A method and system for collecting traction network voltage data

Abstract

The application discloses a kind of acquisition method and system for traction network voltage data, it is related to traction network voltage acquisition technical field.The present application acquires data of different granularity and accuracy according to the different position of information flow, retains high information density data, reduces the recording of repeated waveform, further improves the storage utilization and the information value of unit data, and provides the detection accuracy required for different frequency signals, uses the strategy of hierarchical sampling, retains more valuable sampling data according to conditions;The device topology of acquisition system is simple, strong, easy to use, and the sampling device is easy to obtain, and the voltage data of each frequency band and various characteristics on the traction network can be collected, which can be plugged and played and worked for a long time without supervision, further improving the working efficiency of the system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traction network voltage acquisition, and particularly relates to a traction network voltage data acquisition method and system. BACKGROUND

[0002] The track traction power supply system is the core of the modern electrified railway, and its design and operation are crucial to the efficient and safe operation of the railway. Therefore, the voltage of the train traction network needs to be monitored for a long time with high precision. In the traction power supply system, the monitoring components that need to be completed mainly include the basic 50Hz voltage condition, high-order harmonic resonance, and instantaneous high-frequency impact, i.e., high-frequency impact data, fundamental voltage data, and harmonic resonance data. The harmonic resonance is mainly caused by power electronic devices such as inverters in the traction network, generally at 500Hz-4kHz, while the high-frequency impact generally occurs when the phase is passed, and the frequency can reach more than 60kHz. The envelope characteristics and frequencies of the three types of data components are different. For the fundamental component, the information density in the collected data is low, the repeated components are more, and the occupied space is larger. For the high-frequency impact, the information density in the collected data is large, and there is almost no repeated data. The data occupation and information density of the harmonic resonance data are between the two, so if a single acquisition and storage strategy is used, it is difficult to achieve the purposes of compression storage and accuracy retention. It is very cumbersome to perform long-time measurement in such a situation. The existing technology currently uses a continuous long-time constant sampling rate to collect high-frequency impact data, fundamental voltage data, and harmonic resonance data, which reduces the accuracy of high-frequency impact data and further increases the burden of data storage due to a large amount of repeated sampling data. SUMMARY

[0003] In view of the above problems in the prior art, the present application provides a traction network voltage data acquisition method and system, which solves the problem of a large amount of repeated sampling data caused by continuous long-time sampling at a constant sampling rate in the prior art acquisition method, reduces the work efficiency, and wastes storage space.

[0004] In order to achieve the above-mentioned application purposes, the technical scheme adopted by the present application is as follows:

[0005] A traction network voltage data acquisition method is provided, which includes the following steps:

[0006] S1, setting trigger parameters and sampling parameters and obtaining ARM control instructions through a PC module;

[0007] S2, obtaining FPGA control instructions in real time through an ARM embedded module according to the ARM control instructions; the FPGA control instructions include capture instructions, trigger instructions, and DFT change instructions;

[0008] S3, acquiring the ADC control instruction through the FPGA module according to the capture instruction;

[0009] S4, collecting the traction network voltage through the ADC module according to the ADC control instruction to obtain raw voltage data;

[0010] S5, judging the raw voltage data through the FPGA module according to the trigger instruction to obtain judged voltage data;

[0011] S6, sampling and slicing the judged voltage data through the FPGA module according to the sampling parameter to obtain sliced voltage data windows;

[0012] S7, performing DFT transformation on the sliced voltage data windows through the FPGA module according to the DFT transformation instruction to obtain transformed voltage data windows;

[0013] S8, arranging the transformed voltage data windows through the ARM embedded module to obtain arranged voltage data;

[0014] S9, analyzing the arranged voltage data through the PC module to obtain analyzed voltage data;

[0015] S10, performing block storage on the analyzed voltage data through the SQLite offline database to complete the collection of the traction network voltage.

[0016] Further, the sampling parameter in step S1 includes a sampling frequency, a sampling duration and a sampling duration cycle; and the raw voltage data in step S4 is one-dimensional time discrete data.

