A concentrator intelligent fusion terminal based on multi-source fusion data control

Abstract

The present invention relates to the field of data processing technology, and in particular to a concentrator intelligent fusion terminal based on multi-source fusion data control. The terminal includes a processor and a memory, and the processor executes a computer program in the memory to implement the following steps: obtaining a data fusion moment for any device collected by the concentrator, and obtaining a multidimensional electric energy data sequence for any data fusion moment; recording the multidimensional electric energy data at any data fusion moment as multidimensional target electric energy data, and obtaining the electric energy usage level of each dimension of electric energy data based on data differences in the multidimensional electric energy data sequence; obtaining the fusion weight of each dimension of electric energy data based on data fluctuation characteristics in the multidimensional electric energy data sequence; and obtaining fused data at any data fusion moment based on the electric energy usage level and fusion weight of each dimension of electric energy data in the multidimensional target electric energy data, thereby reducing operating overhead while obtaining more accurate multi-source fusion data.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a concentrator intelligent fusion terminal based on multi-source fusion data regulation. Background Art

[0002] With the large-scale integration of new loads such as distributed photovoltaics and electric vehicle charging stations, distribution networks face the dual challenges of insufficient holographic perception and weak intelligent control capabilities. Multi-source data fusion technology, by integrating data from multiple devices such as smart meters, low-voltage circuit breakers, reactive power compensation devices, and electric vehicle charging stations, can overcome the limitations of a single data source and extract more complete and accurate information. For example, concentrators collect data from smart meters, sensors, and other devices. Through multi-source data fusion and control, they achieve real-time monitoring and analysis of user electricity usage behavior, assisting power companies in load forecasting, electricity bill settlement, and anti-theft efforts, thereby improving the intelligence of power services.

[0003] Traditional multi-source data fusion methods usually use a weighted fusion algorithm with fixed weights to obtain multi-source fused data. However, the concentrator obtains many types of data, and the data sources mainly come from electricity consumption data of different residents, multiple sensor data, etc. The data sources may increase or decrease dynamically. Therefore, traditional multi-source data fusion methods lack dynamic adaptability, resulting in unsatisfactory fusion results and the multi-source fusion data cannot accurately reflect the user's electricity consumption. At the same time, the large amount of data in the concentrator leads to a large computational load, which increases the concentrator's operating overhead.

[0004] Therefore, how to adaptively obtain the dynamic weight of each data, obtain more accurate multi-source fusion data, and adaptively adjust the frequency of multi-source data fusion to reduce operating overhead has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, an embodiment of the present invention provides a concentrator intelligent fusion terminal based on multi-source fusion data regulation to solve the problem of how to adaptively obtain the dynamic weight of each data, obtain more accurate multi-source fusion data, and adaptively adjust the frequency of multi-source data fusion to reduce operating overhead.

[0006] An embodiment of the present invention provides a concentrator intelligent fusion terminal based on multi-source fusion data control, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the following method is implemented:

[0007] For any device collected by the concentrator, a data fusion frequency is obtained based on a preset power consumption frequency. Based on the data fusion frequency, at least two data fusion moments are obtained. For any data fusion moment, multidimensional power data of the device at each moment is obtained to obtain a multidimensional power data sequence at the data fusion moment. The multidimensional power data includes current data, voltage data, and power data.

[0008] Recording the multidimensional electric energy data at any data fusion moment as multidimensional target electric energy data, and obtaining the electric energy usage level of each dimension of electric energy data in the multidimensional target electric energy data according to the data difference of the multidimensional electric energy data in the multidimensional electric energy data sequence;

[0009] obtaining, based on the fluctuation characteristics of the multidimensional electric energy data in the multidimensional electric energy data sequence, a power quality score of each dimension of the electric energy data in the multidimensional target electric energy data, and obtaining, based on the power quality score of each dimension of the electric energy data in the multidimensional target electric energy data, a fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data;

[0010] According to the power usage level and fusion weight of each dimension of the power data in the multi-dimensional target power data, the fusion data of any device at any data fusion moment is obtained, and the fusion data of each device collected by the concentrator at each data fusion moment is obtained for the concentrator to perform regulation.

[0011] Preferably, obtaining the electric energy usage level of each dimension of electric energy data in the multidimensional target electric energy data according to the data difference of the multidimensional electric energy data in the multidimensional electric energy data sequence includes:

[0012] In the multidimensional electric energy data sequence, two multidimensional electric energy data whose electric energy data difference is 0 and whose corresponding time is closest to the time of fusion of any one of the data are obtained, and the data between the two multidimensional electric energy data and the two multidimensional electric energy data are combined into a reference subsequence;

[0013] For any dimension of electric energy data in the multidimensional target electric energy data, obtaining an initial usage level reference value of the electric energy data in any dimension according to the fluctuation characteristics of the multidimensional electric energy data in the reference subsequence;

[0014] Obtain the absolute value of the difference between the electric energy data of any dimension and the initial usage level reference value of the electric energy data of any dimension to obtain the usage difference, obtain the addition result of the usage difference and a constant 1, and obtain the electric energy usage level of the electric energy data of any dimension based on the difference between the constant 1 and the inverse of the addition result.

