CN111965652B - Live working protective cap based on random forest algorithm - Google Patents
Live working protective cap based on random forest algorithm Download PDFInfo
- Publication number
- CN111965652B CN111965652B CN202010844921.1A CN202010844921A CN111965652B CN 111965652 B CN111965652 B CN 111965652B CN 202010844921 A CN202010844921 A CN 202010844921A CN 111965652 B CN111965652 B CN 111965652B
- Authority
- CN
- China
- Prior art keywords
- resistor
- circuit
- signal
- random forest
- amplifier
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007637 random forest analysis Methods 0.000 title claims abstract description 31
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 29
- 230000001681 protective effect Effects 0.000 title claims abstract description 13
- 239000003990 capacitor Substances 0.000 claims description 26
- 230000006698 induction Effects 0.000 claims description 17
- 239000000523 sample Substances 0.000 claims description 17
- 238000006243 chemical reaction Methods 0.000 claims description 14
- 238000001228 spectrum Methods 0.000 claims description 14
- 238000001514 detection method Methods 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 12
- 230000004888 barrier function Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 7
- 238000000034 method Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 230000001939 inductive effect Effects 0.000 claims description 4
- 238000002592 echocardiography Methods 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 3
- 230000005672 electromagnetic field Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims 8
- 238000012423 maintenance Methods 0.000 abstract description 5
- 238000005259 measurement Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 3
- 230000003321 amplification Effects 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/02—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems using reflection of acoustic waves
- G01S15/06—Systems determining the position data of a target
- G01S15/08—Systems for measuring distance only
-
- A—HUMAN NECESSITIES
- A42—HEADWEAR
- A42B—HATS; HEAD COVERINGS
- A42B1/00—Hats; Caps; Hoods
- A42B1/04—Soft caps; Hoods
- A42B1/08—Soft caps; Hoods with protection against blows
-
- A—HUMAN NECESSITIES
- A42—HEADWEAR
- A42B—HATS; HEAD COVERINGS
- A42B1/00—Hats; Caps; Hoods
- A42B1/24—Hats; Caps; Hoods with means for attaching articles thereto, e.g. memorandum tablets or mirrors
-
- A—HUMAN NECESSITIES
- A42—HEADWEAR
- A42B—HATS; HEAD COVERINGS
- A42B1/00—Hats; Caps; Hoods
- A42B1/24—Hats; Caps; Hoods with means for attaching articles thereto, e.g. memorandum tablets or mirrors
- A42B1/242—Means for mounting detecting, signalling or lighting devices
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/145—Indicating the presence of current or voltage
- G01R19/155—Indicating the presence of voltage
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/539—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
Live working protective helmet based on random forest algorithm relates to electric wire netting live working safety tool field. The invention aims to solve the problem that the safety distance between personnel and equipment is controlled only by the experience of a worker in the existing power system maintenance work, and the potential safety hazard exists. According to the live working protective cap based on the random forest algorithm, the number of the obstacles is judged by the random forest algorithm through extracting the time domain and frequency domain information features of the ultrasonic echo signals, and the target distance is obtained. The invention can also alarm the nearby live-line body operation equipment on site, accurately measure the distance between the maintainer and the live-line body, and can send out an alarm in time when the measured distance is smaller than the safe distance.
Description
Technical Field
The invention belongs to the field of safety tools and appliances for live working of a power grid.
Background
In the routine maintenance and overhaul of the power system, it is often necessary to perform an adjacent live overhaul work. In order to enable the maintenance work to be successfully and safely completed, safety distances for hot-line work of various grades are specified in relevant specifications. At present, an accurate distance measurement prompt device is lacking in live working, so that the safety distance during maintenance mainly depends on observation and control of staff. The method for estimating the safe distance through experience can be influenced by uncertain factors in a complex field environment, so that the safe distance is wrongly judged, the requirement of accurate distance measurement cannot be met, and the distance between a worker or auxiliary equipment and high-voltage electrified equipment is too small, so that great potential safety hazards are brought.
Disclosure of Invention
The invention aims to solve the problem that the safety distance between personnel and equipment is controlled only by the experience of a worker in the existing power system maintenance work, and has great potential safety hazard, and provides a live working protective cap based on a random forest algorithm.
