A signal detection device and detection method for primary and secondary deep fusion equipment

Abstract

The present invention discloses a signal detection device and detection method for a primary and secondary deep fusion device. The detection device includes a switching box, and the device to be tested is connected to the switching box; the transformer to be tested is connected to the switching box through a signal conversion device; the circuit breaker to be tested, the terminal to be tested and the complete set of equipment to be tested are directly connected to the switching box; a high-speed recorder and a multi-function meter are connected to the switching box; the switching box includes a switching switch circuit, and the switching switch circuit is connected to an output control circuit; the output control circuit outputs the secondary signal to be tested, and is connected to the control center of the test master station through a control interface; the test master station compares and calculates the waveform similarity collected by the recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested. The detection device of the present invention can meet the fault detection of each component of the primary switch / circuit breaker, secondary terminal, transformer / sensor of the primary and secondary fusion equipment of the power distribution.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution networks, and in particular to a signal sampling and waveform comparison detection method for primary and secondary deep fusion equipment. Background Art

[0002] At present, there is no solution that can be used as a reference for testing the primary and secondary integration equipment of the distribution network. At the same time, the State Grid has put forward the requirements for the deep integration of primary and secondary equipment for power distribution, and has put forward relevant solutions for the deep integration of primary and secondary equipment for power distribution. However, the detection technology of primary and secondary deep integration equipment still has the following problems: First, it is cumbersome to change the wiring when switching the detection items of the complete set of primary and secondary integration equipment, which requires a lot of manpower and poses safety risks; second, there is a lack of automatic detection technology and equipment that is compatible with primary switches / circuit breakers, secondary terminals, transformers and primary and secondary integration equipment. The detection requires split wiring and transportation, and assembly is required after the detection. The detection work is cumbersome and the efficiency is low; third, there is a lack of automatic detection technology and equipment that is compatible with electromagnetic transformers and electronic transformers; fourth, the primary and secondary integration equipment generally uses low-voltage / current signal low-power transformers, which have high requirements for impedance matching of the detection wiring, affecting the detection accuracy and detection efficiency. Summary of the invention

[0003] The technical problem to be solved by the invention is that in the distribution network, when switching detection items of the primary and secondary integrated equipment, rewiring is cumbersome, requires a lot of manpower, and poses safety risks.

[0004] In order to solve the above technical problems, the present invention provides

[0005] A primary and secondary deep fusion device signal detection device, comprising:

[0006] Switch box;

[0007] Connecting the equipment to be tested to the switch box, the equipment to be tested includes a transformer to be tested, a circuit breaker to be tested, a secondary terminal to be tested and a complete set of equipment to be tested;

[0008] The transformer to be tested is connected to the switch box through a signal conversion device; the circuit breaker to be tested, the terminal to be tested and the complete set of equipment to be tested are directly connected to the switch box; the high-speed recorder and the multi-function meter are connected to the switch box;

[0009] The transformer to be tested includes an electromagnetic transformer and an electronic transformer;

[0010] The switching box includes a switching switch circuit, which is connected to the output control circuit and is used to complete the wiring switching of the circuit breaker test, the mutual inductor test, the terminal test and the complete equipment test;

[0011] The output control circuit outputs the secondary signal under test and connects to the control center of the test master station through the control interface;

[0012] The test master station compares and calculates the similarity of waveforms collected by the recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested.

[0013] A method for detecting a signal of a primary and secondary deep fusion device, comprising:

[0014] 1) Connect the equipment to be tested to the switch box, wherein the equipment to be tested includes a transformer to be tested, a circuit breaker to be tested, a secondary terminal to be tested, and a complete set of equipment to be tested; and connect a high-speed recorder to the switch box;

[0015] The switch box includes a switch circuit, which is connected to an output control circuit; the output control circuit outputs a secondary signal to be tested and is connected to a control center of a test master station through a control interface;

[0016] 2) The test master station compares and calculates the similarity of the waveforms collected by the recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested.

