Concentrator with electrical equipment state monitoring function

Abstract

The application relates to the technical field of power monitoring, in particular to a concentrator with an electrical equipment state monitoring function; an unbalance degree is obtained according to state characteristic values among three phases; clustering is carried out according to unbalance degrees at all historical moments; distribution density is obtained according to interval characteristics of different moments in a moment cluster; data similarity is obtained according to unbalance degrees in the moment cluster; reference credibility is obtained according to moment quantity characteristics, distribution density and data similarity in the moment cluster; abnormality degree is obtained according to difference characteristics of unbalance degrees corresponding to the current moment and a reference moment cluster, and reference credibility of the reference moment cluster; abnormal trend values are obtained according to abnormality degrees of adjacent historical time periods of the current moment. The application obtains a warning index of the current moment according to the abnormality degree and the abnormal trend value and carries out warning, and the monitoring accuracy of the concentrator for the electrical equipment power state is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power monitoring, in particular to a concentrator with electrical equipment state monitoring function. BACKGROUND

[0002] Traditional power concentrators are mainly responsible for the collection and transmission of electric energy data, but with the complication of power distribution network structure and the diversification of equipment operating environment, simple energy metering has been difficult to meet the monitoring needs of the health status of electrical equipment, and concentrators with electrical equipment state monitoring function have emerged as the times require. Among many types of electrical faults, three-phase imbalance fault is particularly prominent in terms of its impact on power load balance, equipment life and energy efficiency; three-phase imbalance can cause motor overheating, abnormal increase of neutral line current, harmonic distortion, and even cause equipment damage and safety accidents; therefore, accurate monitoring of three-phase imbalance fault has become the key to improving the intelligence level of the concentrator.

[0003] At present, the monitoring of three-phase imbalance usually adopts a fixed threshold early warning method, which lacks adaptability to different application scenarios and is prone to false positives or false negatives; for example, in industrial sites, production loads often change dynamically, and mild imbalance is a normal phenomenon, so setting a fixed threshold can easily trigger invalid early warning; while in places such as hospitals where the requirement for power quality is high, even a slight imbalance can cause equipment abnormalities, and if the fixed threshold is too high, it can easily fail to provide timely warning. Therefore, the early warning of three-phase imbalance according to the fixed threshold reduces the intelligence level and accuracy of the concentrator. SUMMARY

[0004] In order to solve the above technical problem that the early warning of three-phase imbalance by fixed threshold reduces accuracy, the purpose of the present application is to provide a concentrator with electrical equipment state monitoring function, and the technical solution adopted is as follows:

[0005] The data acquisition module is used to acquire the state characteristic values of three-phase electricity at different times, and the state characteristic values include the amplitude and phase of the voltage;

[0006] The data analysis module is used to obtain the imbalance degree according to the difference characteristics of the state characteristic values between three phases at any time; to obtain different time clusters by clustering the imbalance degrees of all times in history; to obtain the distribution density according to the interval characteristics of different times in the time cluster; to obtain the data similarity according to the difference characteristics of the imbalance degrees corresponding to different times in the time cluster; and to obtain the reference credibility of the time cluster according to the time quantity characteristics of the time cluster, the distribution density and the data similarity;

[0007] anomaly analysis module, configured to obtain a reference time cluster according to time interval features of a current time and time clusters at different times, obtain an anomaly degree of the current time according to difference features of unbalance degrees corresponding to the current time and the reference time cluster and a reference credibility of the reference time cluster, and obtain an anomaly trend value according to change features of anomaly degrees of adjacent historical time periods of the current time;

[0008] an early warning module, configured to obtain an early warning index of the current time according to the anomaly degree and the anomaly trend value, and perform early warning on a three-phase load state according to the early warning index.

[0009] Further, the step of obtaining the unbalance degree according to difference features of state feature values between the three phases at an arbitrary time includes:

[0010] calculating an average value of absolute values of differences between state feature values of an arbitrary phase and other phases at the arbitrary time to obtain a state difference degree of the arbitrary phase, and calculating an average value of state difference degrees of all phases at the arbitrary time to obtain the unbalance degree of the arbitrary time.