The live working protective cap based on the random forest algorithm comprises a protective cap body, an induction electricity testing module and an ultrasonic ranging module, wherein the induction electricity testing module and the ultrasonic ranging module are embedded and fixed between the inner cap body and the outer cap body of the protective cap;
the induction electroscope module comprises: the signal induction circuit is used for inducing the electromagnetic field intensity of the current environment and converting the magnetic field intensity into a voltage signal, the voltage signal output end of the signal induction circuit is connected with the voltage signal input end of the amplifying and filtering circuit, the voltage signal output end of the amplifying and filtering circuit is connected with the voltage signal input end of the rectifying circuit, and the voltage signal output end of the rectifying circuit is connected with the voltage signal input end of the detecting circuit;
the ultrasonic ranging module comprises: the device comprises an FPGA, a transmitting driving circuit, a transmitting probe, a receiving amplifying circuit and an A/D conversion circuit, wherein the transmitting signal output end of the FPGA is connected with the transmitting signal input end of the transmitting driving circuit, the driving signal input end of the transmitting driving circuit is connected with the driving signal input end of the transmitting probe, the transmitting probe emits ultrasonic waves after receiving the driving signal, the receiving probe receives ultrasonic wave echoes and converts ultrasonic wave signals into electric signals, the electric signal output end of the receiving probe is connected with the electric signal input end of the receiving amplifying circuit, the amplifying signal output end of the receiving amplifying circuit is connected with the analog signal input end of the A/D conversion circuit, the digital signal output end of the A/D conversion circuit is connected with the signal storage end of the FPGA, and the frequency control signal output end of the FPGA is connected with the frequency control signal input end of the A/D conversion circuit;
the FPGA comprises the following units realized by software:
and a storage unit: for storing the digital signals stored by the a/D conversion circuit,
and a control unit: for sending a write frequency control signal to the a/D conversion circuit,
a transmitting unit: for transmitting a drive signal to the transmit drive circuit,
NiosII processor:
extracting the characteristic values of all the digital signals stored in the storage unit, marking whether barriers exist in the corresponding time of each digital signal, taking the characteristic values of all the digital signals as input signals, taking the barrier marks of all the digital signals as output signals, training by using a random forest algorithm to obtain a random forest classifier,
extracting the characteristic value of the digital signal S stored in the storage unit at the current moment, inputting the characteristic value of the digital signal S into a random forest classifier to obtain an obstacle mark corresponding to the digital signal S,
when the obstacle mark exists in the time corresponding to the digital signal S, calculating the distance d=0.5 vt of the obstacle by using the time t corresponding to the voltage maximum value in the digital signal S, wherein v is the speed of the emitted wave.
The induction electroscope module further comprises an alarm circuit, and a voltage signal output end of the detection circuit is connected with a voltage signal input end of the alarm circuit.
The ultrasonic ranging module further comprises: temperature sensor and display still include in the FPGA: the temperature sensor is used for acquiring the temperature of the obstacle in front, sending the temperature of the obstacle to the NiosII processor through the temperature measuring unit, and sending the distance of the obstacle to the display through the display unit.
The amplification filter circuit includes: the power supply circuit comprises a resistor R1, a resistor R2, a resistor R3, a resistor R4, a capacitor C1, a capacitor C2 and an amplifier U1A, wherein one end of the resistor R1 is simultaneously connected with one end of the resistor R4, one end of the capacitor C1 and the inverting input end of the amplifier U1A, one end of the resistor R2 is simultaneously connected with one end of the resistor R3, one end of the capacitor C2 and the non-inverting input end of the amplifier U1A, the other end of the resistor R4, the other end of the capacitor C1 and the output end of the amplifier U1A are connected and serve as voltage signal output ends of an amplifying filter circuit, the other end of the resistor R2 serves as voltage signal input ends of the amplifying filter circuit, and the other end of the resistor R1, the other end of the resistor R3 and the other end of the capacitor C2 are all connected with the power supply ground.
The rectifier circuit includes: the resistor R5, the resistor R6, the resistor R7, the resistor R8, the capacitor C3, the amplifier U2A, the diode D1 and the diode D2, one end of the resistor R5 is connected with the positive electrode of the diode D1 and is used as a voltage signal input end of the rectifying circuit together, the other end of the resistor R5 is connected with one end of the resistor R6 and the inverting input end of the amplifier U2A simultaneously, the other end of the resistor R6 is connected with the output end of the amplifier U2A and the positive electrode of the diode D2 simultaneously, the negative electrode of the diode D1 is connected with the negative electrode of the diode D2 and one end of the resistor R7 simultaneously, the other end of the resistor R7 is connected with one end of the capacitor C3 and one end of the resistor R8 simultaneously and is used as a voltage signal output end of the rectifying circuit together, and the non-inverting input end of the amplifier U2A, the other end of the capacitor C3 and the other end of the resistor R8 are all connected with the power supply ground.