[0017] The beneficial effects achieved by the present invention are as follows: Compared with the prior art, the present invention realizes full-item detection of primary wiring of primary and secondary integrated power distribution equipment through an automatic switching device, a high-precision detection method and a programmed detection process, and realizes functions such as real-time recording of the detection process, automatic data identification, and preliminary analysis of defects and faults, saving the time and energy of the detection staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of a primary and secondary deep fusion device detection device provided by the present invention;

[0019] Figure 2 is a schematic diagram of a signal switching device;

[0020] Figure 3 It is a wiring diagram of the primary and secondary deep fusion equipment detection switching device of the present invention;

[0021] Figure 4 The schematic diagram of the switching circuit is shown in FIG.

[0022] Figure 5 This is a schematic diagram of electronic voltage transformer measurement;

[0023] Figure 6 This is a flow chart of the control method for testing the master station in the present invention;

[0024] Figure 7 It is the planning path diagram of two waveforms under the DTW algorithm. DETAILED DESCRIPTION

[0025] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0026] The present invention utilizes the principle and characteristics of electronic transformers to sample large signals of electromagnetic transformers and small voltage / current signals of electronic transformers in an integrated and complete set. The detection system of electronic voltage transformers and electronic current transformers includes electronic transformers, signal cables and distribution terminals. The main reasons for the attenuation of electronic transformer output signals and the difficulty of impedance matching are analyzed according to an impedance network model, and a solution for high-precision impedance matching is proposed. Based on the advantages and existing problems of similarity analysis of recorded data, a dynamic time warping distance algorithm (DTW) is introduced to solve the problems of inconsistent timing spacing and inconsistent duration of waveform sampling data. According to the characteristics of frequent periodic fluctuations of sampled waveform data, the DTW algorithm is optimized and improved based on the longest common substring, thereby improving the similarity matching accuracy.

[0027] like Figure 1 As shown, a primary and secondary deep fusion device signal detection device of the present invention includes:

[0028] Switch box;

[0029] Connecting the equipment to be tested to the switch box, the equipment to be tested includes a transformer to be tested, a circuit breaker to be tested, a secondary terminal to be tested and a complete set of equipment to be tested;

[0030] The transformer to be tested is connected to the switch box through a signal conversion device; the circuit breaker to be tested, the terminal to be tested and the complete set of equipment to be tested are directly connected to the switch box; the high-speed recorder and the multi-function meter are connected to the switch box;

[0031] The transformer to be tested includes an electromagnetic transformer and an electronic transformer;

[0032] The switching box includes a switching switch circuit, which is connected to the output control circuit and is used to complete the wiring switching of the circuit breaker test, the mutual inductor test, the terminal test and the complete equipment test;

[0033] The output control circuit outputs the secondary signal under test and connects to the control center of the test master station through the control interface;

[0034] The test master station compares and calculates the similarity of waveforms collected by the recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested.

[0035] like Figure 2 As shown, the signal conversion device includes:

[0036] The small voltage / current signal output by the electronic transformer under test is connected to the switching circuit of the switching box through the load box;

[0037] The large voltage / current signal output by the electromagnetic transformer under test is sequentially connected to the built-in PT / CT, signal conditioning circuit, and signal conversion circuit to convert the large voltage / current signal into a signal in the same range as the EVT / ECT, and then connected to the switching circuit of the switching box; the signal conditioning circuit is used to convert the analog voltage / current signal into a digital signal for easy collection and control.

[0038] The signal conversion device of the present invention realizes the signal compatibility and switching of traditional electromagnetic and electronic mutual inductors and the automatic switching of wiring modes, thereby realizing the conversion of signal size and type.

[0039] The signal detection device for the primary and secondary deep fusion equipment of the present invention realizes the measurement of small voltage / current signals by adjusting the sampling circuit mode and circuit parameters of the transformer tester; at the same time, the access of EVT is realized by narrowing the PT input range, and the measurement circuit of the original CT is adjusted to the PT mode by using the built-in PT / CT to access the voltage signal of ECT, and the signal type conversion is realized through the signal conditioning circuit and the signal conversion circuit, so as to complete the calibration needs of compatible electromagnetic transformers and electronic transformers.

