CN104215838A - Remote nuclear phase method for intelligent substation - Google Patents

Abstract

The invention provides a remote nuclear phase method for an intelligent substation. The method comprises the following steps: locally acquiring and analyzing; recording the measured amplitude, phase and frequency information of the reference side; remotely acquiring and analyzing the measured amplitude, phase and frequency information of the side to be tested; compensating the error caused by remote measurement through the compensation algorithm, so as to reach the purpose of remote nuclear phase of the intelligent substation. With the adoption of the method, the problem that remote nuclear phase cannot be performed under the condition that related device cannot be normally put into operation at the initial period of construction of the intelligent substation can be solved; the nuclear phase testing efficiency of the intelligent substation can be obviously raised; meanwhile, the compensation algorithm is carried out to correct the measuring error, thus the measuring precision is raised, and the nuclear phase measuring result is accurate.

Description

An off-site nuclear phase method for smart substations

technical field

The invention relates to the technical field of testing of intelligent substations, in particular to a method for detecting remote phases of intelligent substations.

Background technique

Phase verification refers to using related equipment to check whether the phases and phase sequences of two power supplies or loops are the same. After the smart substation is newly built or rebuilt and expanded, a three-phase circuit verification phase test must be carried out before sending power to users to ensure that the phase sequence of the transmission line is consistent with the phase sequence required by the user's three-phase load.

The traditional nuclear phase method of smart substation mostly boosts the voltage on the primary side, and then checks the relevant measured phase on the protection device, oscilloscope or special nuclear phase device, so as to judge the phase difference between the two power sources. These methods have the following problems in the actual operation process:

1. In the initial stage of smart substation construction, the construction of the interval layer and process layer network in the station may not be completed, and the primary side data cannot be transmitted from the transformer to the protection device or wave recorder;

2. The transformers that collect the two power sources may be far apart on site, and it is difficult to monitor the two power signals at the same time when the network at the interval layer and the process layer of the smart substation is not working properly;

3. At present, the relevant standards of smart substations in China are not yet unified. The State Grid Corporation adopts a point-to-point data transmission mode, while the China Southern Power Grid Corporation adopts a networking method to transmit data. The data processing methods in the two modes are different, and the data cannot be processed in the same way;

4. The data sampling rate in the smart substation is only 4000Hz. For power frequency signals, there are only 80 sampling points for each cycle, and the amount of data is small. The accuracy of the data obtained directly by Fourier transform is very poor, especially when the signal output by the voltage or current source is not standard and the frequency is unstable, the error of the obtained data is even greater.

5. Due to the complexity of the power grid, the voltage frequency of the reference side and the side to be tested cannot always maintain the power frequency output, and the frequency on both sides will not be completely consistent. For a frequency deviation of 0.001Hz, an error of 0.001Hz*360°=0.36° will be generated in 1 second. After the time spent in the remote measurement process, this small error will cause a large measurement error.

It can be seen that the traditional smart substation nuclear phase method has high requirements on the on-site test environment, and the operability is not strong. When analyzing data, frequency tracking compensation is not done, and the function of remote phase nuclear phase cannot be well realized.

Contents of the invention

In order to solve the above-mentioned technical problems, the present invention provides a remote phase verification method for smart substations, which includes the following steps:

Step 1: Record the measured real-time data of the reference side and the side to be measured respectively. The currently running power grid is used as the reference side, and the part that needs to be connected to the grid is the side to be measured. Then the measured amplitude, phase, and frequency information;

Step 2: Compensate the off-site measurement error between the side to be measured and the measured side of the reference side;

Step 3: Calculate the measured phase difference between the side to be measured and the reference side, and display it in the form of graphs and tables.

In the above-mentioned method for remote phase verification of a smart substation, the measured real-time data recorded in step 1 is the electrical quantity in at least one of the following signals:

Signal A: the analog small signal output by the photoelectric current transformer;

Signal B: The standard analog signal output by the secondary side of the traditional electromagnetic current transformer;

Signal C: the optical digital signal of the collector output by the photoelectric current transformer;

Signal D: an optical digital SV sampling value signal in the format of IEC61850-9-1, IEC61850-9-2 or IEC60044-8 outputted by the merging unit MU.

According to claim 2, a kind of smart substation off-site nuclear phase method,

The data recording method of signals A and B is: introduce a standard clock source, collect analog data from the second, and record two cycles of data continuously;

The data recording method of signals C and D is as follows: starting from the receipt of the message whose sampling number is 0, the data is recorded, and the data of two cycles is continuously recorded.