The detection circuit includes: resistor R9, varistor R10, resistor R11 and amplifier U3A, the noninverting input of amplifier U3A is as the voltage signal input of detection circuitry, and the positive pole of power is connected simultaneously to resistor R9's one end and resistor R11's one end, and varistor R10's active end and amplifier U3A's inverting input are connected simultaneously to resistor R9's the other end, and amplifier U3A's output is connected resistor R11's the other end to as the voltage signal output of detection circuitry, and varistor R10's an active end is connected power ground.
The method for extracting the characteristic value of the digital signal comprises the following steps:
dividing the digital signal into n segments by step i Dividing the ith signal segment into a plurality of subsections for step length, selecting the maximum voltage value a in each subsection as the characteristic value of the subsection, and combining the characteristic values of all subsections into a set { a } i N is a positive integer, i=1, 2,..n, the time domain feature CT of the i-th signal segment is calculated according to the following formula 1i ：
CT 1i ＝max{a i }-min{a i }
Wherein max { a } i Sum min { a } i Respectively is set { a } i Maximum and minimum values of the },
time domain feature CT 1i Fourier transforming to obtain frequency domain amplitude spectrum, and intercepting to transmit frequency f 0 Section [ f ] with center and left and right length d -d ,f +d ]Extracting frequency domain features of the ith signal segment in the interval according to the following formula, wherein the frequency domain features are spectrum relative area CT 2i Variance CT 3i CT of maximum value 4i Sum-of-range CT 5i ：
CT 4 ＝max{p(x j )}
CT 5 ＝max{p(x j )}-min{p(x j )}
Wherein x is j Is interval [ f -d ,f +d ]The frequency of any point, p (x j ) Is x j The amplitude of d is equal to or less than j is equal to or less than d,
and respectively carrying out normalization processing on the time domain features and the frequency domain features of the n signal segments to obtain feature values of the digital signals.
And respectively carrying out normalization processing on the time domain features and the frequency domain features of the n signal segments according to the following steps:
wherein X is max Is the time domain feature, the frequency spectrum relative area, the variance, the maximum value or the maximum value of the range in the n signal segments, X min X is the time domain feature, the frequency spectrum relative area, the variance, the maximum value or the minimum value of the range in the n signal segments old For the last momentAnd extracting the characteristic value of the digital signal.
The live working protective cap based on the random forest algorithm has the functions of wearing and distance measurement besides the electroscope function. Specifically, by extracting the time domain and frequency domain information characteristics of the ultrasonic echo signals, the number of the obstacles is judged by utilizing a random forest algorithm, and the target distance is obtained, so that the problem of difficult distance measurement caused by long distance and weak echo signals can be effectively solved. In practical application, the distance measuring function of a plurality of barriers within 10m can be realized, the measuring error is within +/-3 cm, and the practical application requirement is met. Meanwhile, the invention can also alarm the nearby live-line body operation equipment on site, and compared with the traditional vision distance measurement and stay wire early warning mode, the invention can accurately measure the distance between the maintainer and the live-line body, and can give an alarm in time when the measured distance is smaller than the safety distance, thereby avoiding the occurrence of safety accidents such as damage of the maintainer and the equipment.
Drawings
FIG. 1 is a schematic structural view of a hot-line work helmet;
FIG. 2 is a circuit diagram of an inductive electroscope module;
FIG. 3 is a block diagram of an ultrasonic ranging module;
FIG. 4 is a time domain diagram of a digital signal;
FIG. 5 is a segmented Fourier transform frequency domain plot;
fig. 6 is a schematic diagram of an obstacle test.
Detailed Description
The first embodiment is as follows: referring to fig. 1 to 6, a specific description is given of the present embodiment, which is a live working helmet based on a random forest algorithm, and includes a helmet body, and an induction electroscope module and an ultrasonic ranging module embedded between an inner helmet body and an outer helmet body of the helmet. The two modules are positioned at the center of the top end of the protective cap, the module shell adopts an integrated design, good insulation and mechanical strength are ensured, battery power is adopted, and the module is arranged as shown in figure 1.
