth cluster of past power supply reference values, Fp j is the scene action coefficient of the jth cluster of past power supply reference values, St j is the differential correlation coefficient of the jth cluster of past power supply reference values, are the importance parameters of the scene action coefficient and the differential correlation coefficient respectively.

[0030] Further, in step 5, the calculation method of the scene action coefficient includes: respectively calculating the KL divergence between the power supply scene value one of the load end and the corresponding power supply scene value two of each cluster of past power supply reference values, and performing standardization processing on the plurality of KL divergences using the Z-score method.

[0031] Further, in step 5, the operation method of the correlation coefficient is distinguished, including, respectively operating the L2 norm between the parameter difference vector one and the plurality of cluster parameter difference vector two, to evaluate the correlation between the parameter difference vectors, and using the Z-score method to perform standardization processing on the plurality of L2 norms, to obtain the difference correlation coefficient of the past power supply reference value of each cluster.

[0032] Further, in step 5, the method for forming the abnormal detection value of the operation parameter value according to the plurality of cluster detection parameter difference vectors and the difference reference coefficient of the past power supply reference value of each cluster, including:

[0033] According to the respective detection parameter difference vector between the operation parameter value and the plurality of past power supply reference values, the difference reference value of the past power supply reference value corresponding to the operation parameter value is determined;

[0034] According to the difference reference coefficient of the past power supply reference value of each cluster, the difference reference value of the past power supply reference value corresponding to the operation parameter value is calibrated, to obtain the target difference value of the operation parameter value corresponding to the past power supply reference value of each cluster in the power supply reference value group, and the operation equation of the target difference value is:

[0035] H k =ν k W k

[0036] In the equation, H k is the target difference value of the kth cluster past power supply reference value, W k is the difference reference value of the kth cluster past power supply reference value, ν k is the difference reference coefficient of the kth cluster past power supply reference value set in advance.

[0037] Further, in step 5, the variance of the plurality of cluster detection parameter difference vectors is calculated, and the variance of the plurality of cluster detection parameter difference vectors is taken as the difference reference value of the past power supply reference value corresponding to the operation parameter value.

[0038] Further, in the face of the control mode forming mode, further including:

[0039] The control mode forming mode is obtained through sample learning, and the sample contains a plurality of cluster operation parameter values, and the target power supply value and the power supply scene value corresponding to each cluster operation parameter value, and the specific steps are as follows:

[0040] The restricted Boltzmann machine is used to perform the construction of the control mode formation mode, to master the operation parameter values of a plurality of clusters in the sample, and the relationship between the target power supply values and the power supply scene values corresponding to the operation parameter values of each cluster, during the mode learning, the target power supply values and the power supply scene values corresponding to the operation parameter values of each cluster are used as the input values of the control mode formation mode to input the restricted Boltzmann machine, and a plurality of cluster operation parameter values are used as the learning target of the control mode formation mode, and the control mode formation mode is learned.

[0041] A source grid load storage power supply reliability evaluation device, comprising:

[0042] The construction module is used for obtaining target power supply values, target operation parameter values and power supply scene values of a load end, and constructing input parameter vector one and detection parameter vector one according to the operation parameter values;

[0043] The extraction module is used for extracting a plurality of cluster past power supply reference values from the power supply values for registering the power supply of the power distribution network to the load end according to the target power supply values, and constructing a power supply reference value group, and extracting input parameter vector two and detection parameter vector two corresponding to each cluster past power supply reference value;

[0044] The analysis module one is used for performing differential analysis one on the target operation parameter values and a plurality of cluster past power supply reference values, extracting parameter difference vector one of the target operation parameter values, and parameter difference vector two of each cluster past power supply reference value;

[0045] The analysis module two is used for performing differential analysis two on a plurality of cluster past power supply reference values according to the target operation parameter values, and extracting detection parameter difference vector corresponding to the target operation parameter values and a plurality of cluster past power supply reference values respectively;

[0046] The control module is used for identifying the difference reference coefficients of each cluster past power supply reference value according to the parameter difference vector one and a plurality of cluster parameter difference vectors two, forming the abnormal detection value of the target operation parameter values according to a plurality of cluster detection parameter difference vectors and the difference reference coefficients of each cluster past power supply reference value, and configuring the control mode of the load end according to the abnormal detection value.