[0017] Further, the judged voltage data in step S5 includes window capture data and trigger data.

[0018] Step S5 further includes:

[0019] S5-1, judging whether the raw voltage data contains high-frequency impact data through the FPGA module according to the trigger instruction; if yes, proceeding to step S5-2; otherwise, taking the raw voltage data as the judged voltage data and proceeding to step S6;

[0020] S5-2, classifying the raw voltage data through the FPGA module to obtain the window capture data and the trigger data; the window capture data includes fundamental voltage data and harmonic resonance data; and the trigger data is high-frequency impact data.

[0021] S5-3, transmitting the trigger data to the PC module through the ARM embedded module and storing the same;

[0022] S5-4, the window capture data is taken as the voltage data after judgment and enters step S6.

[0023] Further, the specific process of step S6 is:

[0024] The sampling duration is taken as the interval and the sampling duration time is taken as the length of each record. The voltage data after judgment is recorded periodically by using the timing function of the FPGA module. After each record is completed, all voltage data obtained by this record is taken as the voltage data window after decomposition this time.

[0025] The voltage data window after fragmentation includes the voltage data window after decomposition corresponding to each record.

[0026] Further, the specific process of step S8 is:

[0027] The window length and the window interval are set. The transformed voltage data is divided into a plurality of windows according to the window interval and the window length by using the timer of the ARM embedded module. The voltage data between the windows is removed and the voltage data within the windows is retained to obtain the sorted voltage data.

[0028] Further, step S9 further includes:

[0029] S9-1, the average value of the sorted voltage data is calculated;

[0030] S9-2, the effective value of the sorted voltage data is calculated;

[0031] S9-3, the sorted voltage data is screened. The processed voltage data with the average value and the effective value less than the threshold value is removed to obtain the screened voltage data;

[0032] S9-4, a time window is set. The screened voltage data in each time window is calculated to obtain the corresponding harmonic component and the amplitude;

[0033] S9-5, each harmonic component and the amplitude are calculated to obtain the corresponding harmonic proportion information;

[0034] S9-6, according to the harmonic proportion information, the screened voltage data is analyzed to obtain the analyzed voltage data.

[0035] A traction network voltage data acquisition system is provided, which includes an ADC module, an FPGA module, an ARM embedded module, a transmission module, a PC module and a storage module; wherein:

[0036] The ADC module is used for acquiring original data.

[0037] FPGA module, for controlling the ADC module to collect and process the collected raw voltage data;

[0038] ARM embedded module, for controlling the FPGA module and arranging the transformed voltage data window output by the FPGA module;

[0039] Transmission module, for transmitting data and instructions between the PC module and the ARM embedded module;

[0040] PC module, for analyzing the arranged voltage data and controlling the ARM embedded module;

[0041] Storage module, for storing the analyzed voltage data and trigger data in blocks by using SQLite offline database, and completing the collection of traction network voltage.

[0042] Further, the data processing includes sequentially judging, sampling and DFT transforming the raw voltage data; the arranging is screening the transformed voltage data by using a window and a timer of the ARM embedded module; and the data analysis is calculating average value, effective value and harmonic proportion information, and performing harmonic analysis on the arranged voltage data according to the average value, effective value and harmonic proportion information.

[0043] The present application has the advantages that: the present application collects data with different intensity and precision according to different positions of information flow, retains data with high information density, reduces the recording of repeated waveforms, further improves the storage utilization rate and the information value of unit data, and provides the required detection precision for different frequency signals, uses a hierarchical sampling strategy, and retains more valuable sampling data according to conditions. The device topology of the collection system is simple, convenient, easy to obtain, can collect voltage data of various frequency bands and characteristics on the traction network, can be used immediately after being plugged in and can work for a long time without supervision, and further improves the working efficiency of the system. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 It is a collection method flowchart;

[0045] Figure 2 It is a sampling process schematic diagram;

[0046] Figure 3 It is a system structure diagram;

[0047] Figure 4 It is a structure schematic diagram of the FPGA module and the ARM embedded module;

[0048] Figure 5 It is a voltage change curve diagram. DETAILED DESCRIPTION

[0049] The specific embodiments of the present application are described below to enable those skilled in the art to understand the present application, but it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, it is obvious that various changes are within the spirit and scope of the present application defined and determined by the appended claims, and all the inventions using the concept of the present application are within the scope of protection.