As shown in fig. 2, the induction electroscope module includes: the signal induction circuit 1-1, the amplifying and filtering circuit 1-2, the rectifying circuit 1-3, the detecting circuit 1-4 and the alarm circuit 1-5, wherein the signal induction circuit 1-1 is used for inducing the electromagnetic field intensity of the current environment and converting the magnetic field intensity into a voltage signal, the voltage signal output end of the signal induction circuit 1-1 is connected with the voltage signal input end of the amplifying and filtering circuit 1-2, the voltage signal output end of the amplifying and filtering circuit 1-2 is connected with the voltage signal input end of the rectifying circuit 1-3, the voltage signal output end of the rectifying circuit 1-3 is connected with the voltage signal input end of the detecting circuit 1-4, and the voltage signal output end of the detecting circuit 1-4 is connected with the voltage signal input end of the alarm circuit 1-5.
The amplification filter circuit 1-2 includes: the power supply circuit comprises a resistor R1, a resistor R2, a resistor R3, a resistor R4, a capacitor C1, a capacitor C2 and an amplifier U1A, wherein one end of the resistor R1 is simultaneously connected with one end of the resistor R4, one end of the capacitor C1 and the inverting input end of the amplifier U1A, one end of the resistor R2 is simultaneously connected with one end of the resistor R3, one end of the capacitor C2 and the non-inverting input end of the amplifier U1A, the other end of the resistor R4, the other end of the capacitor C1 and the output end of the amplifier U1A are connected and serve as voltage signal output ends of the amplifying filter circuits 1-2, the other end of the resistor R2 serves as voltage signal input ends of the amplifying filter circuits 1-2, and the other end of the resistor R1, the other end of the resistor R3 and the other end of the capacitor C2 are all connected with the power supply ground.
The rectifying circuit 1-3 includes: the voltage signal input terminal of the rectifier circuit 1-3 is commonly connected with one end of the resistor R5 and the positive electrode of the diode D1, one end of the resistor R5 is simultaneously connected with one end of the resistor R6 and the inverting input terminal of the amplifier U2A, the other end of the resistor R6 is simultaneously connected with the output terminal of the amplifier U2A and the positive electrode of the diode D2, the negative electrode of the diode D1 is simultaneously connected with the negative electrode of the diode D2 and one end of the resistor R7, the other end of the resistor R7 is simultaneously connected with one end of the capacitor C3 and one end of the resistor R8 and is commonly used as the voltage signal output terminal of the rectifier circuit 1-3, and the non-inverting input terminal of the amplifier U2A, the other end of the capacitor C3 and the other end of the resistor R8 are all connected with power ground.
The detection circuits 1 to 4 include: resistor R9, varistor R10, resistor R11 and amplifier U3A, the noninverting input of amplifier U3A is as the voltage signal input of detection circuit 1-4, and the positive pole of power is connected simultaneously to one end of resistor R9 and one end of resistor R11, and the other end of resistor R9 is connected simultaneously to varistor R10's active end and amplifier U3A's inverting input, and amplifier U3A's output is connected resistor R11's the other end to as the voltage signal output of detection circuit 1-4, and varistor R10's an inactive end is connected power ground.
When the line is charged, the signal sensing circuit 1-1 converts the magnetic field signal into an electrical signal, embodied as a voltage signal Vin that generates an alternating current variation. Since the line is charged, in addition to the 50Hz magnetic field, there is also a mixture of harmonic interference and other noise interference. Therefore, the voltage signal Vin needs to be further amplified and filtered. The ac voltage signal Vin generated by the signal sensing circuit 1-1 is amplified and filtered forward by the op-amp to obtain the voltage signal Vac, and the filter in this embodiment is a second-order low-pass filter, so that high-frequency noise interference can be filtered. The rectification circuit 1-3 adopts full-wave rectification, and is conducted and filtered to Vrec through D1 when the voltage signal Vac is a positive half wave; when Vac is a negative half wave, r6=r5 is amplified by the same proportion by the opposite phase, and is turned on by D2 and filtered to Vrec. Under the action of the first-order RC low-pass filter, vrec is a small-pulsation direct-current voltage and is sent into an alarm circuit for detection. When the input signal Vrec is larger than a preset threshold voltage, vout is high level, the characterization equipment is electrified or is close to the electrified equipment, the buzzer S1 is conducted, and then an alarm is given; when the input Vrec is smaller than the preset voltage, vout is low level, the power failure of the equipment is represented or the equipment is far away from the electrified equipment, the buzzer S1 is not conducted, and no alarm is given. The comparator sends out 2 kinds of signals of power on and power off according to the magnitude of the input signal. The magnitude of the resistance value of the rheostat R10 can be adjusted to adjust the preset voltage value, so that the sensitivity of the device is changed, the buzzer frequency is influenced by the distance between the rheostat R10 and a charged body, and the closer the distance is, the higher the buzzer frequency is. When in use, a worker wearing the device can change the sensitivity of the alarm according to different adjustment gears of the voltage class of the live site.