[0015] Preferably, obtaining the initial usage level reference value of the electric energy data in any dimension according to the fluctuation characteristics of the multi-dimensional electric energy data in the reference subsequence includes:

[0016] If the electric energy data of any dimension is current data, obtaining the mean value of the current data in the reference subsequence as the initial usage degree reference value of the electric energy data of any dimension;

[0017] If the electric energy data of any dimension is voltage data, obtaining the mean value of the voltage data in the reference subsequence as a reference value of the initial usage degree of the electric energy data of any dimension;

[0018] If the electric energy data in any dimension is electric quantity data, an initial usage level reference value of the electric energy data in any dimension is set to 0.

[0019] Preferably, obtaining the power quality score of each dimension of the power data in the multidimensional target power data according to the fluctuation characteristics of the multidimensional power data in the multidimensional power data sequence includes:

[0020] For any dimension of electric energy data in the multidimensional target electric energy data, obtaining electric energy data belonging to the same dimension as the electric energy data in the multidimensional electric energy data sequence, and constructing a target sequence of a first preset length with the electric energy data in the any dimension as the last data;

[0021] Evenly dividing the target sequence into at least two target subsequences of a second preset length, where the second preset length is smaller than the first preset length;

[0022] According to the data fluctuation characteristics of the target sequence and the data fluctuation characteristics of each target subsequence, a power quality score of the power data of any dimension in the multi-dimensional target power data is obtained.

[0023] Preferably, obtaining the power quality score of the power data in any dimension in the multi-dimensional target power data according to the data fluctuation characteristics of the target sequence and the data fluctuation characteristics of each target subsequence includes:

[0024] If the electric energy data of any dimension is current data or voltage data, then for any target subsequence, obtain the variance of the target subsequence, calculate the difference between a constant 1 and the inverse of the variance, and obtain the degree of fluctuation of the target subsequence;

[0025] Numbering the target subsequences in the target sequence according to time, obtaining the sum of the numbers of all target subsequences to obtain a total number value, and calculating the ratio of the number of any target subsequence to the total number value to obtain a fluctuation weight of the fluctuation degree of any target subsequence;

[0026] Obtain the volatility and weight of each target subsequence respectively, and perform weighted summation of the volatility of all target subsequences to obtain the total volatility;

[0027] The degree of fluctuation of the target subsequence including the electric energy data of any dimension is recorded as the real-time fluctuation degree, the real-time fluctuation degree and the total fluctuation degree are weighted and summed to obtain the fluctuation index of the electric energy data of any dimension, and the difference between the constant 1 and the fluctuation index is calculated to obtain the power quality score of the electric energy data of any dimension in the multidimensional target electric energy data.

[0028] Preferably, the obtaining of the power quality score of the power data of any dimension in the multi-dimensional target power data according to the data fluctuation characteristics of the target sequence and the data fluctuation characteristics of each target subsequence further includes:

[0029] If the electric energy data in any dimension belongs to electric quantity data, fitting the target sequence to obtain a target fitting curve, and obtaining fitting values ​​of each electric energy data in the target sequence on the target fitting curve;

[0030] For any electric energy data in the target sequence, obtaining an absolute value of a difference between the electric energy data and its fitting value to obtain a degree of deviation of the electric energy data, obtaining an average value of the reciprocal of the degree of deviation of each electric energy data in the target sequence, and obtaining a power quality usage indicator of the target sequence;

[0031] Recording the target subsequence including the electric energy data of any dimension as a real-time subsequence, and obtaining the power quality usage index of the real-time subsequence;

[0032] A weighted sum is performed on the power quality usage index of the target sequence and the power quality usage index of the real-time subsequence to obtain a power quality score of the power data of any dimension in the multi-dimensional target power data.

[0033] Preferably, obtaining the fusion weight of each dimension of the multi-dimensional target electric energy data according to the power quality score of each dimension of the multi-dimensional target electric energy data includes:

[0034] Obtaining a cumulative value of the power quality score of each dimension of the multi-dimensional target power data to obtain a total power quality score;

[0035] The ratio of the power quality score of each dimension of the power data in the multidimensional target power data to the total power quality score is respectively obtained to obtain the fusion weight of each dimension of the power data in the multidimensional target power data.

[0036] Preferably, the acquiring of the fused data of any device at any data fusion moment according to the power usage level and fusion weight of each dimension of the power data in the multi-dimensional target power data includes:

[0037] The fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data is used as the weight coefficient of the electric energy usage degree of each dimension of the electric energy data in the multidimensional target electric energy data, and the electric energy usage degree of each dimension of the electric energy data in the multidimensional target electric energy data is weighted and summed to obtain the fusion data of any device at any data fusion moment.

[0038] Preferably, obtaining the data fusion frequency according to the preset power consumption frequency includes:

[0039] Divide the time period collected by the concentrator into at least two power consumption intervals according to power demand, and obtain the duration and average power consumption of any power consumption interval;

[0040] Obtain the inverse of the time length, record it as the time index, obtain the difference between the constant 1 and the inverse of the average power consumption, obtain the power consumption index, obtain the average between the time index and the power consumption index, obtain the frequency weight, obtain the product of the frequency weight and the preset power consumption frequency, and obtain the data fusion frequency of any power consumption interval.