[0047] The beneficial effects of the present application are that, compared with the prior art, the technical effects of the present application include:

[0048] Through multi-dimensional analysis of the instantaneous operation parameter value and the past power supply reference value, active improvement and accurate control during the operation of the load end can be achieved, and the reliability and coordination of the power distribution network to the power supply performance of the load end are improved. Through automatic differential analysis operation, the changes in the power supply scene and the reliability of the load end itself are efficiently introduced into the power distribution network to the power supply process of the load end, and the small abnormal values in the values are explored, so that the power distribution network to the load end can perform feedback and configuration control mode on small abnormal values, and the continuous improvement of the power supply performance is ensured. According to the response learning and control mode improvement of the past power supply value, the power supply changes can be matched and the power distribution network to the load end power supply process can be continuously improved and improved, and the power supply performance of the power distribution network to the load end is improved. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 is part of the flowchart of the power supply reliability evaluation method of the power supply network with source, network, load and storage in the present application;

[0050] Figure 2 is part of the structure schematic diagram of the power supply reliability evaluation device of the power supply network with source, network, load and storage in the present application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages of the present application more clear, the technical scheme of the present application will be expressed clearly and completely in combination with the drawings in the embodiments of the present application. The embodiments expressed in the present application are only some embodiments of the present application, not all embodiments. According to the spirit of the present application, other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0052] As shown in Figure 1 , the power supply reliability evaluation method of the power supply network with source, network, load and storage in the present application comprises:

[0053] The power utilization data of the load end are collected, and a model is constructed to perform load power prediction Pload;

[0054] The power supply reliability evaluation method of the power supply network with source, network, load and storage further comprises:

[0055] Step 1: obtain the target power supply value, the target operation parameter value and the power supply scene value of the load end, and obtain the guide parameter vector one and the detection parameter vector one according to the operation parameter value;

[0056] In the preferred but non-limiting embodiment of the present application, in step 1, the target power supply value is used to represent the power supply specification value of the power distribution network for the present power supply to the load end, and the target power supply value includes the types of devices at the load end (such as the types of electrical devices including refrigerators, washing machines, and motors), the power consumption capacity of the load end, the rated parameters of the load end (such as the rated power, the rated voltage, or the rated current), the control specification values of the power distribution network for the load end (such as the reasonable range values of the output power, the output voltage, or the output current of the transformer or the transformer station of the power distribution network for the power supply to the load end), and the like. The target operation parameter value includes an operation leading-in parameter and an operation detection parameter, which respectively represent the parameters (such as the power value, the voltage value, or the current value) sent by the power distribution network to the load end during the power supply of the actual power distribution network to the load end and the working values (such as the working power, the working voltage, or the working current) of the load end obtained by the power transducer, the voltage transducer, the current transducer, or the like. The power supply scenario value includes the scenario parameters of the load end during the use, such as the air temperature, the wind speed, the humidity, and the like. The value extraction is performed on the operation leading-in parameter and the operation detection parameter to obtain a leading-in parameter vector and a detection parameter vector. Each value in the parameter vector represents a parameter, such as the power value, the voltage value, the current value, the working power, the working voltage, or the working current.

[0057] Step 2: According to the target power supply value, a plurality of clusters of past power supply reference values are extracted from the power supply values for the power supply of the power distribution network to the load end, and a power supply reference value group is constructed.

[0058] In the preferred but non-limiting embodiment of the present application, in step 2, the power supply values for the power supply of the power distribution network to the load end record the past power supply values of the power distribution network to the load end under different power supply modes of the power distribution network to the load end (that is, the power supply values of the power distribution network to the load end are different). The power supply values of the power distribution network to the load end are used to give the analysis of the power supply mode of the power distribution network to the load end a certain amount of value basis. In the present application, the same past power supply values as the target power supply value are selected from the power supply values for the power supply of the power distribution network to the load end, and there are no abnormal clusters of past power supply values during the power supply of the power distribution network to the load end. In this way, a plurality of clusters of past power supply reference values are obtained, and a power supply reference value group is constructed.

[0059] In the preferred but non-limiting embodiment of the present application, in step 2, optionally, each cluster of the past power supply reference values includes two past target power supply values, two past operating parameter values, and two power supply scenario values, the two past operating parameter values include two operating inducted parameter vectors and two operating detected parameter vectors, the value extraction is performed on the two past operating parameter values to obtain the two operating inducted parameter vectors and the two operating detected parameter vectors.

