阅读说明：本技术 一种基于工频变化趋势的低压户变关系识别方法及系统 (Low-voltage user variation relation identification method and system based on power frequency variation trend ) 是由 林顺达 何本亮 于 2020-12-25 设计创作，主要内容包括：本发明提供一种基于工频变化趋势的低压户变关系识别方法及系统,属于低压电力线载波通讯设备数据分析领域。本发明方法包括如下步骤：从模块接收主模块数据,所述数据包括正弦电压正负半周过零点时刻以及三相过零点采集数据；定位采集时刻,然后采集该时刻的过零点数据；提取主模块数据中与自身采集的过零点数据相对应的相位过零点数据；计算主从模块该相位过零点数据的工频差及工频变化趋势；据工频差或工频变化趋势获取归属网络；将识别结果上报给主模块。本发明的有益效果为：智能自动化识别,覆盖整个用电网络,实现一对多的全面识别效果,识别精度高。(The invention provides a low-voltage user variable relation identification method and system based on power frequency variation trend, and belongs to the field of data analysis of low-voltage power line carrier communication equipment. The method comprises the following steps: the slave module receives the data of the master module, wherein the data comprises the zero crossing time of positive and negative half cycles of the sinusoidal voltage and three-phase zero crossing acquisition data; positioning the acquisition time, and then acquiring zero crossing point data of the time; extracting phase zero-crossing data corresponding to self-acquired zero-crossing data in the main module data; calculating the power frequency difference and power frequency variation trend of the phase zero-crossing data of the master module and the slave module; acquiring a home network according to the power frequency difference or the power frequency variation trend; and reporting the identification result to the main module. The invention has the beneficial effects that: intelligent automatic identification covers whole power consumption network, realizes the comprehensive identification effect of one-to-many, and the discernment precision is high.)

1. A low-voltage user variation relation identification method based on power frequency variation trend is characterized by comprising the following steps:

the method comprises the following steps that firstly, a slave module receives master module data, wherein the data comprise zero crossing time of positive and negative half cycles of sinusoidal voltage and three-phase zero crossing acquisition data;

positioning the acquisition time, and then acquiring zero crossing point data of the time;

step three, judging whether the phase identification is finished, if so, executing step four, if not, positioning the phase to which the phase belongs, and then returning to execute step one;

extracting phase zero-crossing data corresponding to the self-collected zero-crossing data in the main module data;

calculating the power frequency difference and power frequency variation trend of the phase zero-crossing data of the master module and the slave module;

step six, acquiring a home network according to the power frequency difference or the power frequency change trend;

and step seven, reporting the identification result to the main module.

2. The low-voltage user variation relation identification method based on power frequency variation trend as claimed in claim 1, wherein: after the sixth step is executed, the method also comprises a judging step: and C, judging whether the network data volume is more than a set value m, if so, executing the step seven, otherwise, returning to the step one without reference to the acquired home network.

3. The low-voltage user variation relation identification method based on power frequency variation trend according to claim 1 or 2, characterized in that: in the first step, the slave module and the master module adopt a continuous identification mode, and the master module and the slave module are both provided with broadband carrier communication modules and adopt broadband carrier periodic communication.

4. The low-voltage user variation relation identification method based on power frequency variation trend according to claim 1 or 2, characterized in that: and step two, positioning the acquisition time for synchronizing the acquisition time, wherein the synchronization is not needed when the master module and the slave module are in the same network, otherwise, the acquisition time of the master network is converted into the acquisition time relative to the master network.

5. The low-voltage user variation relation identification method based on power frequency variation trend as claimed in claim 4, wherein: before the zero-crossing point data is collected, a collection time adjusting step is further included, the collection time after positioning is moved forward for a plurality of times, the time Tfuture after moving forward is obtained, and the collection is started by using the time, so that the zero-crossing point data at the same time can be accurately collected.