In the remote phase verification method of an intelligent substation, the measured real-time data of the reference side recorded in step 1 is retained as the reference data, and the recorded real-time measured data of the untested side are updated in real time. There are several lines in a substation, and each line that needs to be connected to the grid is the side to be tested. On-site testing requires special equipment to apply voltage on the line. Generally, after one line is tested, the equipment is transferred to the next line to continue measurement.

In the above-mentioned method for phase verification of a smart substation, in step 1, the analysis of the collected real-time data includes: firstly performing Fourier transform on the original data, and then performing frequency compensation to obtain the amplitude, phase, and frequency exact value.

In the above-mentioned method for phase verification of a smart substation at a different location, in step 2, the method of compensating the measurement error between the voltage or current of the side to be measured and the reference side is serial number alignment or time scale alignment.

In the above-mentioned method for phase verification in different places of smart substations, the serial number alignment method is used to compensate the measurement error data and transfer them to step three for direct use.

The method for phase verification of a smart substation in a remote location includes the following steps:

Step 1: Find the time channel in the data of the reference side and the side to be tested, denoted as t, and the unit is s. If there is no time channel, the time of this side is 0;

Step 2: Obtain the frequency to be measured, denoted as f, and the unit is Hz;

Step 3: The measured phase compensation value Φ=360*t*f, the unit is °;

In the above described phase verification method of a smart substation, when compensating the phase difference, the angle difference caused by the voltage or current frequency deviation between the reference side and the side to be measured should be compensated. The compensation method is:

Step 1: The time when the reference side starts recording data for the last time is recorded as t1, and the unit is s, and the measured frequency obtained after analyzing the data is recorded as f1, and the unit is Hz;

Step 2: The moment when the side under test starts to record data is recorded as t2, and the unit is s;

Step 3: The measured phase compensation value Φ=(t2-t1)*f1*360, divide the obtained value by 360, and take the remainder.

In the remote phase verification method for smart substations, there is at least one measured object on the reference side and the measured side, and one or more measured objects on the tested side can be compared with one or more measured objects on the reference side.

The present invention has obtained following technical effect:

The invention can solve the problem that phase verification cannot be performed in different places when relevant equipment cannot be put into normal operation in the initial stage of intelligent substation construction, and can significantly improve the efficiency of phase verification of the smart substation. At the same time, through the compensation algorithm, the method corrects the measurement error, improves the measurement accuracy, and makes the nuclear phase measurement result more accurate.

Description of drawings

Figure 1 is a schematic diagram of the remote nuclear phase of the smart substation;

Fig. 2 is the schematic diagram when measuring the analog signal;

Fig. 3 is the schematic diagram when measuring the digital quantity signal.

Detailed ways

In order to facilitate the understanding and implementation of the present invention by ordinary counters in the field, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

The present invention provides a remote phase verification method for an intelligent substation, the principle of which is shown in Figure 1, including the following steps:

a) Record the measured real-time data on the reference side, and obtain the measured amplitude, phase and frequency information after analysis;

b) Record the measured real-time data of the side to be tested, and obtain the measured amplitude, phase and frequency information after analysis;

c) Compensate the off-site measurement error between the measured side and the reference side;

d) Calculate and display the measured phase difference between the side to be measured and the reference side in the form of graphs and tables.

In addition, before connecting to the grid, it is necessary to check whether the voltage phases on both sides are the same. If the phases are different, it cannot be connected to the grid, and the wiring of the equipment needs to be checked.

The measured signal includes: the small analog signal output by the photoelectric voltage and current transformer; the standard analog signal output by the secondary side of the traditional electromagnetic voltage and current transformer; the optical digital signal of the collector output by the photoelectric voltage and current transformer; Output optical digital SV sampling value signal in IEC61850-9-1, IEC61850-9-2 or IEC60044-8 format.

When collecting the small analog signal output by the photoelectric voltage and current transformer, or the standard analog signal output by the secondary side of the traditional electromagnetic voltage and current transformer, it is necessary to introduce a standard clock source. clock source. Start to collect analog data from the whole second, and record the data of two cycles continuously. The sampling rate should not be too low, at least 4000Hz.

When collecting the optical digital signal of the collector output by the photoelectric voltage and current transformer, or the optical digital SV sampling value signal output by the merging unit MU in the format of IEC61850-9-1, IEC61850-9-2 or IEC60044-8, use Figure 3 In the shown method, the data is recorded starting from the receipt of the message whose sampling sequence number is 0, and the data of two cycles are continuously recorded.

During the test, first record the measured real-time data on the reference side, and save the measured amplitude, phase, and frequency information. Then, the measured data is updated in real time on the side to be tested, and compared with the data on the reference side.