[0011] Further, the step of obtaining the distribution density according to interval features of different times in the time cluster includes:

[0012] taking an arbitrary time in the time cluster as a center time of a 24-hour time period and constructing a target time period of the arbitrary time, calculating time intervals between corresponding times of other times in the time cluster in the target time period and the arbitrary time to obtain interval lengths of the other times and the arbitrary time, calculating an average value of interval lengths of all other times and the arbitrary time to obtain a comprehensive interval length of the arbitrary time, and calculating an average value of comprehensive interval lengths of all times in the time cluster and performing negative correlation mapping to obtain the distribution density of the time cluster.

[0013] Further, the step of obtaining the data similarity according to difference features of unbalance degrees corresponding to different times in the time cluster includes:

[0014] calculating an average value of absolute values of differences between unbalance degrees of any two times in the time cluster and performing negative correlation mapping to obtain the data similarity of the time cluster.

[0015] Further, the step of obtaining the reference credibility of the time cluster according to time quantity features in the time cluster, the distribution density and the data similarity includes:

[0016] calculating a ratio of a time quantity in the time cluster to a total time quantity to obtain a quantity ratio value, calculating a ratio of the distribution density to the data similarity to obtain a cluster type feature value, and calculating a product of the quantity ratio value and the cluster type feature value to obtain the reference credibility of the time cluster.

[0017] Further, the step of obtaining a reference time cluster according to the time interval feature of the current time and the time clusters at different times comprises:

[0018] calculating the average value of the interval length of each time in the time cluster and the current time, obtaining the interval feature value of the time cluster and the current time; taking the time cluster corresponding to the minimum value of the interval feature value as the reference time cluster of the current time.

[0019] Further, the step of obtaining the abnormality degree of the current time according to the difference feature of the imbalance degree corresponding to the current time and the reference time cluster, and the reference credibility of the reference time cluster comprises:

[0020] calculating the average value of the ratio of the imbalance degree of the current time and all times in the reference time cluster, obtaining the difference degree value; calculating the product of the imbalance degree of the current time, the difference degree value and the reference credibility, obtaining the abnormality degree of the current time.

[0021] Further, the step of obtaining the abnormality trend value according to the change feature of the abnormality degree of the adjacent historical period of the current time comprises:

[0022] calculating the change slope of the abnormality degree of the adjacent historical period of the current time and positively correlating mapping, obtaining the abnormality trend value.

[0023] Further, the step of obtaining the early warning index of the current time according to the abnormality degree and the abnormality trend value comprises:

[0024] taking the maximum value of the normalized abnormality degree and the normalized abnormality trend value as the early warning index of the current time.

[0025] Further, the step of early warning of the three-phase load state according to the early warning index comprises:

[0026] when the early warning index exceeds the preset early warning threshold, early warning of the three-phase load state.

[0027] The present application has the following beneficial effects:

[0028] In the present application, the unbalance degree can represent the three-phase unbalance characteristics at any time; since the user's power consumption behavior has regularity, the periodicity and trend characteristics of the three-phase unbalance characteristics can be analyzed according to the unbalance degrees at different times. The time cluster can determine the distribution time of similar unbalance degrees in the historical period; the distribution density can represent the concentration degree of the time distribution in the cluster, and then judge whether the three-phase unbalance characteristics corresponding to the cluster is caused by normal user power consumption behavior, and preliminarily improve the monitoring accuracy of the power consumption state. The data similarity can represent the difference degree of the unbalance degrees in the time cluster, and further improve the monitoring accuracy of the power consumption state; the reference credibility can represent the reliability degree of the unbalance degrees in the time cluster, so that the calculation accuracy of the abnormality degree is higher. The reference time cluster can determine the time cluster close to the current time, so that the reference range of the power consumption state monitoring is more accurate, and the monitoring accuracy is improved. The abnormality degree can represent the abnormal degree of the three-phase unbalance characteristics at the current time in combination with the distribution characteristics of the unbalance degrees at the similar time in the history; the abnormal trend value can represent the change trend of the abnormal degree of the three-phase unbalance characteristics, so that the monitoring and early warning are more sensitive. Finally, the early warning index is obtained and monitored, which can be compared and analyzed by referring to the data characteristics of the unbalance degrees at the similar time in the history, and then the accuracy of judging the three-phase unbalance degree according to the fixed threshold is higher, and the adaptability to the power consumption scene is stronger. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, the following will briefly introduce the drawings needed in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0030] Figure 1 A concentrator module block diagram with electrical equipment state monitoring function provided by an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following will combine the drawings and the preferred embodiments to specifically describe the specific implementation, structure, features and effects of the concentrator with electrical equipment state monitoring function according to the present application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0032] Unless otherwise defined, 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.