6. The low-voltage user variation relation identification method based on power frequency variation trend as claimed in claim 5, wherein: in the fifth step, the work frequency difference calculation method comprises the following steps:

accurately extracting phase zero-crossing data Fm corresponding to self-collected zero-crossing data Fs in the master module data sequence by the slave module, wherein the number N of data points is the same, calculating the average of the zero-crossing data of the master module and the slave module, and the difference value is reference power frequency difference Fdiff, wherein the calculation formula of the reference power frequency difference is as follows:

wherein i is a data point, i is more than or equal to 0 and less than or equal to N,

the method for calculating the power frequency variation trend Cdiff comprises the following steps:

7. the low-voltage user variation relation identification method based on power frequency variation trend as claimed in claim 6, wherein: and sixthly, firstly judging whether the power frequency difference is not greater than a coefficient K1, if so, increasing the proportion of a weight r1, then executing a step seven, if not, judging that the difference between power frequency change values of the master module and the slave module at the same time point is not greater than a coefficient K2, if not, directly executing the step seven, if so, increasing the proportion of a weight r2, and then executing the step seven, wherein K1 is less than K2, and r1 is greater than r 2.

8. The low-voltage user variation relation identification method based on power frequency variation trend as claimed in claim 7, wherein: and obtaining network similarity R according to the reference power frequency difference, the power frequency change trend and the corresponding weight, sequencing the network similarity R, and selecting the network with the maximum R value as the network of the current slave module.

9. A system for realizing the power frequency change trend-based low-user-variation relationship identification method according to any one of claims 1 to 8 is characterized by comprising the following steps:

a receiving module: the slave module is used for receiving the master module data, and the data comprises positive and negative half-cycle zero-crossing time of sinusoidal voltage and three-phase zero-crossing acquisition data;

a positioning and acquisition module: the device is used for positioning the acquisition time and then acquiring the zero crossing point data of the time;

a first judgment module: used for judging whether the phase identification is finished;

a phase positioning module: used for positioning the phase;

an extraction module: the phase zero-crossing data extraction module is used for extracting phase zero-crossing data corresponding to self-acquired zero-crossing data in the main module data;

a calculation module: the power frequency difference and power frequency variation trend of the phase zero-crossing data of the master module and the slave module are calculated;

an acquisition module: the home network is obtained according to the power frequency difference or the power frequency change trend;

a reporting module: for reporting the recognition result to the main module.

10. The system of claim 9, wherein: the device also comprises a second judgment module: and the method is used for judging whether the network data volume is more than a set value m.

Technical Field

The invention relates to a data analysis technology of low-voltage power line carrier communication equipment, in particular to a low-voltage user variation relation identification method and system based on power frequency variation trend.

Background

In the operation management of the low-voltage distribution network, accurate establishment of the transformer area outdoor variable relationship is a necessary measure for realizing the requirements of fine management and consumption reduction, and a device for developing general survey of the distribution transformer area is indispensable, can be used for clearly surveying information of a transformer, a phase line and the like belonging to a power user, and provides accurate data for a marketing management system. The data are beneficial to analysis of line loss of the transformer area, balance of power utilization load of the transformer area is guaranteed, economic operation level of a power grid is improved, and therefore construction of the smart power grid is achieved.

The instruments or methods put into use in the market at present are mainly a station area identification instrument based on voltage waveform distortion and an identification method based on a signal-to-noise ratio (SNR) algorithm. The transformer area identification instrument monitors the voltage waveform change of the master node and the slave node at the same time through a point-to-point detection mode triggered by external force to obtain a user variable relation. The identification mode has great limitation, an operator needs to go to the site to search the positions of the transformer and the target node of the transformer area, long time is consumed for the round trip if the distance between the transformer and the target node is long, and the identification mode has large volume and mass and is inconvenient to carry. In addition, the transformer area identification instrument requires that operating personnel have higher operation experience, is difficult to learn and use, has potential safety hazards of electricity utilization and higher hardware cost, and the most fundamental problem of the device is that the identification mode only supports point-to-point identification, and the identification efficiency is low.

The identification method based on the SNR algorithm has the advantages that the defects are obvious, due to the fact that planning of laying of an original power line is not uniform, a physical line is complicated, space coupling of an electromagnetic field is easily caused, the condition of a common zero line between transformers in a transformer area exists, line impedance is different due to different user loads in different periods of a low-voltage power grid, the generated noise intensity is different, carrier signals are easily seriously attenuated, the identification method based on the SNR algorithm is inaccurate in identification and large in limitation, in addition, the SNR algorithm does not support phase line attribution judgment, and the function is single.