[0050] As Figure 1 shown, a data acquisition method for traction network voltage acquisition includes the following steps:

[0051] S1, set the trigger parameters and sampling parameters and obtain the ARM control instructions through the PC module; the sampling parameters in step S1 include the sampling frequency, the sampling duration and the sampling duration cycle; wherein the sampling frequency is defined by the ADC module, and the sampling duration and the sampling duration cycle are controllable parameters that need to be set in advance.

[0052] S2, according to the ARM control instructions, real-time acquisition of FPGA control instructions through ARM embedded module; FPGA control instructions include capture instructions, trigger instructions and DFT change instructions;

[0053] S3, according to the capture instructions, obtain the ADC control instructions through the FPGA module;

[0054] S4, according to the ADC control instructions, collect the traction network voltage through the ADC module to obtain the original voltage data; the original voltage data in step S4 is one-dimensional time discrete data.

[0055] S5, according to the trigger instructions, the original voltage data is judged through the FPGA module to obtain the judged voltage data;

[0056] The judged voltage data in step S5 includes window capture data and trigger data;

[0057] Step S5 further includes:

[0058] S5-1, according to the trigger instructions, the FPGA module is used to judge whether the original voltage data contains high-frequency impact data; if yes, go to step S5-2; otherwise, the original voltage data is taken as the judged voltage data and go to step S6;

[0059] S5-2, the original voltage data is classified through the FPGA module to obtain the window capture data and the trigger data; the window capture data includes the fundamental voltage data and the harmonic resonance data; the trigger data is high-frequency impact data, which is in very high frequency;

[0060] S5-3, transmit the trigger data to the PC module through the ARM embedded module and store it;

[0061] S5-4, take the window capture data as the judged voltage data and enter step S6.

[0062] S6, sample and slice the judged voltage data according to the sampling duration and the sampling duration cycle through the FPGA module to obtain the sliced voltage data window;

[0063] As shown in the figure, the specific process of step S6 is as follows: Figure 2

[0064] Take the sampling duration cycle as the interval and the sampling duration as the length of each record, and periodically record the judged voltage data by using the timing function of the FPGA module; after each record is completed, take all the voltage data obtained by this record as the voltage data window of this decomposition;

[0065] The sliced voltage data window includes the decomposed voltage data window corresponding to each record.

[0066] S7, perform DFT transformation on the sliced voltage data window through the FPGA module according to the DFT transformation instruction to obtain the transformed voltage data window;

[0067] The fundamental voltage data and the harmonic resonance data are derived from the original data collected by the same ADC, and are obtained after DFT frequency domain segmentation by the FPGA module, so the sampling data of the fundamental voltage data and the harmonic resonance data are consistent (derived from the same ADC), but after frequency domain segmentation by the FPGA module, the frequency ranges of the two are different, the fundamental voltage data mainly contains the voltage waveform data of the reference frequency of the alternating current power grid, so when measuring a specific alternating current power supply system (such as an alternating current traction power supply system), the specific frequency refers to the reference frequency of the system, and the purposes of the two parameters of sampling duration and sampling duration cycle are to align the sampling slice with the frequency, while the harmonic resonance data mainly contains the data in the frequency domain except the reference frequency of the power grid, which is superimposed in the fundamental voltage data of the sampling slice, but does not align with the sampling slice.