[0033] The specific scheme of the concentrator with the electrical equipment state monitoring function provided by the application is specifically described below in combination with the drawings.

[0034] Please refer to Figure 1 which shows a concentrator module block diagram provided by an embodiment of the application, the concentrator includes the following modules:

[0035] The data acquisition module S1 is configured to acquire state characteristic values of three-phase power at different time points, the state characteristic values including amplitudes and phases of voltages.

[0036] In the embodiment of the application, the implementation scenario is to monitor the power state of electrical equipment through the concentrator to improve the monitoring accuracy. Three-phase power imbalance is usually manifested in the inconsistency of amplitudes or phases of voltages and currents, among which voltage imbalance is the most common, and the causes mainly include voltage amplitude inequality or phase angle deviation. The occurrence probability and manifestation form of imbalance caused by different causes may be different in actual power consumption scenarios. For example, when three-phase imbalance occurs in a low-voltage distribution area of a certain community, if there is a significant difference in three-phase voltage amplitudes while the phase remains normal, it usually indicates that there is a problem such as poor contact, uneven load or local overload in the line of a certain phase, resulting in a decrease in the voltage of the phase. If the three-phase voltage amplitudes are basically the same but the phase angle deviates seriously from 120°, it is more likely to be caused by factors such as user wiring error, equipment failure or harmonic interference. Then, the state characteristic values of three-phase power at different time points are acquired, the state characteristic values including amplitudes and phases of voltages. In the embodiment of the application, the state characteristic values are collected every 0.5 seconds, and the data range of one month before the current time is saved. The implementer can set it according to the implementation scenario. During the collection process, the concentrator has basic data preprocessing capability, can identify and filter missing values and abnormal values, and keeps time sequence synchronization accurate.

[0037] The data analysis module S2 is configured to obtain an imbalance degree according to the difference characteristics of the state characteristic values between three phases at any time point, to obtain different time point clusters by clustering the imbalance degrees of all time points in history, to obtain a distribution density according to the interval characteristics of different time points in the time point cluster, to obtain a data similarity according to the difference characteristics of the imbalance degrees corresponding to different time points in the time point cluster, and to obtain a reference credibility of the time point cluster according to the time point quantity characteristics, the distribution density and the data similarity in the time point cluster.

[0038] Since the power supply area connected by the concentrator usually has a relatively fixed power consumption behavior mode, the power consumption behavior of the user tends to show certain regularity, thereby making the voltage fluctuation in the power grid have time sequence characteristics that can be mined and analyzed. For different causes of imbalance, the three-phase voltage imbalance is first subdivided into "amplitude type" and "phase type", thereby helping the concentrator to more accurately identify different fault types and improve the pertinence and accuracy of the early warning. If the three-phase imbalance characteristics of the amplitude type are analyzed, the state characteristic value in the subsequent step is the voltage amplitude; if the three-phase imbalance characteristics of the phase type are analyzed, the state characteristic value in the subsequent step is the phase of the voltage, and the steps are the same. Taking the voltage amplitude type in the three-phase imbalance as an example, first, the imbalance degree is obtained according to the difference characteristics of the state characteristic values between any three phases at any time.

[0039] Preferably, in the embodiment of the application, the step of obtaining the imbalance degree comprises: calculating the average value of the absolute value of the difference between the state characteristic values of any phase and other phases at any time to obtain the state difference degree of the any phase; wherein the state characteristic value is the voltage amplitude, and the greater the state difference degree means the greater the voltage difference between the any phase and the other two phases. The average value of the state difference degrees of all phases at the any time is calculated to obtain the imbalance degree at the any time; the greater the state difference degrees corresponding to all phases means the more obvious the three-phase imbalance characteristics at the any time, and the greater the imbalance degree.