When calculating the amplitude, phase, and frequency through the recorded real-time data, first obtain the measured amplitude and phase through Fourier transform, then obtain the frequency value through the frequency tracking compensation algorithm, and finally use the frequency value to correct the amplitude and phase. So as to get the precise value of amplitude, phase and frequency. This calculation method of amplitude, phase and frequency is applicable to the measured calculation process of the reference side and the measured side at the same time.

After the above process, the measured amplitude, phase, and frequency information of the reference side and the measured side can be obtained. After obtaining the measured amplitude, phase, and frequency values of the reference side and the measured side, it is necessary to compensate for the operation in different places. caused by the error. The compensation methods include number alignment compensation and time scale alignment compensation. When the serial number compensation method is used, the original data can be directly used for calculation. When using the time scale alignment compensation method, it is necessary to compensate the phase difference caused by the sampling delay at the source. The specific steps are: save the measured information of the reference side, and record the last time when the measured value of the reference side starts to be recorded, denoted as t1, and the unit is s (second); record the measured frequency of the last measurement of the reference side, denoted as f1, the unit is Hz (Hertz). The measurand on the side to be tested is refreshed once per second, compared with the data on the reference side, and the phase difference is obtained after compensation. Find the time channel in the data of the reference side and the side to be measured, denoted as t, and the unit is s. If there is no time channel, the time of this side is 0; obtain the measured frequency, denoted as f, and the unit is Hz (Hertz) ; The measured phase compensation value, Φ=360*t*f, the unit is ° (degree).

After compensating the error caused by the difference in sampling delay between the reference side and the side to be tested, the error caused by the difference in operating frequency between the system on the reference side and the side to be tested must be compensated. The specific compensation method is: record the last time the reference side started recording The moment of the data is recorded as t1, and the unit is s (second), and the measured frequency obtained after analyzing the data is recorded as f1, and the unit is Hz (Hertz); record the moment when the side to be tested starts to record data as t2, and the unit is s (seconds); the measured phase compensation value Φ=(t2-t1)*f1*360, divide the obtained value by 360, and take the remainder.

Two errors need to be compensated for off-site nuclear phase, one is the error caused by the asynchronous acquisition method of the data source; the other is the error caused by the difference in system frequency between the reference side and the side to be measured. When compensating for errors caused by different data source acquisition methods, and because the current domestic smart substation sampling data transmission methods are different, there are two compensation methods: serial number alignment and time scale alignment. These two alignment methods are commonly used at present, and the protection manufacturer also handles them on this basis. Because there will be a certain delay in the process of the actual voltage and current from being collected to the output data of the merging unit, this requires us to restore the data to the moment of its collection after receiving the data, and then perform subsequent analysis, otherwise it will be different Comparing the data at any time loses the meaning of comparison.

The merging unit collects voltage and current signals at a fixed sampling rate, and transmits the data to protection and other equipment in the form of digital messages. When the protection receives the message, it needs to know the data collection time. There are two ways to obtain the data collection time. One way is to judge the collection time according to the serial number in the message. For example, the merging unit collects data at a sampling rate of 4000 Hz, and the sampling time interval is 250 microseconds. , there will be a unique serial number corresponding to it. The sampling serial number is 0 at the whole second, and then accumulated sequentially. At the beginning of the next second, the serial number is reset to zero. The serial number is cycled like this. Multiply the serial number by the sampling time interval to get the The sampling time of the sampling point in the current second. Sequence number alignment is based on this principle. It is believed that packets with the same sampling sequence number have the same collection time; another way is to record the time of receipt after receiving the packet, and then delay the sampling according to the mark in the packet. Estimate the time of data collection. In the ideal case where the primary side voltage is a power frequency of 50Hz, the data of the nth sampling point of this second is the same as the data of the nth sampling point of the next second, so the part above the second can be ignored and passed directly The above two methods obtain the time below the second, and then compare the data at the same time.

In the smart substation that transmits data in the sampling network mode, the transformers and the devices in the station are strictly timed, and the labels in the SV messages actually mark the time of data collection. In this mode, the serial number alignment method is adopted, and the data on the reference side and the side to be tested do not need to be specially processed.

In the smart substation with sampling direct sampling and direct jumping technology, the transformers do not need to be timed, and the devices in the station can also operate normally. The device obtains the original data acquisition time by subtracting the transmission delay in the message from the time when the message is received, and in this way achieves the effect of multi-channel sampling synchronization. In this mode the way the timescales are aligned should be sampled. The time channel value in the measured message is recorded as t2, and the unit is s (second); the measured frequency is recorded as f2, and the unit is Hz (Hz), and the measured phase compensation value Φ=360*t*f, the unit is ° (Spend).