As shown in fig. 3, the ultrasonic ranging module includes: FPGA2-2, emission drive circuit 2-6, emission probe 2-7, receiving probe 2-8, receiving amplifier circuit 2-5, A/D converting circuit 2-4, temperature sensor and display. The transmitting driving circuit 2-6 is a MAX232 chip, the A/D conversion circuit 2-4 is a MAX120 chip, and the model of the FPGA2-2 is EP2C35F672 of Altera. The transmitting signal output end of the FPGA2-2 is connected with the transmitting signal input end of the transmitting driving circuit 2-6, the driving signal input end of the transmitting driving circuit 2-6 is connected with the driving signal input end of the transmitting probe 2-7, and the transmitting probe 2-7 sends out 8 square wave pulse signals of 40kHz after receiving the driving signal. The receiving probe 2-8 receives ultrasonic echoes and converts ultrasonic signals into electric signals, the electric signal output end of the receiving probe 2-8 is connected with the electric signal input end of the receiving amplifying circuit 2-5, the amplifying signal output end of the receiving amplifying circuit 2-5 is connected with the analog signal input end of the A/D converting circuit 2-4, the digital signal output end of the A/D converting circuit 2-4 is connected with the signal storage end of the FPGA2-2, the frequency control signal output end of the FPGA2-2 is connected with the frequency control signal input end of the A/D converting circuit 2-4, the temperature sensor is used for collecting the temperature of a front obstacle and sending the temperature of the obstacle to the NiosII processor through the temperature measuring unit, and the NiosII processor sends the distance of the obstacle to the display through the display unit.
The FPGA2-2 comprises the following software-implemented units:
and a storage unit: for storing the digital signals stored in the a/D conversion circuit 2-4,
and a control unit: for transmitting a write frequency control signal to the a/D conversion circuit 2-4,
a transmitting unit: for transmitting a drive signal to the transmit drive circuit 2-6,
NiosII processor:
extracting the characteristic values of all the digital signals stored in the storage unit, marking whether barriers exist in the corresponding time of each digital signal, taking the characteristic values of all the digital signals as input signals, taking the barrier marks of all the digital signals as output signals, training by using a random forest algorithm to obtain a random forest classifier,
extracting the characteristic value of the digital signal S stored in the storage unit at the current moment, inputting the characteristic value of the digital signal S into a random forest classifier to obtain an obstacle mark corresponding to the digital signal S,
when the digital signal S corresponds to the time when the obstacle mark exists, calculating the distance d=0.5 vt of the obstacle by using the time t corresponding to the voltage maximum value in the digital signal S, wherein v is the speed of the emitted wave.
When a plurality of continuous digital signals are detected in the detection process and the time period corresponding to the digital signals with the obstacle marks is continuous, merging the digital signals with continuous time into an integral time interval, and calculating the distance d of the obstacle by using the time t corresponding to the voltage maximum value in the time interval.
When the ultrasonic ranging module works, the transmitting circuit is started, the storage unit and the A/D conversion circuit 2-4 are started at the same time, converted data are read in from the on-chip SRAM by the NiosII processor, and a specific time delay algorithm is realized after data format conversion and band-pass filtering are carried out; the NiosII reads in the real-time temperature measured by the temperature measuring module and combines the time delay obtained by the data processing to calculate a specific distance value, and the result is output through the display driving circuit.