[0040] like Figure 3 As shown, during the detection process, the program-controlled high-voltage source outputs a primary signal, which is converted into a secondary signal through a switch and a voltage / current transformer. If an electronic transformer is used, the output small voltage / current signal is connected to the switch box after passing through the load box; if an electromagnetic transformer is used, the output large voltage / current signal is connected to the switch box after passing through a signal conversion device; after the secondary signal source to be tested is selected through the switch box, the tested signal enters the transformer calibrator, and then the standard secondary signal and the tested secondary signal are compared to obtain the primary transformer accuracy, ratio difference angle difference and other performance parameters.

[0041] Aiming at the problem that the low-power transformer output of the primary and secondary deep integration equipment has a fast attenuation of small voltage / current signals and high requirements for secondary wiring impedance matching, the present invention proposes a high-precision impedance matching solution. Under the premise of ensuring the wiring accuracy of the secondary cable, the full-item detection of the primary wiring of the primary and secondary integration equipment of the power distribution can be achieved.

[0042] like Figure 5 As shown, the electronic voltage transformer (EVT) uses the principle of capacitor voltage division to achieve voltage conversion. The equivalent impedance model of the entire network is described as follows:

[0043]

[0044]

[0045] Among them, C1 is the high voltage arm capacitor, C2 is the low voltage arm capacitor, C1 and C2 divide the voltage to obtain the low voltage signal US , a precision resistor R is connected in parallel across C2 S Form an energy release channel to improve transient response speed. Electronic voltage transformer EVT output voltage U S After passing through the transmission cable and reaching the terminal side, the measured voltage U2 is obtained. The equivalent resistance of the transmission cable is Rline, the equivalent resistance of the terminal is Rload, and t is the time.

[0046] The electronic current transformer (ECT) uses a low-power electronic current transformer (LPCT), which is different from the traditional current transformer in that the iron core used is an ultra-fine crystal alloy iron core, and a sampling resistor is integrated to convert the current signal into a voltage signal output, reduce the line impedance Rline, and optimize the structure of the primary and secondary fusion equipment. In this way, the transmission distance of small voltage / current signals is shortened to increase the voltage collected at the end, reduce the line impedance, increase the load capacity of the electronic transformer, and reduce the voltage / current signal attenuation. In terms of measuring equipment, the equivalent input impedance of the measuring equipment is increased to reduce the requirements for line impedance and transformer internal resistance, effectively improving the adaptability of the output signal of the electronic transformer. The measuring equipment includes distribution terminals, calibrators, etc.

[0047] like Figure 4 As shown, the core component of the present invention is a switch box, and all signal cables of the primary and secondary fusion equipment to be tested are connected to the switch box through aviation plugs, wherein the transformer to be tested should be connected to the switch box through an electromagnetic / electronic transformer signal conversion device.

[0048] At the same time, the high-speed recorder and multi-function meter are also connected to the switching box.

[0049] The switching circuit is connected to the output control circuit, and the output control circuit outputs the measured secondary signal.

[0050] The output control circuit switches the switch circuit through software control to realize all wiring switching of circuit breaker test, transformer test, terminal test and complete equipment test.

[0051] The output control circuit is realized by the CPU through the quick operation interface (group buttons) and the general control interface (network port, serial port, etc.) to realize various functions of the signal switching circuit, so as to achieve the requirement of completing all test items with one wiring.

[0052] Switch K1 is the power switch control of the device, switch S1 is the primary and secondary complete detection mode control, switch S2 is the switch detection mode control; switch S3 is the terminal detection mode control, and the switching of various test modes is realized by switching switches K1, S1, S2 and S3.

[0053] Figure 6 This is a flow chart of the control method for testing the master station in the present invention, such as Figure 6 As shown, the test items, test methods and test standards issued by the test software of the test master station drive the primary / secondary circuit according to the test requirements, output the specified signal, simulate the power grid production environment, notify the query of the data and information of the tested equipment, and detect the performance, function and other indicators of the primary and secondary fusion equipment through comparison and calculation. At the same time, the test environment data and test result data are uploaded to the test master station, and the waveform similarity is compared and calculated in the test master station, and optimized to obtain the correlation coefficient of the two waveforms. Based on the correlation coefficient, the terminal is further evaluated for performance and fault diagnosis.