[0041] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0042] The present invention is directed to any device collected by the concentrator, obtains a data fusion frequency according to a preset power consumption frequency, obtains at least two data fusion moments according to the data fusion frequency, obtains multidimensional electric energy data of any device at each moment for any data fusion moment, and obtains a multidimensional electric energy data sequence at any data fusion moment, wherein the multidimensional electric energy data includes current data, voltage data, and electric energy data; records the multidimensional electric energy data at any data fusion moment as multidimensional target electric energy data, obtains the electric energy data of each dimension in the multidimensional target electric energy data according to the data difference of the multidimensional electric energy data in the multidimensional electric energy data sequence, and obtains the electric energy data of each dimension in the multidimensional target electric energy data. The degree of electric energy usage; based on the fluctuation characteristics of the multidimensional electric energy data in the multidimensional electric energy data sequence, obtain the electric energy quality score of each dimension of the electric energy data in the multidimensional target electric energy data; based on the electric energy quality score of each dimension of the electric energy data in the multidimensional target electric energy data, obtain the fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data; based on the electric energy usage degree and fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data, obtain the fusion data of any device at any data fusion moment, and obtain the fusion data of each device collected by the concentrator at each data fusion moment, for the concentrator to control. In which, the data fusion frequency is obtained according to the preset power consumption frequency to reduce the operating overhead of the concentrator; by obtaining the electric energy quality score, the electric energy data of different dimensions are converted into the same form of scoring data, and at the same time, the dynamic adaptive fusion weight is obtained according to the electric energy quality score to obtain the fusion data, so that the fusion data can more accurately reflect the degree of use of the fusion grid and the corresponding data status. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a flowchart of a concentrator intelligent fusion method based on multi-source fusion data control provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0045] The embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0046] It should be noted that the terms "first," "second," and the like in the specification of the present disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0047] In order to illustrate the technical solution of the present invention, specific embodiments are provided below.

[0048] An embodiment of the present invention provides a concentrator intelligent fusion terminal based on multi-source fusion data control, comprising a processor and a memory, wherein the processor executes a computer program in the memory to implement a concentrator intelligent fusion method based on multi-source fusion data control, such as Figure 1 As shown, the method includes the following steps:

[0049] Step S101: For any device collected by the concentrator, a data fusion frequency is obtained according to a preset power consumption frequency. According to the data fusion frequency, at least two data fusion moments are obtained. For any data fusion moment, multidimensional electric energy data of any device at each moment is obtained to obtain a multidimensional electric energy data sequence at any data fusion moment, wherein the multidimensional electric energy data includes current data, voltage data and power data.

[0050] With the large-scale integration of new loads such as distributed photovoltaics and electric vehicle charging stations, distribution networks face the dual challenges of insufficient holographic perception and weak intelligent control capabilities. Multi-source data fusion technology, by integrating data from multiple devices such as smart meters, low-voltage circuit breakers, reactive power compensation devices, and electric vehicle charging stations, can overcome the limitations of a single data source and extract more complete and accurate information. For example, concentrators collect data from smart meters, sensors, and other devices. Through multi-source data fusion and control, they achieve real-time monitoring and analysis of user electricity usage behavior, assisting power companies in load forecasting, electricity bill settlement, and anti-theft efforts, thereby improving the intelligence of power services.

[0051] The scenario targeted by the embodiments of the present invention is a concentrator terminal with residential electricity consumption as the main data source. Traditional multi-source data fusion methods usually adopt a weighted fusion algorithm with fixed weights to obtain multi-source fused data. However, the concentrator obtains many types of data, and the data sources mainly come from electricity consumption data of different residents, multiple sensor data, etc. The data sources may increase or decrease dynamically. Therefore, the traditional multi-source data fusion method lacks dynamic adaptability, resulting in unsatisfactory fusion results and the multi-source fusion data cannot accurately reflect the user's electricity consumption. At the same time, the large amount of data in the concentrator leads to a large computational load, which increases the operating overhead of the concentrator.

[0052] Therefore, this embodiment first obtains the data fusion frequency according to the preset power consumption frequency to reduce the operating overhead of the concentrator, and then converts the power data of different dimensions into the same form of scoring data by obtaining the power quality score. At the same time, the dynamic adaptive fusion weight is obtained according to the power quality score, and finally the fused data is obtained, so that the fused data can more accurately reflect the usage level of the fused power grid and the corresponding data status, thereby enabling the concentrator to perform more accurate regulation.

[0053] The core parameters in the power system are current, voltage, and power. These three types of data can, to a certain extent, indicate the degree of grid usage. Therefore, this embodiment obtains the current data, voltage data, and power data of any device collected by the concentrator at each moment, forming multi-dimensional power data at each moment.