[0060] Step 3: performing a differential analysis I on the target operating parameter value and the plurality of clusters of the past power supply reference values to extract a parameter differential vector I of the target operating parameter value and a parameter differential vector II of each cluster of the past power supply reference values;

[0061] In the preferred but non-limiting embodiment of the present application, in step 3, the differential analysis I is to analyze the parameter differential of the power distribution network involved in the power supply to the load, that is, to analyze the differential between the inducted parameter vector and the detected parameter vector, just like the differential of each parameter between the inducted parameter vector I and the detected parameter vector I (the differential of the parameter is the absolute value of the difference between the parameters), and the differential is combined to construct the parameter differential vector I of the target operating parameter value. The differential of each parameter between the inducted parameter vector II and the detected parameter vector II of each cluster of the past power supply reference values is calculated (the differential of the parameter is the absolute value of the difference between the parameters), and the differential is combined to construct the parameter differential vector II of each cluster of the past power supply reference values.

[0062] Step 4: performing a differential analysis II on the target operating parameter value and the plurality of clusters of the past power supply reference values to extract a detected parameter differential vector between the target operating parameter value and each cluster of the past power supply reference values;

[0063] In the preferred but non-limiting embodiment of the present application, in step 4, the key to the differential analysis is to analyze the difference between the current power supply value and the past power supply value, including analyzing the difference between the detection parameter vector one corresponding to the target operating parameter value and the detection parameter vector two corresponding to the past power supply reference value of each cluster (the difference is the quantity obtained by subtracting the detection parameter vector two corresponding to the past power supply reference value of each cluster from the detection parameter vector one corresponding to the target operating parameter value), and according to the difference between the lead-in parameter vector one corresponding to the target operating parameter value and the lead-in parameter vector two corresponding to the past power supply reference value of each cluster (the difference is the quantity obtained by subtracting the lead-in parameter vector two corresponding to the past power supply reference value of each cluster from the lead-in parameter vector one corresponding to the target operating parameter value), the difference between the detection parameter vector corresponding to the target operating parameter value and the past power supply reference value of each cluster is obtained as a calibration reference, and the difference vector between the detection parameter corresponding to the target operating parameter value and the past power supply reference value of each cluster is obtained as a calibration reference.

[0064] Step 5: According to the difference vector one and the difference vector two of each cluster, the difference reference coefficient of each cluster is determined, the abnormal detection value of the target operating parameter value is formed according to the difference vector of each cluster and the difference reference coefficient of each cluster, and the control mode of the load end is configured according to the abnormal detection value.

[0065] In the preferred but non-limiting embodiment of the present application, in step 5, the difference reference coefficient is integrated with the difference between the target operating parameter value and the past power supply reference value of each cluster involved in the power supply process of the primary distribution network to the load end (the power supply process of the primary distribution network to the load end is the period from the start of power supply to the load end to the stop of power supply to the load end), the detection parameter difference vector is the correlation between the operating detection parameter obtained by specifically detecting the target operating parameter value and the past power supply reference value of each cluster after the value calibration of the difference involved in the power supply process of the primary distribution network to the load end, the target difference value corresponding to the past power supply reference value of each cluster in the power supply reference value group is obtained by calculation, the abnormal detection value including the target difference value corresponding to the past power supply reference value of each cluster in the power supply reference value group is constructed, and the control mode of the load end is configured according to the abnormal detection value.

[0066] If the target difference value corresponding to the past power supply reference value of each cluster in the abnormal detection value is lower than the difference threshold set in advance, the control mode of the load end is not configured;

[0067] The target power supply value and the power supply scene value of the load end are sent to the control mode forming mode, and the corresponding operation parameter reference value of the load end is formed.

[0068] If the target difference value is higher, it means that the target operation parameter value deviates from the past power supply value by a higher degree. The deviation degree is set by the preset difference threshold. If the corresponding target difference value of each past power supply reference value is lower than the preset difference threshold, it means that the target operation parameter value deviates from the past power supply value, and it is determined that the value is not abnormal. On the contrary, if the corresponding target difference value of a past power supply reference value is not lower than the preset difference threshold, it means that the target operation parameter value deviates from the past power supply value, and it can be considered that the target operation parameter value is abnormal. In this case, in order to ensure the power supply performance of the power distribution network to the load end, the control mode forming mode learned in advance is used to process the values involved in the power supply task of the power distribution network to the load end, including sending the target power supply value and the power supply scene value of the load end to the control mode forming mode, forming the corresponding operation parameter reference value of the load end, and then configuring the control mode of the load end according to the formed operation parameter reference value, so as to achieve active improvement and accurate control of the power supply process of the power distribution network to the load end, and improve the reliability of the power supply performance of the power distribution network to the load end.