[0054] The test master station regularly updates test items and test methods, receives / calls test data, and saves test data and reports.

[0055] A method for detecting a signal of a primary and secondary deep fusion device, comprising:

[0056] 1) Connect the equipment to be tested to the switch box, wherein the equipment to be tested includes a transformer to be tested, a circuit breaker to be tested, a secondary terminal to be tested, and a complete set of equipment to be tested; and connect a high-speed recorder to the switch box;

[0057] The switch box includes a switch circuit, which is connected to an output control circuit; the output control circuit outputs a secondary signal to be tested and is connected to a control center of a test master station through a control interface;

[0058] 2) The test master station compares and calculates the similarity of the waveforms collected by the recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested.

[0059] Step 2) specifically includes the following steps:

[0060] 21) Establish a distance matrix D for the two waveform time series collected by the wave recorder and the secondary terminal of the primary and secondary fusion device;

[0061] 22) Find a path from the upper left corner to the lower right corner of the distance matrix D so that the sum of the element values ​​​​passed by the path is minimized, and use the step mode algorithm to calculate the distance superposition of each step;

[0062] 23) Calculate the similarity of two waveforms;

[0063] 24) Calculate the common substring and length of two waveforms;

[0064] 25) Define the penalty coefficient and calculate the optimized matching coefficient;

[0065] 26) Optimize and correct the similarity between two waveforms.

[0066] like Figure 7As shown, the waveform time series collected by the recorder and the secondary terminal are calculated using the DTW algorithm to obtain a planned path, which is a straight line approximately 45°. The waveform time series collected by the recorder and the secondary terminal are compared, and the matching path regularization results of the two waveforms under the DTW algorithm show obvious deviation;

[0067] In step 22), the sum of the element values ​​along the path is minimized, that is, the twisted curve is solved.

[0068] X and Y are two waveform sequences collected by the recorder and the secondary terminal respectively. are the curves obtained by distorting waveform sequence X and waveform sequence Y, respectively. m and n are the lengths of the two waveform time series respectively. The lengths of waveform sequence X and waveform sequence Y are |X| and |Y| respectively. The similarity of the two waveforms is compared to solve a curve W=w1,w2,...,w k is a regular path, the length k of the regular path is greater than the length of any of the two waveform sequences and less than the sum of the lengths of the two waveform sequences, and the curve W starts from w1=(1,1) and extends to w k =(|X|,|Y|);

[0069] max(|X|,|Y|) <k<|X|+|Y|

[0070] The path with the smallest regularization cost is:

[0071]

[0072] DTW represents the final waveform matching path, and K is the compensation coefficient, which is used to compensate for regular paths of different lengths;

[0073] The obtained normalized route must be the shortest normalized route:

[0074] D(i,j)=Dist(i,j)+min[D(i-1,j),D(i,j-1),D(i-1,j-1)] (4)

[0075] i represents the i-th waveform time series X; j represents the j-th waveform time series X;

[0076] The cumulative distance D(i, j) is the sum of the current grid point distance Dist(i, j) and the cumulative distance of the smallest adjacent element that can reach the point. The previous grid point distance Dist(i, j) is the distance between point x and the previous grid point distance Dist(i, j). i and point y j The Euclidean distance, i.e. similarity.

[0077] In step 23), from a point (i-1, j-1), (i-1, j) or (i, j-1) of the previous square to a point (i, j) of the next square, if it is horizontal or vertical, the distance between the two waveforms sampled by the recorder and the secondary terminal is d(i, j); if it is oblique or diagonal, the distance between the two waveforms sampled by the recorder and the secondary terminal is 2d(i, j).

[0078]

[0079] Among them, g(i,j) means that both waveform sequences are matched from the starting component to the i component in waveform sequence X and the j component in waveform sequence Y, which means the distance between the two waveform sequences. Add d(i,j) or 2d(i,j) to the result of the previous match, then take the minimum value dist(X,Y), and then divide the similarity distance by Converted into similarity, ω(X,Y) is the similarity of the two compared waveforms.