[0040] Further, since the imbalance degree of three-phase power is usually closely related to the actual power consumption behavior of the user, and the power consumption behavior of the user has strong time regularity, for example, the load difference between morning and evening peaks in residential areas, working days and weekends in industrial areas, etc. The distribution of the imbalance degree at different times in time also obviously shows periodic and trend characteristics, for example, in the residential area, the voltage of a certain phase may drop due to the concentrated use of air conditioners and high-power kitchen electrical equipment, thereby causing amplitude imbalance; in the industrial area, phase imbalance is prone to occur during the start-stop of equipment at the daily start time. Therefore, the imbalance degrees at all times in the history can be clustered to obtain different time clusters; in the embodiment of the application, the existing density peak clustering algorithm is used for clustering, the imbalance degrees in each time cluster are similar, and the specific clustering steps are not described again.

[0041] After obtaining different time clusters, the distribution characteristics of the time in the time cluster can be analyzed, thereby helping to improve the recognition accuracy of the concentrator for three-phase imbalance, and therefore the distribution density is obtained according to the interval characteristics of different times in the time cluster. Preferably, in the embodiment of the present application, the step of obtaining the distribution density comprises: taking any time in the time cluster as the central time of the 24-hour period and constructing the target period of the any time; because the user's power consumption behavior at similar times each day is similar, the target period is constructed to ignore the differences of different times on different days. The time interval between the corresponding time of the other time in the time cluster in the target period and the any time is calculated to obtain the interval length of the other time and the any time; for example, the other time is 22 o'clock on Thursday, the any time is 0 o'clock on Tuesday, the target period of the any time is the time range from 12 o'clock on Monday to 12 o'clock on Tuesday, the other time in the target period is 22 o'clock, and then the interval length of the two is 2 hours. The smaller the interval length, the closer the two times are. The average of the interval lengths of all other times and the any time is calculated to obtain the comprehensive interval length of the any time; the smaller the comprehensive interval length, the closer the any time and the other times are in time distribution, and the higher the probability of similar three-phase imbalance characteristics appearing near the time. The average of the comprehensive interval lengths of all times in the time cluster is calculated and negatively correlated to obtain the distribution density of the time cluster; the larger the distribution density, the more concentrated the time distribution in the time cluster is, the more likely similar three-phase imbalance characteristics appear at similar times, and the more obvious the regularity of three-phase imbalance characteristics at the time distribution in the time cluster is, and the more likely it is caused by the user's normal power consumption behavior.

[0042] Further, the difference between the imbalance degrees corresponding to different times in the same time cluster also affects the result of the distribution density when clustering, and the larger the difference between the corresponding imbalance degrees in the time cluster, the smaller the distribution density, thereby affecting the referenceability of the time cluster and the accuracy of three-phase imbalance monitoring, and therefore the difference characteristics of the imbalance degrees need to be analyzed to make the reference reliability of the time cluster more reliable. Therefore, the data similarity is obtained according to the difference characteristics of the imbalance degrees corresponding to different times in the time cluster; preferably, in the embodiment of the present application, the step of obtaining the data similarity comprises: calculating the average of the absolute value of the difference between the imbalance degrees of any two times in the time cluster and negatively correlating to obtain the data similarity of the time cluster; in the embodiment of the present application, the negatively correlating is calculated by function, which represents an exponential function with a natural constant as the base, and a represents the mapping object. The smaller the difference between the imbalance degrees of any two times, the more similar the corresponding imbalance degrees of the time cluster, and the larger the data similarity.