The signal has characteristics of a time domain and a frequency domain. In the time domain, the signal f (t) is a function of time, describing the amplitude, frequency and phase of the signal as a function of time. In the frequency domain, the signal F (j ω) is a function of frequency, and the amplitude and phase of the signal are discussed as a function of frequency. The signal may be transformed between the time and frequency domains by a fourier transform. It has been found through research that when the ultrasonic wave encounters an obstacle and returns, the frequency ff of the reflected wave is at the frequency f of the emitted wave 0 Occurs nearby, resulting in a frequency f 0 The nearby spectral amplitude becomes large. For example, at a frequency f 0 The pulse square wave is emitted at 48kHz, resulting in the band-pass filtered original echo signal waveform S, as shown in fig. 4. It can be seen that there is one obstacle in each of the periods 18200 μs to 19100 μs and 55200 μs to 56100 μs. For ranging, the highest point is taken as the flight time in the two time periods, and the distances between the two obstacles can be calculated to be about 3.2m and 9.5m according to the formula d=0.5vt. Therefore, finding the period of time in which the obstacle is located is a key to solving the problem. Thus, the above-mentioned liftingThe method for taking the characteristic value of the digital signal comprises the following steps:
dividing the digital signal (echo signal) into n segments, each segment corresponding to time t 0 ，t 0 The size of the obstacle in the echo signal corresponds to the time period t from the starting time to the ending time of the occurrence obs Related to, for example, t in FIG. 4 obs =900 μs, if t 0 If the value is too large, a plurality of barriers are contained in one section, and the number of the barriers is influenced. Otherwise, if the value is too small, the time period t for an obstacle to appear is caused obs Being divided into successive segments complicates the computation. So take t generally 0 Suitably, about 12 tobs. Such an obstacle will be at most covered by a succession of 3 segments t 0 If it is determined that an obstacle appears in the adjacent time period in the subsequent processing, the two time periods can be combined into one time period. Select good t 0 After that, the total duration of the echo signal is tall, then n=tall/t 0 . Then FFT is carried out on each section of signal, when an obstacle appears, f 0 The frequency spectrum amplitude nearby is stronger, as shown in fig. 5, the graph shows only the segments nearby the time intervals of the two obstacles in fig. 4, and it can be seen that the frequency spectrum amplitude of 43 segments and 44 segments of the time intervals of the obstacles is obviously stronger than that of 42 segments and 45 segments; similarly, the spectral amplitude of 128 segments and 129 segments of another obstacle is higher than that of 127 segments and 130 segments.
Based on the above analysis, some characteristics of the signal in the time and frequency domains can be extracted. With a segment S in the digital signal S i For example, the calculation method comprises the following steps:
by step i Dividing the ith signal segment into a plurality of subsections for step length, selecting the maximum voltage value a in each subsection as the characteristic value of the subsection, and combining the characteristic values of all subsections into a set { a } i N is a positive integer, i=1, 2,..n, the time domain feature CT of the i-th signal segment is calculated according to the following formula 1i ：
CT 1i ＝max{a i }-min{a i }
Wherein max { a } i Sum min { a } i Respectively is set { a } i Maximum and minimum values of the },
time domain feature CT 1i Fourier transforming to obtain frequency domain amplitude spectrum, and intercepting to transmit frequency f 0 Section [ f ] with center and left and right length d -d ,f +d ]Extracting frequency domain features of the ith signal segment in the interval according to the following formula, wherein the frequency domain features are spectrum relative area CT 2i Variance CT 3i CT of maximum value 4i Sum-of-range CT 5i ：
CT 4 ＝max{p(x j )}
CT 5 ＝max{p(x j )}-min{p(x j )}
Wherein x is j Is interval [ f -d ,f +d ]The frequency of any point, p (x j ) Is x j The amplitude of d is equal to or less than j and is equal to or less than d, and the interval [ f ] -d ,f +d ]The magnitudes corresponding to the frequencies of all points in (a) are combined into a set { p (x) j ) Then there is max { p (x) j ) Sum min { p (x) j ) Respectively, is set { p (x) j ) Maximum and minimum values in }.
Respectively carrying out normalization processing on the time domain features and the frequency domain features of the n signal segments according to the following steps of:
wherein X is max Is the time domain feature, the frequency spectrum relative area, the variance, the maximum value or the maximum value of the range in the n signal segments, X min For the time domain features, the relative areas of the frequency spectrums, the variances, the maxima or the minimum of the range in the n signal segments,X old is the characteristic value of the digital signal extracted at the last moment.