Disclosure of Invention

In order to solve the problems of unstable identification and low accuracy of the existing instrument or method, the invention provides a low-voltage user variable relation identification method based on power frequency variation trend and a system for realizing the low-voltage user variable relation identification method.

The invention relates to a low-voltage user variable relation identification method based on power frequency variation trend, which comprises the following steps:

the method comprises the following steps that firstly, a slave module receives master module data, wherein the data comprise zero crossing time of positive and negative half cycles of sinusoidal voltage and three-phase zero crossing acquisition data;

positioning the acquisition time, and then acquiring zero crossing point data of the time;

step three, judging whether the phase identification is finished, if so, executing step four, if not, positioning the phase to which the phase belongs, and then returning to execute step one;

extracting phase zero-crossing data corresponding to the self-collected zero-crossing data in the main module data;

calculating the power frequency difference and power frequency variation trend of the phase zero-crossing data of the master module and the slave module;

step six, acquiring a home network according to the power frequency difference or the power frequency change trend;

and step seven, reporting the identification result to the main module.

The invention is further improved, after the step six is executed, the method also comprises a judging step: and C, judging whether the network data volume is more than a set value m, if so, executing the step seven, otherwise, returning to the step one without reference to the acquired home network.

The invention is further improved, in the first step, the slave module and the master module adopt a continuous identification mode, and the master module and the slave module are both provided with a broadband carrier communication module and adopt broadband carrier periodic communication.

The invention is further improved, in the second step, the positioning acquisition time is used for synchronizing the acquisition time, when the master module and the slave module are in the same network, the synchronization is not needed, otherwise, the acquisition time of the master network is converted into the acquisition time relative to the master network.

The invention is further improved as follows: before the zero-crossing point data is collected, a collection time adjusting step is further included, the collection time after positioning is moved forward for a plurality of times, the time Tfuture after moving forward is obtained, and the collection is started by using the time, so that the zero-crossing point data at the same time can be accurately collected.

The invention is further improved, and in the fifth step, the method for calculating the work frequency difference comprises the following steps:

accurately extracting phase zero-crossing data Fm corresponding to self-collected zero-crossing data Fs in the master module data sequence by the slave module, wherein the number N of data points is the same, calculating the average of the zero-crossing data of the master module and the slave module, and the difference value is reference power frequency difference Fdiff, wherein the calculation formula of the reference power frequency difference is as follows:

wherein i is a data point, i is more than or equal to 0 and less than or equal to N,

the method for calculating the power frequency variation trend Cdiff comprises the following steps:

Cdiff(i,0≤i≤N)＝Cs(i)-Cm(i)

Cs(i,0≤i≤N)＝Fs(i+1)-Fs(i)。

Cm(i,0≤i≤N)＝Fm(i+1)-Fm(i)

the invention is further improved, in the sixth step, judge whether the power frequency difference is not greater than coefficient K1 at first, if yes, increase the proportion of weight r1, then carry out the seventh step, if no, judge that the difference of power frequency change value of master slave module at the same time point is not greater than coefficient K2, if no, carry out the seventh step directly, if yes, increase the proportion of weight r2, then carry out the seventh step, wherein, K1< K2, and r1> r 2.

The invention is further improved, the network similarity R is obtained according to the reference power frequency difference, the power frequency change trend and the corresponding weight, the network similarity R is sequenced, and the network with the maximum R value is selected as the network of the current slave module.

The invention also provides a system for realizing the low-voltage user variation relation identification method based on the power frequency variation trend, which is characterized by comprising the following steps of:

a receiving module: the slave module is used for receiving the master module data, and the data comprises positive and negative half-cycle zero-crossing time of sinusoidal voltage and three-phase zero-crossing acquisition data;

a positioning and acquisition module: the device is used for positioning the acquisition time and then acquiring the zero crossing point data of the time;

a first judgment module: used for judging whether the phase identification is finished;

a phase positioning module: used for positioning the phase;

an extraction module: the phase zero-crossing data extraction module is used for extracting phase zero-crossing data corresponding to self-acquired zero-crossing data in the main module data;

a calculation module: the power frequency difference and power frequency variation trend of the phase zero-crossing data of the master module and the slave module are calculated;

an acquisition module: the home network is obtained according to the power frequency difference or the power frequency change trend;

a reporting module: for reporting the recognition result to the main module.