[0043] After the distribution density and the data similarity of the time point cluster are obtained, the reference credibility of the time point cluster can be obtained according to the number of time points in the time point cluster, the distribution density and the data similarity; preferably, in the embodiment of the present application, the step of obtaining the reference credibility comprises: calculating the ratio of the number of time points in the time point cluster to the total number of time points to obtain the number ratio; the more the number of time points in the time point cluster, the greater the number ratio, which means that the scale is larger and the number of similar three-phase imbalance characteristics in the historical time points is greater, so that the corresponding three-phase imbalance characteristics in the time point cluster are more likely to be caused by the normal power consumption behavior of the user. The cluster feature value is obtained by calculating the ratio of the distribution density and the data similarity; the smaller the data similarity and the greater the distribution density, the greater the cluster feature value, which can further represent that the time point distribution of the time point cluster is more concentrated, and the corresponding three-phase imbalance characteristics in the time point cluster are more likely to be caused by the normal power consumption behavior of the user, and the time point distribution regularity is more obvious. The reference credibility of the time point cluster is obtained by calculating the product of the number ratio and the cluster feature value; the greater the reference credibility, the more likely the three-phase imbalance characteristics in the time point cluster are caused by the normal power consumption behavior of the user, and the more credible the imbalance degree.

[0044] The abnormality analysis module S3 is configured to obtain a reference time point cluster according to the time interval feature of the current time point and different time point clusters; obtain the abnormality degree of the current time point according to the difference feature of the imbalance degree corresponding to the current time point and the reference time point cluster and the reference credibility of the reference time point cluster; and obtain the abnormal trend value according to the change feature of the abnormality degree of the adjacent historical time period of the current time point.

[0045] After the reference credibility of all time point clusters is obtained, the reference time point cluster can be obtained according to the time interval feature of the current time point and different time point clusters; preferably, in the embodiment of the present application, the step of obtaining the reference time point cluster comprises: calculating the average value of the interval length of each time point in the time point cluster and the current time point to obtain the interval feature value of the time point cluster and the current time point; the smaller the interval feature value, the closer the current time point to the time point distribution of the time point cluster. The time point cluster corresponding to the minimum value of the interval feature value is taken as the reference time point cluster of the current time point, and the closer the time point distribution of the current time point and the reference time point cluster, the more the three-phase imbalance characteristics in the reference time point cluster can be compared and analyzed with the imbalance degree of the current time point to judge the similarity degree of the imbalance characteristics of the current time point and the historical situation; therefore, the abnormality degree of the current time point is obtained according to the difference feature of the imbalance degree corresponding to the current time point and the reference time point cluster and the reference credibility of the reference time point cluster; it should be noted that if the imbalance degree of the current time point is 0, it means that the current time point does not appear three-phase imbalance, and the power consumption behavior is normal, so that the subsequent calculation steps are not performed.

[0046] Preferably, in the embodiments of the present application, the step of obtaining the abnormality degree comprises: calculating the average value of the ratio of the unbalance degree of the current moment to the unbalance degree of all moments in the reference moment cluster to obtain a difference degree value; the greater the difference degree value, the greater the three-phase unbalanced state of the current moment than the three-phase unbalanced state of the historical similar moment, and the more abnormal the three-phase unbalanced state of the current moment; the product of the unbalance degree of the current moment, the difference degree value and the reference credibility is calculated to obtain the abnormality degree of the current moment; the greater the unbalance degree of the current moment, the more likely the user electricity state of the current moment is abnormal; the greater the difference degree value and the reference credibility, the greater the unbalance degree of the current moment than the three-phase unbalance degree of the historical similar moment and the higher the data reliability, the more likely the user electricity state of the current moment is abnormal; therefore, the greater the abnormality degree, the more likely the electricity state of the current moment is abnormal. The formula for obtaining the abnormality degree comprises:

[0047]

[0048] In the formula, R represents the abnormality degree of the current moment, G represents the unbalance degree of the current moment, K represents the reference credibility of the reference moment cluster, N represents the number of moments in the reference moment cluster, represents the unbalance degree of the nth moment in the reference moment cluster, d represents a preset minimum positive number, and when the denominator is 0, it participates in the calculation, which is 0.01 in the embodiments of the present application, represents the difference degree value. The calculation of the abnormality degree can refer to the data characteristics of the unbalance degree of the historical similar moment for comparison and analysis, so that the accuracy of judging the three-phase unbalance degree according to the fixed threshold is higher, and the adaptability to the electricity scene is stronger.