Experiments were performed according to the above embodiment, in which the oscillation frequency of the pulse square wave emitted was 48kHz, the sampling frequency was 680MHz, and the sampling time was 62638.55 mus. The original data points after band-pass filtration are collected, the number of sampling points is 43200, and the drawn waveform is similar to that of fig. 4. Dividing the original data into 144 segments, t 0 About 500 μs, 300 data points per segment, and the frequency domain interval is selected to be [46kHz,50kHz]The distance of the obstacle is calculated according to the algorithm of the present embodiment. The experimental results obtained by measuring a plurality of different obstacles at different distances are shown in table 1. The results show that the present embodiment can effectively complete distance measurement of a plurality of obstacles at the same time within 10 m. The error is within +/-3 cm, so that the practical application requirement is met. Multiple measurements were made at the same location to evaluate the accuracy of the random forest algorithm, and the experimental results are shown in table 2. The result shows that the algorithm has good obstacle recognition degree.
Table 1 measurement results
Table 2 random forest algorithm accuracy
The method can realize the electroscope function, and has the wearable and ranging functions, and the problem of difficult ranging caused by long distance and weak echo signals can be effectively solved by extracting time domain and frequency domain information characteristics of ultrasonic echo signals through an algorithm and then judging the number of obstacles and solving the target distance through a random forest algorithm. The distance measuring function of a plurality of barriers within 10m can be realized, the measuring error is within +/-3 cm, the practical application requirement is met, the good effect is achieved, and the device has high practical value and theoretical reference significance.
Through the on-the-spot test to neighbouring live body operation equipment alarm device, compare with traditional stadia measurement and act as go-between early warning mode, the device can accurately measure the distance between maintainer and the live body to can in time send out the alarm when measuring the distance and be less than safe distance, avoid the emergence of the incident such as damage of maintainer and equipment.
Claims (9)
1. The live working protective cap based on the random forest algorithm comprises a protective cap body and is characterized by further comprising an induction electricity testing module and an ultrasonic ranging module which are embedded and fixed between the inner cap body and the outer cap body of the protective cap;
the induction electroscope module comprises: the device comprises a signal induction circuit (1-1), an amplifying and filtering circuit (1-2), a rectifying circuit (1-3) and a detecting circuit (1-4), wherein the signal induction circuit (1-1) is used for inducing the electromagnetic field intensity of the current environment and converting the magnetic field intensity into a voltage signal, the voltage signal output end of the signal induction circuit (1-1) is connected with the voltage signal input end of the amplifying and filtering circuit (1-2), the voltage signal output end of the amplifying and filtering circuit (1-2) is connected with the voltage signal input end of the rectifying circuit (1-3), and the voltage signal output end of the rectifying circuit (1-3) is connected with the voltage signal input end of the detecting circuit (1-4);
the ultrasonic ranging module comprises: the FPGA (2-2), the transmission driving circuit (2-6), the transmission probe (2-7), the receiving probe (2-8), the receiving amplifying circuit (2-5) and the A/D converting circuit (2-4), wherein the transmission signal output end of the FPGA (2-2) is connected with the transmission signal input end of the transmission driving circuit (2-6), the driving signal input end of the transmission driving circuit (2-6) is connected with the driving signal input end of the transmission probe (2-7), the transmission probe (2-7) emits ultrasonic waves after receiving the driving signals, the receiving probe (2-8) receives ultrasonic wave echoes and converts ultrasonic signals into electric signals, the electric signal output end of the receiving probe (2-8) is connected with the electric signal input end of the receiving amplifying circuit (2-5), the amplified signal output end of the receiving amplifying circuit (2-5) is connected with the analog signal input end of the A/D converting circuit (2-4), the digital signal output end of the A/D converting circuit (2-4) is connected with the signal storage end of the FPGA (2-2), and the frequency control signal output end of the FPGA (2-2) is connected with the frequency control signal input end of the A/D converting circuit (2-4).