[0069] In the preferred but non-limiting embodiment of the present application, in step 4, the target operation parameter value is used to perform differential analysis on the plurality of past power supply reference values to extract the corresponding detection parameter difference vector between the target operation parameter value and the plurality of past power supply reference values, which comprises the following steps:

[0070] The difference between each parameter in the detection parameter vector one and the plurality of detection parameter vector two is calculated (the difference between the parameters is the absolute value of the difference between the parameters), and the plurality of detection difference vectors are obtained by combining the differences. The difference between each parameter in the guide parameter vector one and the plurality of guide parameter vector two is calculated (the difference between the parameters is the absolute value of the difference between the parameters), and the plurality of guide difference vectors are obtained by combining the differences.

[0071] The detection difference vector represents the difference between the target operation parameter value and the detected parameter of the plurality of past power supply reference values, and the sending difference vector represents the difference between the target operation parameter value and the sent parameter of the plurality of past power supply reference values. The detection difference vector is calibrated according to the difference between the target operation parameter value and the corresponding second guide parameter vector of the plurality of past power supply reference values, so as to obtain the corresponding detection parameter difference vector between the target operation parameter value and the plurality of past power supply reference values.

[0072] In the preferred but non-limiting embodiment of the present application, in step 5, the method for determining the difference reference coefficient of each cluster past power supply reference value according to the parameter difference vector one and the plurality of parameter difference vectors two comprises:

[0073] The scene action coefficient of each cluster past power supply reference value is determined according to the power supply scene value one of the load end and the power supply scene value two corresponding to each cluster past power supply reference value.

[0074] In the preferred but non-limiting embodiment of the present application, in step 5, the calculation method of the scene action coefficient comprises: calculating the KL divergence between the power supply scene value one of the load end and the power supply scene value two corresponding to each cluster past power supply reference value, respectively. The KL divergence is suitable for the case where the difference of the scene detection value is very important. The scene elements such as temperature, wind speed, humidity, etc. of the scene can affect the function of the load end and the performance of the power distribution network to the load end. According to the KL divergence analysis of the scene element difference between different values, the scene difference characteristics can be well represented. In the face of a plurality of KL divergences obtained by calculation, it is suitable for subsequent value analysis. The Z-score method is used to perform standardization processing on the plurality of KL divergences, which is suitable for comparison between values. Thus, the scene action coefficient of each cluster past power supply reference value is obtained.

[0075] The difference correlation coefficient of each cluster past power supply reference value is determined according to the approximation analysis of the parameter difference vector one on the plurality of parameter difference vectors two.

[0076] In the preferred but non-limiting embodiment of the present application, in step 5, the calculation method of the difference correlation coefficient comprises: calculating the L2 norm between the parameter difference vector one and the plurality of parameter difference vectors two, respectively, to evaluate the correlation between the parameter difference vectors. Similarly, the Z-score method is used to perform standardization processing on the plurality of L2 norms, so as to obtain the difference correlation coefficient of each cluster past power supply reference value.

[0077] After the scene effect coefficient and the difference correlation coefficient of each cluster past power supply reference value are obtained, the difference reference coefficient of each cluster past power supply reference value is determined according to the scene effect coefficient and the difference correlation coefficient, as shown in the following equation:

[0078]

[0079] In the equation, Esf j is the difference reference coefficient of the jth cluster past power supply reference value, Fp j is the scene effect coefficient of the jth cluster past power supply reference value, St j is the difference correlation coefficient of the jth cluster past power supply reference value, respectively, are the importance parameters of the scene effect coefficient and the difference correlation coefficient set in advance.

[0080] The difference reference coefficient obtained by the above method is used to determine the scene effect coefficient, and the load end is efficiently locked in the scene temperature, humidity and other external effects of the power distribution network, thereby improving the matching and reliability of the power distribution network to the load end during power supply, and the difference reference coefficient integrates the multi-dimensional differences between the target operation parameter value and the past power supply reference value. Through numerical processing and analysis, a multi-dimensional integrated evaluation system is formed, which can completely and comprehensively reflect the hidden dangers of the power distribution network to the load end during power supply.

[0081] In the preferred but non-limiting embodiment of the present application, in step 5, the method for forming the abnormal detection value of the target operation parameter value according to the difference vector of the plurality of clusters of detection parameters and the difference reference coefficient of each cluster past power supply reference value comprises:

[0082] According to the corresponding detection parameter difference vector between the target operation parameter value and the plurality of clusters of past power supply reference values, the difference reference value of each cluster past power supply reference value corresponding to the target operation parameter value is determined;

[0083] In the preferred but non-limiting embodiment of the present application, in step 5, the variance of each cluster detection parameter difference vector is calculated, and the variance of each cluster detection parameter difference vector is taken as the difference reference value of each cluster past power supply reference value corresponding to the target operation parameter value.