The invention is further improved, and the invention also comprises a second judgment module: and the method is used for judging whether the network data volume is more than a set value m.

Compared with the prior art, the invention has the beneficial effects that: the existing high-speed broadband carrier communication equipment of HPLC in the low-voltage distribution network is utilized, and the identification method of the invention is matched, so that the manual operation mode is converted into intelligent automatic identification, the whole power utilization network is covered, the one-to-many comprehensive identification effect is realized, and the problems of unstable identification and low accuracy of the existing instrument or method are solved.

Drawings

FIG. 1 is a flow chart of the method of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples.

The invention provides a technical scheme for realizing automatic identification of the user variable relationship by using an HPLC high-speed broadband carrier communication device by utilizing the existing low-voltage power centralized meter reading system in order to overcome the defects of low identification accuracy, high economic cost and low identification efficiency in the prior art.

The invention utilizes the low-voltage power centralized meter reading system channel to transmit the start-stop operation and the identification result of the household change identification command through the power utilization acquisition system or the concentrator, and the core method is mainly embodied in the interoperation between the master module and the slave module of the HPLC broadband carrier communication equipment. The main module sends the latest three-phase alternating current zero crossing point time and a plurality of groups of data, namely the difference value between the zero crossing point time and the last zero crossing point time and the difference value of a 20ms theoretical period, which are acquired by the main module in a broadcasting mode, the slave module acquires single-phase zero crossing point data of relative time according to the three-phase alternating current zero crossing point time and the data provided by the main module, converts the data into working frequency and then compares the working frequency with the frequency variation trend of the three-phase alternating current zero crossing point data and the reference frequency, so that the similarity of physical line characteristics is obtained and a record is formed, the attribution result is obtained after a certain data amount is met, and the main module is actively informed or inquired the. The present invention will be described in detail below.

As shown in fig. 1, the method for identifying a low-voltage user variation relationship based on a power frequency variation trend of the present invention includes the following steps:

1. after the main module starts a user change identification command, three-phase zero crossing point acquisition data of positive and negative half cycle zero crossing points of sinusoidal voltage, namely M, L1, L2 and L3, are sent to the network in a broadcasting mode according to the arrangement sequence shown in Table 1. The initial acquisition time represents the zero crossing point acquisition time under the counting frequency of 25MHz, and each 20ms is 500000 counting units; the positive and negative half cycle collected data Tn represents the deviation of the time length of a zero-crossing cycle of Lx phase and an ideal cycle of 20ms, and adopts HEX format, signed integer number and two-byte representation, and the counting frequency is 8 frequency division of 25MHz, namely 3.125 MHz.

Field(s)	Size (byte)
		Negative half cycle start acquisition time M1	4
Number of negative half-cycle L1 acquisitions	1
		Number of negative half-cycle L2 acquisitions	1
Number of negative half-cycle L3 acquisitions	1
		Negative half cycle T1 (data 1.)	2
Negative half cycle T2 (data 2)	2
		…	2
Negative half cycle Tn (nth data)	2
		Negative half cycle L2 phase T1-Tn	-
Negative half cycle L3 phase T1-Tn	-
		Positive half cycle start acquisition time M2	4
Number of positive half-cycle L1 acquisitions	1
		Number of positive half-cycle L2 acquisitions	1
Number of positive half-cycle L3 acquisitions	1
		Positive half cycle T1 (data No. 1)	2
Positive half cycle T2 (data 2)	2
		…	2
Positive half cycle Tn (nth data)	2
		Positive half cycle L2 phase T1-Tn	-
Positive half cycle L3 phase T1-Tn	-

TABLE 1 arrangement format of zero crossing data of voltage sine wave

Of course, other data formats can be adopted in the embodiment, and data acquisition only needs to be carried out from the zero-crossing time of positive and negative half cycles of the sinusoidal voltage and the three-phase zero-crossing point which can be identified by the module.