[0068] S8, arrange the transformed voltage data through the ARM embedded module to obtain the arranged voltage data;

[0069] The specific process of step S8 is as follows:

[0070] ​Setting window length and window interval; dividing the transformed voltage data into several windows according to the window interval and the window length through the timer of the ARM embedded module, eliminating the voltage data between the windows and retaining the voltage data within the windows, and obtaining the arranged voltage data. The values of the window length and the window interval are the same as the values of the sampling duration and the sampling duration period, respectively;

[0071] S9, analyzing the arranged voltage data through the PC module to obtain analyzed voltage data;

[0072] Step S9 further comprises:

[0073] S9-1, calculating the average value of the arranged voltage data; wherein the formula of the average value is:

[0074]

[0075] Vaver represents the average value of the arranged voltage data, Vi represents the voltage amplitude of the i-th sampling point, n represents the number of sampling points in the window (window width), and ∑(·) represents the summation function;

[0076] S9-2, calculating the effective value of the arranged voltage data; wherein the formula of the effective value is:

[0077]

[0078] V rms represents the effective value of the arranged voltage data, M represents the maximum frequency of DFT decomposition, represents the square of the peak value of the m-th frequency;

[0079] S9-3, screening the arranged voltage data, removing the processed voltage data whose average value and effective value are less than the threshold value, and obtaining screened voltage data;

[0080] S9-4, setting a time window, calculating the screened voltage data in each time window, and obtaining the corresponding harmonic component and its amplitude;

[0081] S9-5, calculating each harmonic component and its amplitude to obtain corresponding harmonic proportion information; wherein the formula of the harmonic proportion information is:

[0082]

[0083] V s1 = V p,1

[0084] V sh represents the total harmonic content, V s1 represents the fundamental content, Vp,1 represents a fundamental peak value;

[0085] S9-6, according to the harmonic proportion information, performing harmonic analysis on the screened voltage data to obtain analyzed voltage data.

[0086] The harmonic analysis refers to recording the harmonic distortion rate value of each window, monitoring and recording the change thereof; according to the change of the harmonic distortion rate value, performing secondary screening on the screened voltage data, eliminating the screened voltage data whose change of the harmonic distortion rate value exceeds the threshold range, and retaining the screened voltage data whose change of the harmonic distortion rate value is within the threshold range. The formula of the harmonic distortion rate value is:

[0087]

[0088] THD represents the harmonic distortion rate value, reflecting the proportion of the harmonic component.

[0089] S10, using the SQLite offline database to perform block storage on the analyzed voltage data, to complete the collection of the traction network voltage.

[0090] As shown in Figure 3 , a collection system for traction network voltage data includes an ADC module, an FPGA module, an ARM embedded module, a transmission module, a PC module and a storage module; wherein:

[0091] The ADC module is used for collecting original data.

[0092] The FPGA module is used for controlling the ADC module to collect and performing data processing on the collected original voltage data; the data processing includes sequentially performing judgment, sampling fragmentation and DFT transformation on the original voltage data; the function of the FPGA module is realized by Verilog language.

[0093] The ARM embedded module is used for controlling the FPGA module through the internal interrupt controller and the state register, and sorting the transformed voltage data window output by the FPGA module; the sorting is screening the transformed voltage data by using the window and the timer of the ARM embedded module.

[0094] As shown in Figure 4As shown, the FPGA module includes a root mean square calculation module RMS, a static random access memory BRAM, a first-in-first-out structure FIFO, a discrete Fourier transform module DFT, a dynamic random access memory DRAM, a storage module, a delay module DELAY, a gating switch and a timer TIMER, and the storage module includes a dynamic random access memory DRAM and an advanced extension interface AXI4. The signal receiving end of the ADC module is connected to the control instruction and the original data output by the timer TIMER, and the signal output end of the ADC module is connected to the signal input end of the static random access memory BRAM, the signal input end of the root mean square calculation module RMS and the signal input end of the first-in-first-out structure FIFO; the signal output end of the static random access memory BRAM and the signal output end of the root mean square calculation module RMS are connected to the ARM embedded module; the timer TIMER has the functions of synchronous input and synchronous output, the clock signal end of the timer TIMER is connected to the output end of the ARM embedded module, the dft_store register of the timer TIMER is connected to the signal input end of the delay module DELAY and the first signal input end of the gating switch; the signal output end of the delay module DELAY is connected to the first signal input end of the storage module; the signal output end of the first-in-first-out structure FIFO is connected to the signal input end of the discrete Fourier transform module DFT; the signal output end of the discrete Fourier transform module DFT is connected to the signal input end of the dynamic random access memory DRAM; the signal output end of the dynamic random access memory DRAM is divided into three paths, and each is connected to the second signal input end of the gating switch; the signal output end of the gating switch is connected to the second signal input end of the storage module; the signal output end of the storage module is connected to the ARM embedded module.