[0084] According to the difference reference coefficient of each cluster past power supply reference value, the difference reference value of each cluster past power supply reference value corresponding to the target operation parameter value is calibrated, and the target difference value of the target operation parameter value corresponding to each cluster past power supply reference value in the power supply reference value group is obtained. Here, the operation equation of the target difference value is:

[0085] H k = v k W k

[0086] H is the difference value of the kth cluster past power supply reference value, W k is the difference reference value of the kth cluster past power supply reference value, v k is the difference reference value of the kth cluster past power supply reference value, v k is the difference reference coefficient of the kth cluster past power supply reference value set in advance.

[0087] Through the above method, an abnormal detection value containing a plurality of difference values can be obtained, wherein the difference value is obtained by performing a calibration operation on the difference reference value corresponding to the target operating parameter value of each cluster past power supply reference value, and can accurately reflect the error condition of the power distribution network to the load end power supply performance between the past power supply value and the current target operating parameter value. The higher the error, the higher the difference value. The operation of the difference value comprehensively evaluates the role of the scene action coefficient and the difference correlation coefficient. Based on the analysis operation of the past power supply value and the current operating parameter value, the multi-directional evaluation can comprehensively analyze the mutual action of various elements during the power supply of the power distribution network to the load end, and provide better numerical evidence for the improvement of the power supply performance. The difference value can immediately reflect the error condition of the power supply performance of the power distribution network to the load end, and the integration evaluation of the past reference data has good recyclable improvement function in specific application. By continuously collecting and analyzing new power supply values of the power distribution network to the load end, the difference reference coefficient and the difference critical value set in advance can be dynamically configured, and the power supply process of the power distribution network to the load end is better improved.

[0088] In the preferred but non-limiting embodiment of the present application, the control mode forming mode also includes:

[0089] The control mode forming mode is obtained by learning the input sample, and the sample contains a plurality of cluster operating parameter values, target power supply values corresponding to each cluster operating parameter value, and power supply scene values. Specifically as follows:

[0090] The restricted Boltzmann machine is used to construct the control mode forming mode to master the relationship between the plurality of cluster operating parameter values, the target power supply values corresponding to each cluster operating parameter value, and the power supply scene values in the sample. During the mode learning, the target power supply values corresponding to each cluster operating parameter value and the power supply scene values are used as the input values of the control mode forming mode and input into the restricted Boltzmann machine, and the plurality of cluster operating parameter values are used as the learning target of the control mode forming mode. The control mode forming mode is learned, which can be used to evaluate the corresponding operating parameter value according to the input target power supply value and the power supply scene value, so as to obtain the accurate control mode of the power distribution network to the load end power supply machine.

[0091] The application is suitable for performing abnormal detection on the target operation parameter value during the operation of the load end, can be used to find the hidden abnormal value in the value, the load end is affected by factors such as scene elements and self damage during operation, there is often a small difference between the parameters sent in and the specific parameters of the load end during operation, various factors make the fluctuation of the value often cover the small abnormal value in the value, so the abnormal detection can be performed on the value collected at present combined with the power supply value in the past, a number of past power supply reference values with reference are selected from the information table according to the target power supply value, which is related to the effect of the load end factors and external scene factors on the load end during operation, during the power supply to the load end according to the same target power supply value, the error between the specific parameters and the principle parameters sent under a number of factors will often be calibrated, and small changes will often be performed on the principle parameters sent.

[0092] To find the hidden abnormal value often covered in the value, prevent the power distribution network from adversely affecting the power supply performance of the load end, the present application can achieve active improvement and accurate control of the load end during operation through multi-dimensional analysis of the real-time operation parameter value and the past power supply reference value, improve the reliability and coordination of the power distribution network to the power supply performance of the load end. Through automatic differential analysis operation, the change of the power supply scene and the reliability of the load end itself are efficiently introduced to affect the power supply process of the power distribution network to the load end, and the small abnormal value in the value is detected, so that the power supply of the power distribution network to the load end can perform feedback and configuration control mode on the small abnormal value, and the continuous improvement of the power supply performance is ensured. According to the response learning and control mode improvement of the past power supply value, the power supply change can be matched and the power supply process of the power distribution network to the load end can be continuously improved and improved, and the power supply performance of the power distribution network to the load end can be improved.