2. And when the slave module receives the data acquired by the zero crossing point of the master module, positioning the acquisition time, wherein the step relates to the synchronization of the acquisition time. If the master module and the slave module are in the same network, synchronization is not needed, otherwise, the acquisition time of the other network needs to be converted into the acquisition time relative to the network. The slave module can calculate the difference in the network reference time between the networks, in one beacon per second, which is a fixed value, without taking into account the error, with which the "negative half-cycle start acquisition time M1" in table 1 can be converted into the relative time of acquisition from the module itself.

The network reference time occupies 4 bytes, and the value range is as follows: 0 to 232-1, i.e., 0 to 4294967295. The time is circulated under the counting frequency of 25MHz, and the network reference time of the other network is set as T1, the network reference time of the network is set as T2, the difference value of the two is Tdiff, and the negative half-cycle initial acquisition time M1 is Tnow.

The specific calculation formula is as follows:

when T1> T2:

Tdiff＝T1-T2；

if Tnow > -Tdiff, Tnow-Tdiff;

if Tnow < Tdiff, Tnow ═ 0 xfffffffff- (Tdiff-Tnow);

when T1< ═ T2:

Tdiff＝T2-T1；

Tnow＝Tnow+Tdiff。

3. after the calculation, the "negative half cycle start acquisition time M1" Tnow is already the time relative to the local network, and because crystal oscillators in different hardware individuals have errors, zero crossing point acquisition directly using the "negative half cycle start acquisition time M1" Tnow may randomly occur to obtain zero crossing point data or acquire the next zero crossing point data, so the acquisition time needs to be moved forward by a plurality of times to start searching, the value of the forward movement time is usually slightly larger than the maximum deviation of the crystal oscillators of the master and slave modules, but should be smaller than 10ms and not be close to the maximum deviation, the forward movement time in this example is 555us, the time tfont after the forward movement is obtained, and the tfont is Tnow- (62500/360 × 10), and zero crossing point data at the same time can be accurately acquired by using the time.

4. After the first data acquisition, if the phase is not identified yet, it is first determined which phase the slave module is in relative to the master module. Since the master module data format is acquired according to the sequence of L1-L2-L3, and the slave module only has one-phase zero-crossing acquisition, the phase and fire-zero reverse connection condition can be known by comparing the self acquisition time Tself with the negative half cycle starting acquisition time M1 Tnow. The phase L2 is 0 degrees relative to the phase L1, 120 degrees relative to the phase L3, and the complementary angle is opposite. Taking the reference angle as the center and plus or minus 30 degrees as the belonging area, and each phase (including the positive phase and the negative phase) occupies 60 degrees. The phase angle is calculated as:

Tc_diff＝Tself-Tnow；

angle＝Tc_diff(us)*360/20000(us)。

5. after the phase identification step is completed, the slave module accurately extracts phase zero crossing data Fm corresponding to the self-collected zero crossing data Fs in the master module data sequence, the number of data points N is the same, the average of the zero crossing data of the master module and the slave module is calculated, and the difference value is the reference power frequency difference Fdiff.

The calculation formula of the reference power frequency difference is as follows:

wherein i is a data point, i is more than or equal to 0 and less than or equal to N,

the method for calculating the power frequency variation trend Cdiff comprises the following steps:

if Fdiff is less than the factor K1, meaning that the current slave approaches the master network sending the data more, then the weight r1 is added to the network, otherwise the trend of change Cdiff needs to be compared, if the Cdiff (x) at individual moments in a set of trends of change is less than the factor K2, meaning that the current slave has a partial probability of approaching the master network sending the data, then the weight r2 is added to the network, typically K1< K2, and r1> r 2.

6. When the network similarity R obtained by the conditions of the reference power frequency difference and the power frequency change trend in the step 5 is stored, the judgment step is added, whether the data volume of the current network similarity R meets the set value m or not is judged, if the data volume of the current network similarity R is smaller than the set value m, the data volume is less, the data volume is not suggested to be used as a reference of the user variable identification relationship, when the data volume is larger than or equal to m, the data volume is sorted according to the network similarity R, the network with the largest R value is determined as the network to which the current slave module belongs, and the identification result is informed to the master module of the network where the slave module belongs.