[0054] In theory, residential electricity consumption is calculated from voltage and current. However, in reality, there can be discrepancies between electricity consumption and current and voltage. These discrepancies are primarily due to various factors, including measurement errors, device characteristics, grid fluctuations, and data collection and processing methods. For example, an insufficient meter sampling frequency may fail to capture rapidly changing current and voltage, leading to discrepancies in electricity consumption calculations. Electronic devices (such as variable-frequency air conditioners and LED lights) generate harmonic currents, causing inaccurate meter readings. Inductive loads (such as motors) cause current to lag behind voltage, leading to an overestimation of electricity consumption if the meter is not power factor compensated. Grid voltage instability (such as voltage drops during peak hours) affects actual power consumption, but voltmeters typically calculate based on the rated voltage, leading to discrepancies. Time desynchronization between the meter and the concentrator can lead to misalignment of electricity consumption data. Therefore, electricity consumption analysis based solely on single data is incomplete and inaccurate. Fusion of data is necessary to more accurately reflect grid usage and the corresponding data status.

[0055] Since the acquisition of fused data involves resource loss, if calculations are performed at every moment, the operating overhead of the concentrator will be high. Therefore, the frequency of obtaining fused data can be adjusted according to the actual situation of residents' electricity consumption to reduce operating overhead.

[0056] Since residential electricity consumption has peak and off-peak periods, the time of concentrator collection is divided into different electricity consumption intervals according to residents' electricity demand and electricity usage habits, namely peak and off-peak periods. During peak periods, the frequency of obtaining fusion data can be increased, and during off-peak periods, the computing nodes of the fusion task can be reduced. The fusion data can be obtained in an environment with fixed power checks, ensuring the real-time nature of the data while reducing operating expenses.

[0057] Since the duration and power consumption of each power consumption interval are different, for any power consumption interval, the duration and average power consumption of any power consumption interval can be obtained, and the data fusion frequency can be obtained based on the duration and average power consumption of any power consumption interval and the preset power consumption frequency.

[0058] The method for obtaining the data fusion frequency based on the duration and average power consumption of any power consumption interval and the preset power consumption frequency is as follows:

[0059] Obtain the inverse of the time length, record it as the time index, obtain the difference between the constant 1 and the inverse of the average power consumption, obtain the power consumption index, obtain the average between the time index and the power consumption index, obtain the frequency weight, obtain the product of the frequency weight and the preset power consumption frequency, and obtain the data fusion frequency of any power consumption interval.

[0060] In one embodiment, taking the peak period of electricity consumption as an example, the peak period of electricity consumption is generally 4 hours, and the average electricity consumption is 10 kWh. The calculation formula for the data fusion frequency during the peak period of electricity consumption is: f , where f is the data fusion frequency during the peak period of electricity consumption, 4 is the duration of the peak period of electricity consumption, and 10 is the average power consumption during the peak period of electricity consumption. This is the preset power consumption frequency (i.e., the maximum frequency during peak power consumption). It needs to be set based on actual conditions. There is no restriction here and it can be set according to the specific implementation scenario.

[0061] Furthermore, according to the data fusion frequency, at least two data fusion moments at which data fusion is required are obtained. The main purpose of the present invention is to obtain fused data. Hereinafter, for any data fusion moment, the multidimensional electric energy data of any device at each moment is obtained, and the multidimensional electric energy data of each moment between any data fusion moment and the previous data fusion moment are combined to form a multidimensional electric energy data sequence at any data fusion moment, which is used to obtain analysis of the fused data of any device at any data fusion moment, wherein the last multidimensional electric energy data in the multidimensional electric energy data sequence belongs to the multidimensional electric energy data collected at any data fusion moment. In this embodiment, the acquisition frequency of the multidimensional electric energy data is set to once per second, which is not limited here and can be set according to the specific implementation scenario.

[0062] Step S102 , recording the multidimensional power data at any data fusion moment as multidimensional target power data, and obtaining the power usage level of each dimension of power data in the multidimensional target power data according to the data difference of the multidimensional power data in the multidimensional power data sequence.

[0063] When performing data fusion, direct fusion is not possible due to the inconsistent formats of current, voltage, and power data. However, these three types of data can all reflect the degree of electricity usage, so the corresponding degree of electricity usage can be obtained based on the performance characteristics of each type of data. Among them, the main manifestation of current data when used in the power grid is that the current increases instantly (which may exceed the rated current), then returns to the normal value and remains stable, with the possibility of slight fluctuations; voltage performance is similar, but the voltage will be lower during peak power consumption and higher during low power consumption; the accumulated power continues to rise; therefore, based on the data differences of the multi-dimensional power data in the multi-dimensional power data sequence, the method for obtaining the degree of electricity usage of each dimension of the multi-dimensional target power data is as follows:

[0064] (1) Obtain the voltage and current values ​​when the residential electricity consumption is 0 as the reference values ​​when the usage level is 0.