[0080] Due to the constant fluctuation of the amplitude of temporal fluctuations in time series data, the monotonic shift speed of the curve direction is also different. In the step mode operation promotion process, the path direction can often only maintain a short-term monotonic characteristic, but the adjustment process is cumbersome, resulting in a "pathological matching" situation in which the peak and trough directions are different when calculating the time curve distance. Considering the electromagnetic interference caused by the opening and closing state to the secondary circuit, the current disturbance of the electronic sensor is large when the switch is opened and closed, and the waveform will have a phase error after entering the steady state time. In this invention, a penalty coefficient is defined to optimize the DTW algorithm, and the penalty coefficient is used to adjust the distance between the two waveform sequences. The new distance is obtained by multiplying the penalty coefficient by the distance.

[0081] 24) Calculate the common substring and length of two waveforms;

[0082] (a) Calculate the maximum standard deviation sd max , first calculate the average values ​​x and y of the two waveform sequences, the calculation formula is as follows:

[0083]

[0084]

[0085] Then we have:

[0086] same:

[0087] In the formula, sd X 、sd YThey represent the standard deviation of waveform sequence X and waveform sequence Y respectively, n represents the number of elements of waveform sequence X and waveform sequence Y, and the maximum standard deviation is the one with the larger data standard deviation in the two time series, that is,

[0088] sd max =max(sd X ,sd Y ) (6)

[0089] (b) Define the offset range. Since both waveform sequence X and waveform sequence Y are numerical sequences, set the maximum standard deviation to an appropriate offset range. That is, when the difference between the i-th value of sequence X and the j-th value of sequence Y is within the offset range, the i-th value of sequence X and the j-th value of sequence Y are included in the range of the common substring. Define the matrix:

[0090]

[0091] Use dynamic programming to solve the path in the matrix. When the elements in the path are less than the offset tolerance, they are continuously superimposed. The final value is the length L of the common substring.

[0092] 25) Define the penalty coefficient and calculate the optimized matching coefficient;

[0093] Define the penalty coefficient α as follows:

[0094]

[0095] |X| and |Y| are the sequence lengths of waveform sequence X and waveform sequence Y respectively;

[0096] 26) Optimize and correct the similarity of the two waveforms. Perform similarity correction according to equations (9) and (10), and use α*dist(X,Y) instead of dist(X,Y) (9)

[0097]

[0098] In the present invention, according to the path principle of the planning matrix step-by-step progress, the optimal path of the matrix dq[i][j] from the upper left corner to the lower right corner can be obtained, the longest common substring and its length can be obtained, the penalty coefficient can be calculated, and the distance algorithm can be optimized.

[0099] Based on the improved algorithm, a database is established in the cloud to obtain massive data, and the waveform similarity is matched with the fault condition to achieve the effect of fault preprocessing.

[0100] Example 2

[0101] The primary and secondary deep fusion device signal detection device is the same as that in Example 1.

[0102] A method for detecting a signal of a primary and secondary deep fusion device comprises the following steps:

[0103] 1) Connect the equipment to be tested to the switch box, wherein the equipment to be tested includes a transformer to be tested, a circuit breaker to be tested, a secondary terminal to be tested, and a complete set of equipment to be tested; and connect a high-speed recorder to the switch box;

[0104] The switch box includes a switch circuit, which is connected to an output control circuit; the output control circuit outputs a secondary signal to be tested and is connected to a control center of a test master station through a control interface;

[0105] 2) The test master station compares and calculates the similarity of the waveforms collected by the recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested.

[0106] The method also includes the following steps: in the pole switch fusion detection stage, the circuit breaker body is in a closed state, the three-phase incoming line side of the circuit breaker body is connected to the output end of the current regulation system, the outgoing line side of the circuit breaker body is short-circuited and grounded, the circuit breaker protection delay setting is set to 0s, the protection setting value I0 is set, the oscilloscope recorder is turned on, the current probe parameters (10A: 1V), the voltage probe parameters (1:1), the recording time parameters (120s), and the sampling rate (200kS / s) are set, the file name is saved, the oscilloscope recorder acquisition function is turned on, and the power circuit breaker Q is connected at the same time, so that the current regulation system gradually increases the current at a rate of 20A / s until the circuit breaker is disconnected, the oscilloscope recorder acquisition data is stored, and the oscilloscope recorder is turned off.