[0095] The ARM embedded module includes a first ARM embedded module ARM0, a memory controller and a second ARM embedded module ARM1 connected in series. The first ARM embedded module ARM0 includes an interrupt program and a main program, the interrupt program is used for reading a sampling value, correcting a bias, converting an instantaneous value, judging an overvoltage trigger logic and recording an overvoltage, and the main program is used for checking a correction ratio and a frequency check; the memory controller includes an overvoltage recording memory, a frequency detection memory, a ratio / bias / instantaneous effective value threshold, a frequency detection and a DFT result memory; the second ARM embedded module ARM1 includes an interrupt program and a main program, the interrupt program is used for processing a 485 receiving program and decoding, and the main program is used for modifying a ratio, a bias, an instantaneous overvoltage threshold, an effective overvoltage threshold, uploading a DFT, an overvoltage waveform and a frequency.

[0096] The FPGA module and the ARM embedded module transmit data according to two standard bus protocols of AXI-GP and AXI-HP. The EMIO DFT establishes a bridge between the FPGA module and the ARM embedded module, realizes the combination of the FPGA module and the ARM embedded module, and further achieves the purpose of collecting traction network voltage data.

[0097] The transmission module is used for transmitting data and instructions between the PC module and the ARM embedded module.

[0098] The PC module is used for data analysis on the sorted voltage data, and controls the ARM embedded module through serial communication 485. The data analysis is to calculate average value, effective value and harmonic proportion information, and to perform harmonic analysis on the sorted voltage data according to the average value, the effective value and the harmonic proportion information.

[0099] The storage module is used for block storage of the analyzed voltage data and trigger data by using a SQLite offline database, and completes the collection of the traction network voltage. The storage module is based on a file system, performs block storage on the analyzed voltage data and trigger data according to the storage space of a single data block, and stores the block-stored data into each single data block. After the data is block-stored, the collection system can support long-time recording, so that the recording time is only affected by the upper limit of the disk size. The single data block can store up to 100,000 data.

[0100] In one embodiment of the present application, the data recorded by the device is as follows: a large amount of linear time voltage data is calculated and decomposed to obtain a plurality of analyzed data in Table 1. The analyzed data items are greatly reduced (taking the first 8 data recorded in the example as an example). The reduced data still effectively maintains the monitoring function of the voltage, and a voltage change curve as shown in FIG. 1 can be drawn. Figure 5

[0101] Table 1

[0102] Effective value Harmonic content Amplitude 0.002142046 0.82140936 0.00234085 0.002007925 0.75756632 0.002263461 0.002078233 0.786952887 0.002309651 0.002042977 0.834423199 0.00221836 0.002086521 0.789971843 0.00231546 0.002469993 0.799361211 0.002728501 0.002087808 0.352335783 0.002784808 0.002428858 0.329472488 0.003262414 0.001241298 0.443645489 0.001604636

[0103] In summary, the present application collects data with different intensity and precision according to different positions of information flow, retains data with high information density, reduces the recording of repeated waveforms, further improves the storage utilization rate and the information value of unit data, and provides the required detection precision for different frequency signals. The hierarchical sampling strategy is used to retain more valuable sampling data according to conditions, and the working efficiency of the system is improved.​