[0065] Specifically, in the multidimensional electric energy data sequence, two multidimensional electric energy data whose electric energy data difference is 0 and whose corresponding time is closest to the time of fusion of any data are obtained, and the data between the two multidimensional electric energy data and the two multidimensional electric energy data are combined into a reference subsequence;

[0066] For any dimension of electric energy data in the multi-dimensional target electric energy data, if the electric energy data in any dimension is current data, obtaining a mean value of the current data in the reference subsequence as a reference value of the initial usage level of the electric energy data in any dimension;

[0067] If the electric energy data of any dimension is voltage data, obtaining the mean value of the voltage data in the reference subsequence as a reference value of the initial usage degree of the electric energy data of any dimension;

[0068] If the electric energy data in any dimension is electric quantity data, an initial usage level reference value of the electric energy data in any dimension is set to 0.

[0069] (2) Obtaining the absolute value of the difference between the electric energy data of any dimension and the initial usage level reference value of the electric energy data of any dimension to obtain the usage difference, obtaining the addition result of the usage difference and a constant 1, and obtaining the electric energy usage level of the electric energy data of any dimension according to the difference between the constant 1 and the reciprocal of the addition result.

[0070] In one embodiment, taking the voltage data in the multi-dimensional target electric energy data as an example, the calculation formula of the electric energy usage degree of the voltage data is:

[0071]

[0072] in, is the degree of electric energy usage of voltage data; V is the voltage data in the multi-dimensional target electric energy data; is the initial usage level reference value of the voltage data in the multi-dimensional target electric energy data; 1 is a constant; is the absolute value symbol.

[0073] It should be noted that the greater the difference between the voltage data in the multi-dimensional target electric energy data and its initial usage level reference value, the more electricity is used, and the greater the electric energy usage level of the voltage data.

[0074] Similarly, the power usage level of each dimension of the multi-dimensional target power data is obtained.

[0075] Step S103: obtaining a power quality score of each dimension of the power data in the multidimensional target power data according to the fluctuation characteristics of the multidimensional power data in the multidimensional power data sequence; and obtaining a fusion weight of each dimension of the power data in the multidimensional target power data according to the power quality score of each dimension of the power data in the multidimensional target power data.

[0076] Since traditional multi-source data fusion methods lack dynamic adaptability, the fusion results obtained are not ideal and the multi-source fusion data cannot accurately reflect the user's electricity consumption. Therefore, according to the fluctuation characteristics of the multi-dimensional power data in the multi-dimensional power data sequence, the power quality score of each dimension of power data in the multi-dimensional target power data can be obtained to reflect the data quality of each power data. Then, according to the power quality score of each dimension of power data in the multi-dimensional target power data, the fusion weight of each dimension of power data in the multi-dimensional target power data can be obtained, thereby obtaining fused data.

[0077] Since voltage and current data usually remain in a relatively stable state, their power quality scores are mainly based on the volatility of the data, while electricity consumption data is a changing quantity, and its power quality score is mainly based on the trend stability of the data. Therefore, for any dimension of electric energy data in the multidimensional target electric energy data, electric energy data belonging to the same dimension as the electric energy data of any dimension can be obtained in the multidimensional electric energy data sequence, and the electric energy data of any dimension is taken as the last data to construct a target sequence with a first preset length of 30 minutes (electric energy data collected within 30 minutes, i.e., including 1,800 electric energy data); the target sequence is evenly divided into at least two target sub-sequences with a second preset length of 60 seconds (electric energy data collected within 60 seconds, i.e., including 60 electric energy data). There is no restriction here and it can be set according to the specific implementation scenario; based on the data fluctuation characteristics of the target sequence and the data fluctuation characteristics of each target sub-sequence, the power quality score of the electric energy data of any dimension in the multidimensional target electric energy data is obtained.

[0078] If any dimension of electric energy data is current data or voltage data, the method for obtaining the power quality score of any dimension of electric energy data is as follows:

[0079] For any target subsequence, obtain the variance of the target subsequence, calculate the difference between a constant 1 and the inverse of the variance, and obtain the degree of fluctuation of the target subsequence;

[0080] Numbering the target subsequences in the target sequence according to time, obtaining the sum of the numbers of all target subsequences to obtain a total number value, and calculating the ratio of the number of any target subsequence to the total number value to obtain a fluctuation weight of the fluctuation degree of any target subsequence;

[0081] Obtain the volatility and weight of each target subsequence respectively, and perform weighted summation of the volatility of all target subsequences to obtain the total volatility;

[0082] The degree of fluctuation of the target subsequence including the electric energy data of any dimension is recorded as the real-time fluctuation degree, the real-time fluctuation degree and the total fluctuation degree are weighted and summed to obtain the fluctuation index of the electric energy data of any dimension, and the difference between the constant 1 and the fluctuation index is calculated to obtain the power quality score of the electric energy data of any dimension in the multidimensional target electric energy data.

[0083] In one embodiment, taking the voltage data in the multi-dimensional target electric energy data as an example, the calculation formula for the power quality score of the voltage data is:

[0084]

[0085] in, Score the power quality of voltage data; is the variance of the last target subsequence in the target sequence (i.e., the target subsequence containing voltage data in the multidimensional target electric energy data); is the variance of the i-th target subsequence in the target sequence; M is the number of target subsequences in the target sequence; is the weight coefficient of the real-time volatility, is the weight coefficient of the total fluctuation degree. Since the volatility of voltage data at the current time (at any data fusion moment) is more important, in this embodiment, , , there is no restriction here and it can be set according to the specific implementation scenario.