[0049] Further, the abnormality degree represents the abnormal state of the current moment, but the monitoring ability of the potential enhanced abnormal trend is weak, for example, the feature that the unbalance degree has an upward trend is difficult to discover in time according to the abnormality degree of a single moment; therefore, the abnormal trend value is obtained according to the change feature of the abnormality degree of the adjacent historical period of the current moment. Preferably, in the embodiments of the present application, the step of obtaining the abnormal trend value comprises: calculating the change slope of the abnormality degree of the historical adjacent period of the current moment and positively correlating the mapping to obtain the abnormal trend value, in the embodiments of the present application, the historical adjacent period is the previous 5 minutes of the current moment, and the greater the change slope of the abnormality degree of the moment, the more obvious the trend of the gradually increasing abnormality degree, and the greater the abnormal trend value; in the embodiments of the present application, the positive correlation mapping is calculated by the function, represents an exponential function with a natural constant as the base, and b represents the mapping object.

[0050] The warning module S4 is used for obtaining the warning index of the current moment according to the abnormality degree and the abnormal trend value; and warning the three-phase load state according to the warning index.

[0051] The greater the abnormality degree or abnormality trend value at the current moment, the greater the possibility of the current power consumption abnormality or the more obvious the abnormality trend. Both cases need timely warning to ensure the normal operation of the electrical equipment. Therefore, the warning index at the current moment is obtained according to the abnormality degree and the abnormality trend value, specifically including: taking the maximum value of the normalized abnormality degree and the normalized abnormality trend value as the warning index at the current moment; the greater the warning index, the more abnormal the current power consumption state or the more obvious the abnormality trend, which needs timely warning. Then, the three-phase load state can be warned according to the warning index, and the three-phase load state is warned when the warning index exceeds the preset warning threshold; in the embodiment of the present application, the preset warning threshold is 0.6, which can be determined by the implementer according to the implementation scene; when the preset warning threshold is exceeded, it means that the three-phase imbalance degree is obviously abnormal and needs to be handled in time. It should be noted that the amplitude type and phase type three-phase imbalance states need to be monitored at the same time, so as to identify more power consumption abnormalities and quickly locate the causes, and improve the monitoring accuracy of three-phase imbalance. Therefore, the abnormality degree and the abnormality trend value at the current moment can more accurately represent the abnormality degree of the power consumption state at the current moment, and the accuracy is higher than that of monitoring according to the fixed threshold, and the adaptability to the power consumption scene is stronger.

[0052] In summary, the embodiment of the present application provides a concentrator with an electrical equipment state monitoring function; the imbalance degree is obtained according to the state characteristic value between three phases; clustering is performed according to the imbalance degree of all historical moments, the distribution density is obtained according to the interval characteristics of different moments in the moment cluster; the data similarity is obtained according to the imbalance degree in the moment cluster; the reference credibility is obtained according to the moment quantity characteristics, the distribution density and the data similarity in the moment cluster; the abnormality degree is obtained according to the difference characteristics of the imbalance degree corresponding to the current moment and the reference moment cluster, and the reference credibility of the reference moment cluster; and the abnormality trend value is obtained according to the abnormality degree of the adjacent historical period of the current moment. The present application obtains the warning index at the current moment according to the abnormality degree and the abnormality trend value and performs warning, thereby improving the monitoring accuracy of the concentrator for the electrical equipment power consumption state.

[0053] It should be noted that the above-mentioned embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0054] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other, and each embodiment mainly describes the difference from other embodiments.