[0107] Calculation of the entire set of action time: Since there are certain differences in the three-phase current regulation rates, taking the phase with the largest current during protection action as an example, in the instantaneous current and voltage waveforms of the protection action recorded by the recorder, ts is the start time of the protection action, and tc is the end time of the protection action.

[0108] The current regulation system is connected to the primary side of the circuit breaker body to simulate the short-circuit fault current. The current and voltage waveforms are measured and recorded at high speed using an oscilloscope recorder, and the quick-break protection characteristics are obtained through the waveform data.

[0109] The above describes the specific module design and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited by the above examples. The above examples and descriptions are only for explaining the idea of ​​the present invention. Without departing from the spirit and scope of the technical solution of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A signal detection device for a primary and secondary deep fusion device, characterized in that ,include: Switch box; The equipment under test is connected to the switch box, and the equipment under test includes a transformer under test, a circuit breaker under test, a secondary terminal under test and a complete set of equipment under test; The transformer to be tested is connected to the switch box through a signal conversion device; the circuit breaker to be tested, the terminal to be tested and the complete set of equipment to be tested are directly connected to the switch box; the high-speed recorder and the multi-function meter are connected to the switch box; The transformer to be tested includes an electromagnetic transformer and an electronic transformer; The switching box includes a switching switch circuit, which is connected to the output control circuit and is used to complete the wiring switching of the circuit breaker test, the mutual inductor test, the terminal test and the complete equipment test; The output control circuit outputs the secondary signal under test and connects to the control center of the test master station through the control interface; The test master station compares and calculates the waveform similarity collected by the recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested, and the performance evaluation includes the following steps: 21) Establish a distance matrix D for the two waveform time series collected by the wave recorder and the secondary terminal of the primary and secondary fusion device; 22) Find a path from the upper left corner to the lower right corner of the distance matrix D so that the sum of the element values ​​​​passed by the path is minimized, and use the step mode algorithm to calculate the distance superposition of each step; 23) Calculate the similarity of two waveforms; 24) Calculate the common substring and length of two waveforms; 25) Define the penalty coefficient and calculate the optimized matching coefficient; 26) Optimize and correct the similarity between two waveforms; In step 22), the sum of the element values ​​along the path is minimized, that is, the twisted curve is solved. X and Y are two waveform sequences collected by the recorder and the secondary terminal respectively. are the curves obtained by distorting waveform sequence X and waveform sequence Y, respectively. m and n are the lengths of the two waveform time series respectively. The lengths of waveform sequence X and waveform sequence Y are |X| and |Y| respectively. The similarity of the two waveforms is compared to solve a curve W=w1,w2,...,w k is a regular path, the length k of the regular path is greater than the length of any of the two waveform sequences and less than the sum of the lengths of the two waveform sequences, and the curve W starts from w1=(1,1) and extends to w k =(|X|,|Y|); max(|X|,|Y|) <k<|X|+|Y| The path with