[0086] It should be noted that is the fluctuation degree of the i-th target subsequence in the target sequence, The larger the value is, the more unstable the power data in the target subsequence i in the target sequence is, that is, the stronger the volatility is. The power data may be interfered or have equipment quality problems, and the data quality is low. The bigger it is, the The smaller it is; is the fluctuation weight of the fluctuation degree of the i-th target subsequence in the target sequence. The larger i is, the closer the i-th target subsequence is to the voltage data in the multidimensional target electric energy data, and the smaller the time decay is. The bigger it is, the The smaller it is.

[0087] If the electric energy data in any dimension is electric quantity data, the method for obtaining the electric energy quality score of the electric energy data in any dimension is as follows:

[0088] The target sequence is fitted to obtain a target fitting curve, and fitting values ​​of each electric energy data in the target sequence are respectively obtained on the target fitting curve. Obtaining the fitting curve belongs to the prior art and will not be described in detail here;

[0089] For any electric energy data in the target sequence, obtaining an absolute value of a difference between the electric energy data and its fitting value to obtain a degree of deviation of the electric energy data, obtaining an average value of the reciprocal of the degree of deviation of each electric energy data in the target sequence, and obtaining a power quality usage indicator of the target sequence;

[0090] Recording the target subsequence including the electric energy data of any dimension as a real-time subsequence, and obtaining the power quality usage index of the real-time subsequence;

[0091] A weighted sum is performed on the power quality usage index of the target sequence and the power quality usage index of the real-time subsequence to obtain a power quality score of the power data of any dimension in the multi-dimensional target power data.

[0092] In one embodiment, the calculation formula for the power quality score of the electric quantity data is:

[0093]

[0094] in, Score the power quality of electricity data; is the jth power data in the real-time subsequence; is the fitted value of the jth electric quantity data in the real-time subsequence; N is the number of electric quantity data in the real-time subsequence; is the kth power data in the target sequence; is the fitted value of the kth electric quantity data in the target sequence; T is the number of electric quantity data in the target sequence; is the weight coefficient of the power quality usage index of the real-time subsequence, The weight coefficient of the target sequence power quality indicator is set. Since the degree of deviation of the power data at the current time (at any data fusion moment) is more important, , ,There is no restriction here and it can be set according to the specific implementation scenario; is the absolute value symbol.

[0095] It should be noted that It is the power quality usage index of the target sequence, indicating the quality of the power used within the time corresponding to the target sequence. The larger the value is, the greater the deviation between the target sequence power data and its fitted value is. The smaller it is, the The smaller it is.

[0096] Furthermore, according to the power quality score of each dimension of the power data in the multi-dimensional target power data, a method for obtaining the fusion weight of each dimension of the power data in the multi-dimensional target power data is as follows:

[0097] Obtaining a cumulative value of the power quality score of each dimension of the multi-dimensional target power data to obtain a total power quality score;

[0098] The ratio of the power quality score of each dimension of the power data in the multidimensional target power data to the total power quality score is respectively obtained to obtain the fusion weight of each dimension of the power data in the multidimensional target power data.

[0099] In one embodiment, taking the h-th dimension electric energy data in the multi-dimensional target electric energy data as an example, the calculation formula for the fusion weight of the h-th dimension electric energy data in the multi-dimensional target electric energy data is:

[0100]

[0101] in, is the fusion weight of the h-th dimension electric energy data in the multi-dimensional target electric energy data; is the power quality score of the h-th dimension power data in the multidimensional target power data; 3 is the total number of dimensions in the multidimensional target power data.

[0102] It should be noted that The larger it is, the better the data quality of the h-th dimension electric energy data in the multi-dimensional target electric energy data is. The bigger it is.

[0103] Similarly, the fusion weight of each dimension of the electric energy data in the multi-dimensional target electric energy data is obtained.

[0104] Step S104, based on the power usage level and fusion weight of each dimension of the power data in the multi-dimensional target power data, obtain the fusion data of any device at any data fusion moment, and obtain the fusion data of each device collected by the concentrator at each data fusion moment, for the concentrator to control.

[0105] In step S103, the fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data is obtained, and in step S102, the electric energy usage level of each dimension of the electric energy data in the multidimensional target electric energy data is obtained. Furthermore, the fusion data corresponding to the multidimensional target electric energy data can be obtained based on the electric energy usage level and fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data.

[0106] Among them, according to the power usage degree and fusion weight of each dimension of power data in the multi-dimensional target power data, the method for obtaining the fusion data corresponding to the multi-dimensional target power data is as follows:

[0107] The fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data is used as the weight coefficient of the electric energy usage degree of each dimension of the electric energy data in the multidimensional target electric energy data, and the electric energy usage degree of each dimension of the electric energy data in the multidimensional target electric energy data is weighted and summed to obtain the fusion data corresponding to the multidimensional target electric energy data.