The invention has three important innovation points, so that the identification method of the invention has very high accuracy, which is as follows:

1. location of acquisition point time

The slave module positions self zero-crossing data according to the zero-crossing acquisition time of the master module, requires the uniformity of the acquisition time of the master module and the slave module, and avoids invalid comparison of subsequent data. The invention is characterized in that the network reference time difference between different networks is a fixed value, the relative time of other networks relative to the network can be obtained, and the crystal oscillators in different hardware individuals have errors, so that the acquisition time needs to be shifted forward by a plurality of times to start searching, thereby accurately acquiring the zero crossing point data of the same time.

2. Identification of associated phases

The master module broadcasts the positive and negative phase data of the three phases to the network, and since the target slave node can appear on any one of the three phases, after the acquisition point time is determined, the acquired data needs to be compared with the data of the corresponding phase of the master module, otherwise, the comparison is still invalid.

Since the phase difference among the three phases of L1, L2, and L3 is 120 °, each sine wave period is 20ms, the sine wave angle is 360 °, the phase is divided into 6 regions by angle calculation, the phase of L1 is set as relative 0 °, plus or minus 30 degrees is assigned to the range of L1 phase in consideration of error, and the complementary angle region is set as opposite phase. Similarly, L2 is 120 degrees, and L3 is 240 degrees. The slave module compares the acquisition time with the determined phase time of the master module L1, and can obtain the time difference which is converted into an angle to be accurately determined as the phase line and whether the phase line is reversed, so that the required partial data is extracted from all the data of the master module for analysis.

3. Analysis of power frequency variations

After obtaining the zero-crossing data of the same phase at the same moment, by utilizing the reciprocal relation between the frequency F and the time T, the main module and the slave module respectively reversely reduce the data of the difference value between the zero-crossing point and the last zero-crossing point time and the difference value of the theoretical period of 20ms into the frequency, the average frequency of a plurality of groups of frequencies is taken as the reference frequency, the power frequency difference between the main frequency and the slave module and the change trend of the frequency in continuous time are recorded, when the reference power frequency difference Fdiff of the master module and the slave module at the same moment is less than or equal to a certain coefficient K1, the proportion of the weight r1 belonging to the local area is increased, when the reference power frequency difference Fdiff is greater than the coefficient K1, the similarity of the change trends of the master module and the slave module is compared, and when the difference of the 'change value of an acquisition point' of the two points. And when the data volume of different networks and the network reaches a certain degree m, the network with the highest weight proportion in the local area is regarded as the network. Thus, the identification method of the invention has higher precision.

Compared with the existing instruments or methods, the invention utilizes the existing HPLC high-speed broadband carrier communication equipment in the low-voltage distribution network and is matched with the identification method of the invention, and the invention has the following three advantages:

1. intelligent automation

The method adopts a continuous identification mode, the broadband carrier master-slave modules periodically communicate in the whole process, the slave modules automatically identify the station change to which the slave modules belong, the master module informs the judgment result to the master module after the identification is finished, the slave modules which do not belong to the station area are screened out by the master module, and the encapsulated data format is reported to an acquisition system.

2. Accuracy of recognition

For the situations that the power line physical line is complicated to cause space coupling of an electromagnetic field and a common zero line exists between transformers in a transformer area, crosstalk easily occurs in an HPLC high-speed broadband carrier signal to cause misjudgment of a user variable relationship. Aiming at the problem, a power frequency change analysis algorithm is a user variable identification method obtained by utilizing the relation between the power load and the power frequency. Because the instantaneous active power demand of different transformers at the same moment is different due to different power loads, the instantaneous rotating speed of the generator set is reduced or increased, and the instantaneous rotating speed is different at the same moment. On the contrary, the power frequency of the physical line of the same transformer is unique at the same time, and on the basis, a high-precision counting means of a high-speed carrier communication module of 25MHz is matched, data are periodically collected, and the similarity of the line characteristics between the target node and each transformer is analyzed by multiple sections of data, so that the accuracy of identification of the user variable is improved.

3. Convenience and safety

The invention combines the core power frequency change analysis algorithm on the basis of the prior high-speed broadband HPLC carrier communication equipment, effectively reduces the use of other instruments, reduces the labor cost, improves the accuracy, and eliminates the potential safety hazard of power utilization.

The above-described embodiments are intended to be illustrative, and not restrictive, of the invention, and all changes that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.