[0093] As shown in Figure 2 , the power supply reliability evaluation device of the power distribution network containing source, load and storage, comprising:

[0094] The construction module is used to obtain the target power supply value of the load end, the target operation parameter value and the power supply scene value one, and to obtain the guide parameter vector one and the detection parameter vector one according to the operation parameter value;

[0095] The extraction module is used to extract a number of past power supply reference values from the power supply value for registering the power supply of the power distribution network to the load end according to the target power supply value, and to obtain the power supply reference value group, and to extract the corresponding guide parameter vector two and detection parameter vector two of each cluster of past power supply reference values;

[0096] The analysis module one is used to perform differential analysis one on the target operation parameter value and a number of past power supply reference values, extract the parameter difference vector one of the target operation parameter value, and the parameter difference vector two of each cluster of past power supply reference values;

[0097] a second analysis module configured to perform a second differential analysis on the plurality of clusters of past power supply reference values according to the target operation parameter value, to extract a corresponding detection parameter differential vector between the target operation parameter value and the plurality of clusters of past power supply reference values;

[0098] a control module configured to determine a differential reference coefficient of each cluster of past power supply reference values according to the parameter differential vector and a plurality of parameter differential vectors, to form an abnormal detection value of the target operation parameter value according to the plurality of detection parameter differential vectors and the differential reference coefficient of each cluster of past power supply reference values, and to configure a control mode of the load end according to the abnormal detection value.

[0099] The technical effects of the present application include:

[0100] Through the multi-dimensional analysis of the real-time operation parameter value and the past power supply reference value, the active improvement and accurate control of the load end during operation can be achieved, and the reliability and coordination of the power distribution network to the power supply performance of the load end can be improved. Through the automatic differential analysis operation, the change of the power supply scene and the reliability of the load end itself are efficiently introduced to affect the power supply process of the power distribution network to the load end, and the small abnormal value in the value is searched, so that the power supply of the power distribution network to the load end can perform feedback and configure the control mode for the small abnormal value, guarantee the continuous improvement of the power supply performance, and according to the response learning and control mode improvement of the past power supply value, the power supply change can be matched, and the power supply process of the power distribution network to the load end can be continuously improved and improved, and the power supply performance of the power distribution network to the load end can be improved.

[0101] The present disclosure can be a system, a method, and / or a computer program product. The computer program product can include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present disclosure.

[0102] Computer readable backup medium can be a tangible computer readable storage medium capable of maintaining and backing up instructions for use by an instruction execution system. Computer readable backup medium can be, but is not limited to, an electronic, magnetic, optical, electromagnetic, semiconductor, or further proper combination of the foregoing. Further examples (a non-exhaustive list) of computer readable backup medium include: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanism that is encoded with instructions (e.g., a punch card or a hole in a punch card), and any suitable combination of the foregoing. A computer readable backup medium as used herein is not, and should not be construed as being, a transitory signal per se, such as a radio wave or other freely propagating electromagnetic wave, a electromagnetic wave propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0103] Computer readable program instructions expressed on the computer readable backup medium can be downloaded to a computing / processing device from an external computer or external backup computer via a wireless network, e.g., the Internet, a local area network, a wide area network, and / or a wireless network. The wireless network can include copper transmission cables, optical transmission cables, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A wireless network adapter card or wireless network interface in a computing / processing device receives computer readable program instructions from the wireless network and forwards the computer readable program instructions for storage in a computer readable backup medium within the computing / processing device.

[0104] Computer readable program instructions for carrying out operations of the present disclosure can be assembly-level instructions, instructions-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, conditionally- set values for configuration registers, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) can execute the computer readable program instructions by utilizing state

[0105] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the present application. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the technical solutions of the present application can still be modified or equivalent replaced without departing from the spirit and scope of the present application, and any modification or equivalent replacement should be covered in the protection scope of the claims of the present application.