Claims

1. A method for collecting traction network voltage data, characterized by: The method comprises the following steps: S1, setting trigger parameters and sampling parameters and obtaining ARM control instructions through a PC module; S2, obtaining FPGA control instructions in real time through an ARM embedded module according to the ARM control instructions; the FPGA control instructions comprise capture instructions, trigger instructions and DFT change instructions; S3, obtaining ADC control instructions through the FPGA module according to the capture instructions; S4, collecting traction network voltage through the ADC module according to the ADC control instructions to obtain original voltage data; S5, judging the original voltage data through the FPGA module according to the trigger instructions to obtain judged voltage data; S6, sampling and slicing the judged voltage data through the FPGA module according to the sampling parameters to obtain sliced voltage data windows; S7, performing DFT transformation on the sliced voltage data windows through the FPGA module according to the DFT transformation instructions to obtain transformed voltage data windows; S8, arranging the transformed voltage data windows through the ARM embedded module to obtain arranged voltage data; S9, analyzing the arranged voltage data through the PC module to obtain analyzed voltage data; S10, performing block storage on the analyzed voltage data through a SQLite offline database to complete traction network voltage collection; The judged voltage data in the step S5 comprises window capture data and trigger data; The step S5 further comprises: S5-1, judging whether the original voltage data contains high-frequency impact data through the FPGA module according to the trigger instructions; if yes, proceeding to step S5-2; otherwise, taking the original voltage data as the judged voltage data and proceeding to step S6; S5-2, classifying the original voltage data through the FPGA module to obtain window capture data and trigger data; the window capture data comprises fundamental voltage data and harmonic resonance data; the trigger data is high-frequency impact data; S5-3, transmitting the trigger data to the PC module through the ARM embedded module and storing the same; S5-4, taking the window capture data as the judged voltage data and proceeding to step S6.

2. The method for collecting traction grid voltage data according to claim 1, characterized in that: The sampling parameters in the step S1 comprise a sampling frequency, a sampling duration and a sampling duration cycle; The original voltage data in the step S4 is one-dimensional time discrete data.

3. The method for collecting traction grid voltage data according to claim 1, characterized in that: The specific process of the step S6 is: taking the sampling duration cycle as an interval and taking the sampling duration as the length of each record, periodically recording the judged voltage data by using the timing function of the FPGA module; after each record is completed, taking all the voltage data obtained by the record as the voltage data window of this time decomposition; the sliced voltage data window comprises the voltage data windows of each record corresponding to the decomposition.

4. The method for collecting traction grid voltage data according to claim 1, characterized in that: The specific process of the step S8 is: setting a window length and a window interval; dividing the transformed voltage data into a plurality of windows according to the window interval and the window length by using the timer of the ARM embedded module, eliminating the voltage data between the windows and retaining the voltage data within the windows to obtain the arranged voltage data.

5. The method for collecting traction grid voltage data according to claim 1, characterized in that: The step S9 further comprises: S9-1, calculate the average value of the sorted voltage data; S9-2, calculate the effective value of the sorted voltage data; S9-3, screen the sorted voltage data, remove the processed voltage data whose average value and effective value are less than the threshold value, and obtain the screened voltage data; S9-4, set a time window, calculate the screened voltage data in each time window, and obtain the corresponding harmonic component and its amplitude; S9-5, calculate each harmonic component and its amplitude to obtain corresponding harmonic ratio information; S9-6, according to the harmonic ratio information, perform harmonic analysis on the screened voltage data to obtain the analyzed voltage data.

6. A collection system for a collection method of traction network voltage data according to any one of claims 1 to 5, characterized in that: The device comprises an ADC module, an FPGA module, an ARM embedded module, a transmission module, a PC module and a storage module, wherein: The ADC module is used for collecting original data; The FPGA module is used for controlling the ADC module to collect and process the collected original voltage data; The ARM embedded module is used for controlling the FPGA module and arranging the transformed voltage data window output by the FPGA module; The transmission module is used for transmitting data and instructions between the PC module and the ARM embedded module; The PC module is used for data analysis of the arranged voltage data and control of the ARM embedded module; The storage module is used for block storage of the analyzed voltage data and trigger data by using SQLite offline database, and completes the collection of traction network voltage.

7. The collection system of claim 6, wherein: The data processing comprises judging, sampling and DFT transforming the original voltage data in sequence; The arrangement is screening the transformed voltage data by using the window and the timer of the ARM embedded module; The data analysis is calculating the average value, effective value and harmonic ratio information, and performing harmonic analysis on the arranged voltage data according to the average value, effective value and harmonic ratio information.

Images