the smallest regularization cost is: DTW represents the final waveform matching path, and K is the compensation coefficient, which is used to compensate for regular paths of different lengths; The obtained normalized route is the one with the shortest distance: D(i,j)=Dist(i,j)+min[D(i-1,j),D(i,j-1),D(i-1,j-1)] (4) i represents the i-th waveform time series X; j represents the j-th waveform time series X; The cumulative distance D(i, j) is the sum of the current grid point distance Dist(i, j) and the cumulative distance of the smallest adjacent element that can reach the point. The previous grid point distance Dist(i, j) is the distance between point x and the previous grid point distance Dist(i, j). i and point y j The Euclidean distance, i.e., similarity; In step 23), from a point (i-1, j-1), (i-1, j) or (i, j-1) of the previous square to a point (i, j) of the next square, if it is horizontal or vertical, the distance between the two waveforms sampled by the recorder and the secondary terminal is d(i, j); if it is oblique or diagonal, the distance between the two waveforms sampled by the recorder and the secondary terminal is 2d(i, j). Among them, g(i,j) means that both waveform sequences are matched from the starting component to the i component in waveform sequence X and the j component in waveform sequence Y, which means the distance between the two waveform sequences. Add d(i,j) or 2d(i,j) to the result of the previous match, then take the minimum value dist(X,Y), and then divide the similarity distance by Converted into similarity, ω(X,Y) is the similarity of the two compared waveforms; In step 24), the common substrings and lengths of the two waveforms are calculated, including the following steps: (a) Calculate the maximum standard deviation sd max , first calculate the average values ​​x and y of the two waveform sequences, the calculation formula is as follows: Then we have: In the formula, sd X 、sd Y They represent the standard deviation of waveform sequence X and waveform sequence Y respectively, n represents the number of elements of waveform sequence X and waveform sequence Y, and the maximum standard deviation is the one with the larger data standard deviation in the two time series, that is, sd max =max(sd X ,sd Y )(6) (b) Define the offset range. Since both waveform sequence X and waveform sequence Y are numerical sequences, set the maximum standard deviation to an appropriate offset range. That is, when the difference between the i-th value of sequence X and the j-th value of sequence Y is within the offset range, the i-th value of sequence X and the j-th value of sequence Y are included in the range of the common substring. Define the matrix: Use dynamic programming to solve the path in the matrix. When the elements in the path are less than the offset tolerance, they are continuously superimposed. The final value is the length L of the common substring. In step 25), the penalty coefficient α is defined as follows: |X| and |Y| are the sequence lengths of waveform sequence X and waveform sequence Y respectively; In step 26), the similarity between the two waveforms is optimized and corrected according to equations (9) and (10). Use α*dist(X,Y) instead of dist(X,Y) (9) 2. The signal detection device for a primary and secondary deep fusion device according to claim 1, characterized in that: The signal conversion device comprises: The small voltage / current signal output by the electronic transformer under test is connected to the switching circuit of the switching box through the load box; The large voltage / current signal output by the electromagnetic transformer under test is sequentially connected to the built-in PT / CT, signal conditioning circuit, and signal conversion circuit to convert the large voltage / current signal into a signal in the same range as the EVT / ECT, and then connected to the switching circuit of the switching box; the signal conditioning circuit is used to convert the analog voltage / current signal into a digital signal.