Claims

1. A method for evaluating the reliability of power supply in a distribution network containing source, grid, load, and energy storage, characterized in that, include: Collect power usage data at the load end and construct a model to perform load power prediction P load; The reliability assessment method for power supply in a power grid including source, grid, load, and energy storage distribution network also includes: Step 1: Obtain the target power supply value, target operating parameter value, and power supply scenario value 1 at the load end; construct the input parameter vector 1 and the detection parameter vector 1 based on the operating parameter value. Step 2: Based on the target power supply value, extract several clusters of past power supply reference values ​​from the power supply values ​​used to register the power supply to the load end of the distribution network and construct a power supply reference value group. Extract the corresponding input parameter vector 2 and detection parameter vector 2 for each cluster of past power supply reference values. Step 3: Perform differential analysis on the target operating parameter value and several clusters of previous power supply reference values, and extract the parameter difference vector one of the target operating parameter value and the parameter difference vector two of the previous power supply reference values ​​of each cluster. Step 4: Based on the target operating parameter values, perform differential analysis on several clusters of past power supply reference values, and extract the corresponding detection parameter difference vectors between the target operating parameter values ​​and several clusters of past power supply reference values; Step 5: Based on parameter difference vector one and several cluster parameter difference vector two, determine the difference base coefficient of the previous power supply base value of each cluster. Based on the difference base coefficient of the previous power supply base value of each cluster, form the abnormal detection value of the target operating parameter value. Configure the control mode of the load side based on the abnormal detection value. In step 5, it is determined whether the target difference value corresponding to the previous power supply reference value of each cluster is lower than the preset difference threshold. If so, the control mode of the load side will not be configured. Conversely, the target power supply value and power supply scenario value of the load end are sent into the control mode to form the mode, forming the corresponding operating parameter reference value of the load end, and the control mode of the load end is configured according to the operating parameter reference value. In step 5, the method for determining the reference coefficients for the historical power supply reference values ​​of each cluster based on parameter difference vector one and several cluster parameter difference vectors two includes: The scenario effect coefficient of each cluster's past power supply benchmark value is determined based on the power supply scenario value one at the load end and the corresponding power supply scenario value two at the previous power supply benchmark value of each cluster. Based on the parameter difference vector pair, an approximation analysis is performed on several cluster parameter difference vectors. Based on the approximation between parameter difference vector one and several cluster parameter difference vectors two, the difference correlation coefficient of the previous power supply reference values ​​of each cluster is determined. After calculating the scenario effect coefficient and the difference correlation coefficient of the historical power supply baseline values ​​for each cluster, the difference baseline coefficient of the historical power supply baseline values ​​for each cluster is determined based on the scenario effect coefficient and the difference correlation coefficient, as shown in the following equation: Within the equation, Esf j Fp is the reference coefficient that distinguishes the historical power supply reference values ​​of the j-th cluster. j St is the scenario effect coefficient of the historical power supply reference value of the j-th cluster. j It is the correlation coefficient between the historical power supply reference values ​​of the j-th cluster, θ, These are the pre-set importance parameters for the scene effect coefficient and the difference correlation coefficient; In step 5, the calculation method of the scenario effect coefficient includes: calculating the KL divergence between the power supply scenario value one at the load end and the corresponding power supply scenario value two of the previous power supply benchmark values ​​of each cluster, and performing standardization processing on several KL divergences using the Z-score method. In step 5, the method for calculating the correlation coefficient includes calculating the L2 norm between parameter difference vector one and several cluster parameter difference vector two to evaluate the correlation between parameter difference vectors, and performing standardization processing on several L2 norms using the Z-score method to obtain the correlation coefficient of the previous power supply reference values ​​of each cluster. In step 5, the method for generating abnormal detection values ​​of target operating parameters based on the difference reference coefficients between several cluster detection parameter difference vectors and the historical power supply reference values ​​of each cluster includes: Based on the target operating parameter values ​​and the corresponding detection parameter difference vectors between several clusters of previous power supply reference values, the difference reference values ​​between the previous power supply reference values ​​of each cluster and the target operating parameter values ​​are determined. Based on the reference coefficients for the historical power supply reference values ​​of each cluster, the reference values ​​corresponding to the target operating parameter values ​​of each cluster are calibrated to obtain the target difference values ​​of the target operating parameter values ​​corresponding to the historical power supply reference values ​​of each cluster in the power supply reference value group. Here, the calculation equation for the target difference value is: H k =n k W k Within the equation, H k W is the target difference value of the historical power supply reference value of the k-th cluster. k It is the reference value that distinguishes the previous power supply reference value of the k-th cluster, ν k It is a pre-set reference coefficient that distinguishes the previous power supply reference values ​​of the k-th cluster; In step 5, the variance of the difference vector of the detection parameters of each cluster is calculated, and the variance of the difference vector of the detection parameters of each cluster is used as the difference reference value of the previous power supply reference value of each cluster corresponding to the target operating parameter value. The extraordinary detection value includes several target differentiation values.