When compensating the error caused by the system frequency difference between the reference side and the side to be measured, the moment when the side to be measured starts to record the measured data is recorded as t3, the unit is s (second), and the measured phase compensation value Φ=(t3-t1) *f1*360, divide the obtained value by 360, and take the remainder.

Through two compensations, the error in the process of exogeus nuclear phase can basically be eliminated. Finally, the measured results are displayed graphically. When displayed graphically, it is displayed in the form of a commonly used phasor diagram, with the reference side data as the background, and the three-phase data are separated by yellow, green and red colors; The angular difference and the magnitude difference between the side measurands. When displayed in table form, it is displayed in the form of amplitude + phase angle. The measured amplitude and phase information of the reference side and the side to be tested are displayed in two screens, and the calculation results of the same-phase and phase-to-phase phasor differences are provided. for debuggers to analyze.

During the test process, under the condition that the reference side remains unchanged, after testing the data on one side, the test equipment can be moved to the next side to continue measuring the data on this side, which improves the work efficiency during the test process.

Claims (10)

1. an intelligent substation strange land nuclear-phase method, is characterized in that, comprises the following steps:

Step one: record the measured real time data of reference side and side to be measured respectively, wherein with the electrical network of current operation for reference side, need grid-connected part to be side to be measured, then obtain measured amplitude, phase place, frequency information respectively through analysis;

Step 2: compensate side to be measured and reference side measured between strange land measuring error;

Step 3: calculate phase differential measured between side to be measured and reference side, and show with the form of figure and form.

2. a kind of intelligent substation strange land according to claim 1 nuclear-phase method, is characterized in that: recording measured real time data in described step one is electric parameters in following at least one signal:

Signal A: the simulation small-signal that For The Rogowski Optical Current Transformer exports;

Signal B: the standard analog signal that conventional electromagnetic Current Transformer Secondary side exports;

Signal C: the collector light digital signal that For The Rogowski Optical Current Transformer exports;

Signal D: through the light numeral SV sampled value signal of IEC61850-9-1, IEC61850-9-2 or IEC60044-8 form that merge cells MU exports.

3. a kind of intelligent substation strange land according to claim 2 nuclear-phase method, is characterized in that:

The data recording fashion of signal A and B is: introduce standard clock source, from whole second, gather analog data, continuously the data of record two cycles;

The data recording fashion of signal C and D is: from receiving the start of message (SOM) that sampling sequence number is 0, record data, continuously the data of record two cycles.

4. a kind of intelligent substation strange land according to claim 1 nuclear-phase method, it is characterized in that: the measured real time data of the reference side recorded in step one is retained as reference data, the measured real time data in the side to be measured of recording is real-time update.

5. a kind of intelligent substation strange land according to claim 1 nuclear-phase method, it is characterized in that: in step one, the analysis of gathered real time data is comprised: first by raw data through Fourier transform, then carry out frequency compensation, to obtain the exact numerical of amplitude, phase place, frequency.

6. a kind of intelligent substation strange land according to claim 1 nuclear-phase method, is characterized in that: in step 2, and compensating side to be measured with the method for the measuring error between reference side voltage or electric current is that sequence number is alignd or markers is alignd.

7. a kind of intelligent substation strange land according to claim 6 nuclear-phase method, is characterized in that: the data adopting sequence number alignment thereof to compensate measuring error turn and directly used by step 3.

8. a kind of intelligent substation strange land according to claim 6 nuclear-phase method, is characterized in that: the method adopting markers alignment to compensate measuring error comprises following steps:

Step 1: find out the time channel in reference side and side data to be measured, be designated as t, unit is s, if not free passage, then this side time is 0;

Step 2: obtain measured frequency, be designated as f, unit is Hz;

Step 3: measured phase compensation value Φ=360*t*f, unit is °.

9. a kind of intelligent substation strange land according to claim 1 nuclear-phase method, is characterized in that: during compensation of phase difference, compensate the differential seat angle caused by the frequency departure of reference side and side to be measured voltage or electric current, compensation method is:

Step 1: the moment that reference side starts to record data is for the last time designated as t1, unit is s, and the measured frequency obtained after analyzing data is designated as f1, and unit is Hz;

Step 2: the side to be measured current moment starting to record data is designated as t2, and unit is s;

Step 3: measured phase compensation value Φ=(t2-t1) * f1*360, the value obtained divided by 360, remainder number.

10. intelligent substation strange land according to claim 1 nuclear-phase method, is characterized in that: reference side and the measured of side to be measured are at least one, and the one or more measured of side to be measured can compare with the one or more measured of reference side.