[0108] In one embodiment, the calculation formula for the fusion data corresponding to the multi-dimensional target electric energy data is:

[0109]

[0110] in, It is the fusion data corresponding to the multi-dimensional target electric energy data; is the fusion weight of the h-th dimension electric energy data in the multi-dimensional target electric energy data; is the energy usage level of the hth dimension of the multidimensional target energy data; 3 is the total number of dimensions in the multidimensional target energy data.

[0111] Similarly, the fused data collected by the concentrator for each device at each data fusion moment is obtained and used for the concentrator's regulation, such as load forecasting and electricity bill settlement by power companies; optimized scheduling of distributed energy; real-time detection of equipment status and fault warning, power scheduling, etc.

[0112] Because fused data can more accurately reflect energy usage and the corresponding data status, it's closer to real data. Once the fused data is obtained, to view the real data for each dimension, the fused data is used as the energy usage for each dimension, and the energy usage calculation formula is inverted to obtain the real data for each dimension. The main purpose of this invention is to adaptively obtain dynamic weights for each data point to obtain more accurate fused data. The concentrator's use of fused data for control and the inverse calculation of the formula are prior art and will not be further elaborated here.

[0113] In summary, the embodiment of the present invention obtains a data fusion frequency according to a preset power consumption frequency for any device collected by the concentrator, obtains at least two data fusion moments according to the data fusion frequency, obtains multidimensional electric energy data of any device at each moment for any data fusion moment, and obtains a multidimensional electric energy data sequence at any data fusion moment, wherein the multidimensional electric energy data includes current data, voltage data and electric energy data; the multidimensional electric energy data at any data fusion moment is recorded as multidimensional target electric energy data, and the electric energy data of each dimension in the multidimensional target electric energy data is obtained according to the data difference of the multidimensional electric energy data in the multidimensional electric energy data sequence. The method comprises the following steps: obtaining the degree of electric energy usage of the energy data; obtaining the electric energy quality score of each dimension of the electric energy data in the multidimensional target electric energy data according to the fluctuation characteristics of the multidimensional electric energy data in the multidimensional electric energy data sequence; obtaining the fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data according to the electric energy quality score of each dimension of the electric energy data in the multidimensional target electric energy data; obtaining the fusion data of any device at any data fusion moment according to the degree of electric energy usage and the fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data, and obtaining the fusion data of each device collected by the concentrator at each data fusion moment for the concentrator to control. The data fusion frequency is obtained according to the preset power consumption frequency to reduce the operating overhead of the concentrator; by obtaining the electric energy quality score, the electric energy data of different dimensions are converted into the same form of scoring data, and the dynamic adaptive fusion weight is obtained according to the electric energy quality score to obtain the fusion data, so that the fusion data can more accurately reflect the degree of use of the fusion grid and the corresponding data status.

[0114] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A concentrator intelligent fusion terminal based on multi-source fusion data control, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the following method is implemented: For any device collected by the concentrator, a data fusion frequency is obtained based on a preset power consumption frequency. Based on the data fusion frequency, at least two data fusion moments are obtained. For any data fusion moment, multidimensional power data of the device at each moment is obtained to obtain a multidimensional power data sequence at the data fusion moment. The multidimensional power data includes current data, voltage data, and power data. Recording the multidimensional electric energy data at any data fusion moment as multidimensional target electric energy data, and obtaining the electric energy usage level of each dimension of electric energy data in the multidimensional target electric energy data according to the data difference of the multidimensional electric energy data in the multidimensional electric energy data sequence; obtaining, based on the fluctuation characteristics of the multidimensional electric energy data in the multidimensional electric energy data sequence, a power quality score of each dimension of the electric energy data in the multidimensional target electric energy data, and obtaining, based on the power quality score of each dimension of the electric energy data in the multidimensional target electric energy data, a fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data; According to the power usage level and fusion weight of each dimension of the power data in the multi-dimensional target power data, the fusion data of any device at any data fusion moment is obtained, and the fusion data of each device collected by the concentrator at each data fusion moment is obtained for the concentrator to perform regulation.

2. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 1 is characterized in that: The obtaining, based on the data difference of the multidimensional electric energy data in the multidimensional electric energy data sequence, the electric energy usage level of each dimension of the electric energy data in the multidimensional target electric energy data includes: In the multidimensional electric energy data sequence, two multidimensional electric energy data whose electric energy data difference is 0 and whose corresponding time is closest to the time of fusion of any one of the data are obtained, and the data between the two multidimensional electric energy data and the two multidimensional electric energy data are combined into a reference subsequence; For any dimension of electric energy data in the multidimensional target electric energy data, obtaining an initial usage level reference value of the electric energy data in any dimension according to the fluctuation characteristics of the multidimensional electric energy data in the reference subsequence; Obtain the absolute value of the difference between the electric energy data of any dimension and the initial usage level reference value of the electric energy data of any dimension to obtain the usage difference, obtain the addition result of the usage difference and a constant 1, and obtain the electric energy usage level of the electric energy data of any dimension based on the difference between the constant 1 and the inverse of the addition result.

3. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 2 is characterized in that: The obtaining, based on the fluctuation characteristics of the multi-dimensional electric energy data in the reference subsequence, an initial usage level reference value of the electric energy data in any dimension includes: If the electric energy data of any dimension is current data, obtaining the mean value of the current data in the reference subsequence as the initial usage degree reference value of the electric energy data of any dimension; If the electric energy data of any dimension is voltage data, obtaining the mean value of the voltage data in the reference subsequence as a reference value of the initial usage degree of the electric energy data of any dimension; If the electric energy data in any dimension is electric quantity data, an initial usage level reference value of the electric energy data in any dimension is set to 0.

4. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 1 is characterized in that: The obtaining, according to the fluctuation characteristics of the multidimensional power data in the multidimensional power data sequence, a power quality score of each dimension of the power data in the multidimensional target power data, comprises: For any dimension of electric energy data in the multidimensional target electric energy data, obtaining electric energy data belonging to the same dimension as the electric energy data in the multidimensional electric energy data sequence, and constructing a target sequence of a first preset length with the electric energy data in the any dimension as the last data; Evenly dividing the target sequence into at least two target subsequences of a second preset length, where the second preset length is smaller than the first preset length; According to the data fluctuation characteristics of the target sequence and the data fluctuation characteristics of each target subsequence, a power quality score of the power data of any dimension in the multi-dimensional target power data is obtained.

5. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 4 is characterized in that: The obtaining, based on the data fluctuation characteristics of the target sequence and the data fluctuation characteristics of each target subsequence, a power quality score of the power data of any dimension in the multi-dimensional target power data includes: If the electric energy data of any dimension is current data or voltage data, then for any target subsequence, obtain the variance of the target subsequence, calculate the difference between a constant 1 and the inverse of the variance, and obtain the degree of fluctuation of the target subsequence; Numbering the target subsequences in the target sequence according to time, obtaining the sum of the numbers of all target subsequences to obtain a total number value, and calculating the ratio of the number of any target subsequence to the total number value to obtain a fluctuation weight of the fluctuation degree of any target subsequence; Obtain the volatility and weight of each target subsequence respectively, and perform weighted summation of the volatility of all target subsequences to obtain the total volatility; The degree of fluctuation of the target subsequence including the electric energy data of any dimension is recorded as the real-time fluctuation degree, the real-time fluctuation degree and the total fluctuation degree are weighted and summed to obtain the fluctuation index of the electric energy data of any dimension, and the difference between the constant 1 and the fluctuation index is calculated to obtain the power quality score of the electric energy data of any dimension in the multidimensional target electric energy data.

6. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 5, characterized in that: The step of obtaining a power quality score of the power data of any dimension in the multi-dimensional target power data according to the data fluctuation characteristics of the target sequence and the data fluctuation characteristics of each target subsequence further includes: If the electric energy data in any dimension belongs to electric quantity data, fitting the target sequence to obtain a target fitting curve, and obtaining fitting values ​​of each electric energy data in the target sequence on the target fitting curve; For any electric energy data in the target sequence, obtaining an absolute value of a difference between the electric energy data and its fitting value to obtain a degree of deviation of the electric energy data, obtaining an average value of the reciprocal of the degree of deviation of each electric energy data in the target sequence, and obtaining a power quality usage indicator of the target sequence; Recording the target subsequence including the electric energy data of any dimension as a real-time subsequence, and obtaining the power quality usage index of the real-time subsequence; A weighted sum is performed on the power quality usage index of the target sequence and the power quality usage index of the real-time subsequence to obtain a power quality score of the power data of any dimension in the multi-dimensional target power data.

7. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 1 is characterized in that: The obtaining, according to the power quality score of each dimension of the power data in the multi-dimensional target power data, a fusion weight of each dimension of the power data in the multi-dimensional target power data, comprises: Obtaining a cumulative value of the power quality score of each dimension of the multi-dimensional target power data to obtain a total power quality score; The ratio of the power quality score of each dimension of the power data in the multidimensional target power data to the total power quality score is respectively obtained to obtain the fusion weight of each dimension of the power data in the multidimensional target power data.

8. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 1 is characterized in that: The obtaining, according to the power usage degree and fusion weight of each dimension of the power data in the multi-dimensional target power data, fused data of any device at any data fusion moment, includes: The fusion weight of each dimension of the electric energy data in the multidimensional target electric energy data is used as the weight coefficient of the electric energy usage degree of each dimension of the electric energy data in the multidimensional target electric energy data, and the electric energy usage degree of each dimension of the electric energy data in the multidimensional target electric energy data is weighted and summed to obtain the fusion data of any device at any data fusion moment.

9. The concentrator intelligent fusion terminal based on multi-source fusion data control according to claim 8, characterized in that: The obtaining of the data fusion frequency according to the preset power consumption frequency includes: Divide the time period collected by the concentrator into at least two power consumption intervals according to power demand, and obtain the duration and average power consumption of any power consumption interval; Obtain the inverse of the time length, record it as the time index, obtain the difference between the constant 1 and the inverse of the average power consumption, obtain the power consumption index, obtain the average between the time index and the power consumption index, obtain the frequency weight, obtain the product of the frequency weight and the preset power consumption frequency, and obtain the data fusion frequency of any power consumption interval.

Images