3. A method for detecting signals of a primary and secondary deep fusion device, characterized in that: The following steps are involved: 1) Connect the equipment to be tested to the switch box, wherein the equipment to be tested includes a transformer to be tested, a circuit breaker to be tested, a secondary terminal to be tested, and a complete set of equipment to be tested; and connect a high-speed recorder to the switch box; The switch box includes a switch circuit, which is connected to an output control circuit; the output control circuit outputs a secondary signal to be tested and is connected to a control center of a test master station through a control interface; 2) The test master station compares and calculates the similarity of the waveforms collected by the wave recorder and the secondary terminal to be tested of the primary and secondary fusion device, and performs performance evaluation on the secondary terminal to be tested; In step 2), the following steps are specifically included: 21) Establish a distance matrix D for the two waveform time series collected by the wave recorder and the secondary terminal of the primary and secondary fusion device; 22) Find a path from the upper left corner to the lower right corner of the distance matrix D so that the sum of the element values ​​​​passed by the path is minimized, and use the step mode algorithm to calculate the distance superposition of each step; 23) Calculate the similarity of two waveforms; 24) Calculate the common substring and length of two waveforms; 25) Define the penalty coefficient and calculate the optimized matching coefficient; 26) Optimize and correct the similarity between two waveforms; In step 22), the sum of the element values ​​along the path is minimized, that is, the twisted curve is solved. X and Y are two waveform sequences collected by the recorder and the secondary terminal respectively. are the curves obtained by distorting waveform sequence X and waveform sequence Y, respectively. m and n are the lengths of the two waveform time series respectively. The lengths of waveform sequence X and waveform sequence Y are |X| and |Y| respectively. The similarity of the two waveforms is compared to solve a curve W=w1,w2,...,w k is a regular path, the length k of the regular path is greater than the length of any of the two waveform sequences and less than the sum of the lengths of the two waveform sequences. The curve W starts from w1=(1,1) and extends to w k =(|X|,|Y|); max(|X|,|Y|) <k<|X|+|Y| The path with the smallest regularization cost is: DTW represents the final waveform matching path, and K is the compensation coefficient, which is used to compensate for regular paths of different lengths; The obtained normalized route is the one with the shortest distance: D(i,j)=Dist(i,j)+min[D(i-1,j),D(i,j-1),D(i-1,j-1)] (4) i represents the i-th waveform time series X; j represents the j-th waveform time series X; The cumulative distance D(i, j) is the sum of the current grid point distance Dist(i, j) and the cumulative distance of the smallest adjacent element that can reach the point. The previous grid point distance Dist(i, j) is the distance between point x and the previous grid point distance Dist(i, j). i and point y j The Euclidean distance, i.e., similarity; In step 23), from a point (i-1, j-1), (i-1, j) or (i, j-1) of the previous square to a point (i, j) of the next square, if it is horizontal or vertical, the distance between the two waveforms sampled by the recorder and the secondary terminal is d(i, j); if it is oblique or diagonal, the distance between the two waveforms sampled by the recorder and the secondary terminal is 2d(i, j). Among them, g(i,j) means that both waveform sequences are matched from the starting component to the i component in waveform sequence X and the j component in waveform sequence Y, which means the distance between the two waveform sequences. Add d(i,j) or 2d(i,j) to the result of the previous match, then take the minimum value dist(X,Y), and then divide the similarity distance by Converted into similarity, ω(X,Y) is the similarity of the two compared waveforms; In step 24), the common substrings and lengths of the two waveforms are calculated, including the following steps: (a) Calculate the maximum standard deviation sd max , first calculate the average values ​​x and y of the two waveform sequences, the calculation formula is as follows: Then we have: In the formula, sd X 、sd Y They represent the standard deviation of waveform sequence X and waveform sequence Y respectively, n represents the number of elements of waveform sequence X and waveform sequence Y, and the maximum standard deviation is the one with the larger data standard deviation in the two time series, that is, sd max =max(sd X ,sd Y )(6) (b) Define the offset range. Since both waveform sequence X and waveform sequence Y are numerical sequences, set the maximum standard deviation to an appropriate offset range. That is, when the difference between the i-th value of sequence X and the j-th value of sequence Y is within the offset range, the i-th value of sequence X and the j-th value of sequence Y are included in the range of the common substring. Define the matrix: Use dynamic programming to solve the path in the matrix. When the elements in the path are less than the offset tolerance, they are continuously superimposed. The final value is the length L of the common substring. In step 25), the penalty coefficient α is defined as follows: |X| and |Y| are the sequence lengths of waveform sequence X and waveform sequence Y respectively; In step 26), the similarity between the two waveforms is optimized and corrected according to equations (9) and (10). Use α*dist(X,Y) instead of dist(X,Y) (9) 4. The method for detecting a signal of a primary and secondary deep fusion device according to claim 3, characterized in that: The following steps are included in the column switch fusion detection phase: Put the circuit breaker body in the closed state, connect the three-phase incoming line side of the circuit breaker body to the output end of the current regulation system, short-circuit the outgoing line side of the circuit breaker body to ground, set the circuit breaker protection delay setting value to 0s, set the protection setting value I0, turn on the oscilloscope recorder, set the current probe parameters, voltage probe parameters, recording time parameters, sampling rate, save the file name, turn on the oscilloscope recorder acquisition function, and connect the power circuit breaker Q at the same time, so that the current regulation system gradually increases the current at the set rate until the circuit breaker is disconnected, store the oscilloscope recorder acquisition data, and turn off the oscilloscope recorder; Calculate the whole set of action time: including the protection action start time ts and protection action end time tc recorded by the recorder; The current regulation system is connected to the primary side of the circuit breaker body to simulate the short-circuit fault current. The current and voltage waveforms are measured and recorded at high speed using a recorder, and the quick-break protection characteristics are obtained through the waveform data.

Images