Claims

1. A concentrator having an electrical equipment state monitoring function, characterized by comprising: The concentrator comprises the following modules: a data acquisition module, configured to acquire state characteristic values of three-phase electricity at different time instants, the state characteristic values comprising amplitudes and phases of voltages; a data analysis module, configured to acquire an unbalance degree according to difference characteristics of the state characteristic values among the three phases at an arbitrary time instant; to cluster according to the unbalance degrees of all time instants in history, obtain different time instant clusters; acquire a distribution density according to interval characteristics of different time instants in the time instant clusters; acquire a data similarity according to difference characteristics of the unbalance degrees corresponding to different time instants in the time instant clusters; acquire a reference credibility of the time instant cluster according to time instant quantity characteristics in the time instant cluster, the distribution density and the data similarity; an abnormality analysis module, configured to acquire a reference time instant cluster according to a time interval characteristic of a current time instant and different time instant clusters; acquire an abnormality degree of the current time instant according to difference characteristics of the unbalance degrees corresponding to the current time instant and the reference time instant cluster, and the reference credibility of the reference time instant cluster; acquire an abnormality trend value according to a change characteristic of the abnormality degree of a neighboring historical time period of the current time instant; a pre-warning module, configured to acquire a pre-warning index of the current time instant according to the abnormality degree and the abnormality trend value; to pre-warn the three-phase load state according to the pre-warning index; the step of acquiring the reference credibility of the time instant cluster according to the time instant quantity characteristics in the time instant cluster, the distribution density and the data similarity comprises: calculating a ratio of the number of time instants in the time instant cluster to the total number of time instants to obtain a quantity proportion value; calculating a ratio of the distribution density to the data similarity to obtain a cluster class characteristic value; calculating a product of the quantity proportion value and the cluster class characteristic value to obtain the reference credibility of the time instant cluster; the step of acquiring the abnormality degree of the current time instant according to the difference characteristics of the unbalance degrees corresponding to the current time instant and the reference time instant cluster, and the reference credibility of the reference time instant cluster comprises: calculating an average value of ratios of the unbalance degrees of the current time instant to all time instants in the reference time instant cluster to obtain a difference degree value; calculating a product of the unbalance degree of the current time instant, the difference degree value and the reference credibility to obtain the abnormality degree of the current time instant.

2. The concentrator with electrical equipment state monitoring function according to claim 1, characterized in that, the step of acquiring the unbalance degree according to the difference characteristics of the state characteristic values among the three phases at an arbitrary time instant comprises: calculating an average value of absolute values of differences between state characteristic values of an arbitrary phase and other phases at the arbitrary time instant to obtain a state difference degree of the arbitrary phase; calculating an average value of state difference degrees of all phases at the arbitrary time instant to obtain the unbalance degree of the arbitrary time instant.

3. The concentrator with electrical equipment state monitoring function according to claim 1, characterized in that, the step of acquiring the distribution density according to interval characteristics of different time instants in the time instant clusters comprises: Taking any time point in the time point cluster as a central time point of a 24-hour time period and constructing a target time period of the any time point; calculating time intervals of other time points in the time point cluster in corresponding time points in the target time period and the any time point, obtaining interval lengths of the other time points and the any time point; calculating an average of the interval lengths of all other time points and the any time point, obtaining a comprehensive interval length of the any time point; calculating an average of comprehensive interval lengths of all time points in the time point cluster and negatively correlating mapping, obtaining a distribution density of the time point cluster.

4. The concentrator with electrical equipment state monitoring function according to claim 1, characterized in that, The step of obtaining the data similarity according to the difference characteristics of the unbalance degrees of different time points in the time point cluster comprises: calculating an average of absolute values of difference values of unbalance degrees between any two time points in the time point cluster and negatively correlating mapping, obtaining the data similarity of the time point cluster.

5. The concentrator with electrical equipment state monitoring function according to claim 3, characterized in that, The step of obtaining the reference time point cluster according to time interval characteristics of the current time point and different time point clusters comprises: calculating an average of the interval lengths of each time point in the time point cluster and the current time point, obtaining interval characteristic values of the time point cluster and the current time point; taking a time point cluster corresponding to a minimum value of the interval characteristic values as a reference time point cluster of the current time point.

6. The concentrator with electrical equipment state monitoring function according to claim 1, characterized in that, The step of obtaining the abnormal trend value according to change characteristics of the abnormal degree of adjacent historical time periods of the current time point comprises: calculating a change slope of the abnormal degree of adjacent historical time periods of the current time point and positively correlating mapping, obtaining the abnormal trend value.

7. The concentrator with electrical equipment state monitoring function according to claim 1, characterized in that, The step of obtaining the early warning index of the current time point according to the abnormal degree and the abnormal trend value comprises: taking a maximum value of the abnormal degree and the abnormal trend value after normalization as the early warning index of the current time point.

8. The concentrator with electrical equipment state monitoring function according to claim 1, characterized in that, The step of early warning of the three-phase load state according to the early warning index comprises: when the early warning index exceeds a preset early warning threshold, early warning of the three-phase load state.

Images