2. The method for evaluating the reliability of power supply in a distribution network containing source, grid, load, and energy storage as described in claim 1, is characterized in that, In step 1, the target power supply value is used to represent the power supply specification value of the distribution network currently supplying power to the load. The target power supply value includes the type of device at the load, the power consumption capacity of the load, the rated parameters of the load, and the control specification value of the distribution network for the load. The target operation parameter value includes operation input parameter 1 and operation detection parameter 1, which respectively represent the parameters sent to the load by the distribution network and the operating values ​​of the load obtained through detection during the actual power supply of the distribution network to the load. The power supply scenario value 1 includes the scenario parameters of the load during the operation. The operation input parameter 1 and operation detection parameter 1 are extracted to obtain input parameter vector 1 and detection parameter vector 1. Each value in the parameter vector represents a parameter.

3. The method for evaluating the reliability of power supply in a distribution network containing source, grid, load, and energy storage as described in claim 2, is characterized in that, In step 2, the power supply data used to register the power supply of the distribution network to the load end includes the power supply data of the distribution network to the load end under different power supply modes. The power supply data used to register the power supply of the distribution network to the load end selects the previous power supply data that is the same as the target power supply data, and it is a number of previous power supply data that have not exceeded the power supply during the period of power supply of the distribution network to the load end. In this way, a number of previous power supply reference data are obtained and a power supply reference data group is constructed. In step 2, any set of past power supply reference values ​​includes past target power supply values, past operating parameter values, and power supply scenario value 2. The past operating parameter values ​​include operating input parameter 2 and operating detection parameter 2. Value extraction is performed on the past operating parameter values ​​to obtain the input parameter vector 2 corresponding to the operating input parameter 2 and the detection parameter vector 2 corresponding to the operating detection parameter 2.

4. The method for evaluating the reliability of power supply in a distribution network containing source, grid, load, and energy storage as described in claim 3, is characterized in that, In step 3, the difference between the input parameter vector and the detection parameter vector is analyzed. Just as the difference between each parameter of the input parameter vector and the detection parameter vector is calculated, the difference is combined to construct the parameter difference vector 1 that obtains the target operating parameter value. The difference between each parameter of the input parameter vector 2 and the detection parameter vector 2 that obtain the historical power supply reference value of each cluster is calculated. The difference is combined to construct the parameter difference vector 2 that obtains the historical power supply reference value of each cluster.

5. The method for evaluating the reliability of power supply in a distribution network containing source, grid, load, and energy storage according to claim 4, characterized in that, In step 4, the differential analysis includes analyzing the difference between the detection parameter vector one corresponding to the target operating parameter value and the detection parameter vector two corresponding to several clusters of previous power supply reference values. The difference between the input parameter vector one corresponding to the target operating parameter value and the input parameter vector two corresponding to several clusters of previous power supply reference values ​​is used as the calibration reference to obtain the detection parameter difference vectors corresponding to the target operating parameter value and several clusters of previous power supply reference values.

6. The method for evaluating the reliability of power supply in a distribution network containing source, grid, load, and energy storage as described in claim 5, is characterized in that, In step 4, a method for performing discriminative analysis on several clusters of historical power supply reference values ​​based on the target operating parameter values, and extracting the corresponding detection parameter difference vectors between the target operating parameter values ​​and the several clusters of historical power supply reference values, includes: The differences between each parameter of the first detection parameter vector and several clusters of second detection parameter vectors are calculated separately, and the differences are combined to obtain several detection difference vectors. The differences between each parameter of the first input parameter vector and several second input parameter vectors are calculated separately, and the differences are combined to obtain multiple input difference vectors.

7. The method for evaluating the reliability of power supply in a distribution network containing source, grid, load, and energy storage as described in claim 6, is characterized in that, The formation of modalities in the face of control models also includes: The control mode is formed through learning from the input samples. The samples contain several clusters of operating parameter values, and the corresponding target power supply values ​​and power supply scenario values ​​for each cluster of operating parameter values, as detailed below: The Restricted Boltzmann Machine (RBM) is used to construct the control mode formation mode to understand the relationship between the values ​​of several clusters of operating parameters within the sample and the corresponding target power supply values ​​and power supply scenario values. During mode learning, the target power supply values ​​and power supply scenario values ​​corresponding to the values ​​of each cluster of operating parameters are used as the input values ​​for the control mode formation mode and fed into the RBM. The values ​​of several clusters of operating parameters are used as the learning target for the control mode formation mode, and the control mode formation mode is learned.

Images