The FPGA (2-2) comprises the following units realized by software:
and a storage unit: for storing the digital signals stored in the A/D conversion circuit (2-4),
and a control unit: for transmitting a write frequency control signal to the A/D conversion circuit (2-4),
a transmitting unit: for transmitting a drive signal to the transmit drive circuit (2-6),
NiosII processor:
extracting the characteristic values of all the digital signals stored in the storage unit, marking whether barriers exist in the corresponding time of each digital signal, taking the characteristic values of all the digital signals as input signals, taking the barrier marks of all the digital signals as output signals, training by using a random forest algorithm to obtain a random forest classifier,
extracting the characteristic value of the digital signal S stored in the storage unit at the current moment, inputting the characteristic value of the digital signal S into a random forest classifier to obtain an obstacle mark corresponding to the digital signal S,
when the obstacle mark exists in the corresponding time of the digital signal S, calculating the distance d=0.5 vt of the obstacle by using the time t corresponding to the maximum voltage value in the digital signal S, wherein v is the speed of the emitted wave;
the method for extracting the characteristic value of the digital signal comprises the following steps:
dividing the digital signal into n segments by step i Dividing the ith signal segment into a plurality of subsections for step length, selecting the maximum voltage value a in each subsection as the characteristic value of the subsection, and combining the characteristic values of all subsections into a set { a } i N is a positive integer, i=1, 2,..n, the time domain feature CT of the i-th signal segment is calculated according to the following formula 1i ：
CT 1i ＝max{a i }-min{a i }
Wherein max { a } i Sum min { a } i Respectively is set { a } i Maximum and minimum values of the },
time domain feature CT 1i Fourier transforming to obtain frequency domain amplitude spectrum, and intercepting to transmit frequency f 0 Section [ f ] with center and left and right length d -d ,f +d ]Extracting frequency domain features of the ith signal segment in the interval according to the following formula, wherein the frequency domain features are spectrum relative area CT 2i Variance CT 3i CT of maximum value 4i Sum-of-range CT 5i ：
CT 4i ＝max{p(x j )}
CT 5i ＝max{p(x j )}-min{p(x j )}
Wherein x is j Is interval [ f -d ,f +d ]The frequency of any point, p (x j ) Is x j The amplitude of d is equal to or less than j is equal to or less than d,
and respectively carrying out normalization processing on the time domain features and the frequency domain features of the n signal segments to obtain feature values of the digital signals.
2. The live working helmet based on the random forest algorithm according to claim 1, wherein the induction electroscope module further comprises an alarm circuit (1-5), and a voltage signal output end of the detection circuit (1-4) is connected with a voltage signal input end of the alarm circuit (1-5).
3. The live working helmet based on random forest algorithm according to claim 1 or 2, wherein the ultrasonic ranging module further comprises: the temperature sensor and the display are also included in the FPGA (2-2): a display unit and a temperature measuring unit,
the temperature sensor is used for collecting the temperature of the obstacle in front and sending the temperature of the obstacle to the NiosII processor through the temperature measuring unit,
the NiosII processor sends the distance of the obstacle to the display through the display unit.
4. Live working helmet based on random forest algorithm according to claim 1, characterized in that the amplifying filter circuit (1-2) comprises: resistor R1, resistor R2, resistor R3, resistor R4, capacitor C1, capacitor C2 and amplifier U1A,
one end of the resistor R1 is simultaneously connected with one end of the resistor R4, one end of the capacitor C1 and the inverting input end of the amplifier U1A,
one end of the resistor R2 is simultaneously connected with one end of the resistor R3, one end of the capacitor C2 and the non-inverting input end of the amplifier U1A,
the other end of the resistor R4, the other end of the capacitor C1 and the output end of the amplifier U1A are connected and used as the voltage signal output end of the amplifying and filtering circuit (1-2),
the other end of the resistor R2 is used as a voltage signal input end of the amplifying and filtering circuit (1-2),
the other end of the resistor R1, the other end of the resistor R3 and the other end of the capacitor C2 are all connected with power ground.
5. Live working helmet based on random forest algorithm according to claim 1, characterized in that the rectifying circuit (1-3) comprises: resistor R5, resistor R6, resistor R7, resistor R8, capacitor C3, amplifier U2A, diode D1 and diode D2,
one end of the resistor R5 is connected with the anode of the diode D1 and is commonly used as a voltage signal input end of the rectifying circuit (1-3),
the other end of the resistor R5 is connected to both one end of the resistor R6 and the inverting input terminal of the amplifier U2A,
the other end of the resistor R6 is connected with the output end of the amplifier U2A and the anode of the diode D2 at the same time,
the cathode of the diode D1 is connected to both the cathode of the diode D2 and one end of the resistor R7,
the other end of the resistor R7 is connected with one end of the capacitor C3 and one end of the resistor R8 at the same time and is used as the voltage signal output end of the rectifying circuit (1-3) together,
the non-inverting input terminal of the amplifier U2A, the other end of the capacitor C3 and the other end of the resistor R8 are all connected with power ground.
6. Live working helmet based on random forest algorithm according to claim 1, characterized in that the detection circuit (1-4) comprises: resistor R9, varistor R10, resistor R11 and amplifier U3A,
the non-inverting input terminal of the amplifier U3A is used as the voltage signal input terminal of the detection circuit (1-4),
one end of the resistor R9 and one end of the resistor R11 are simultaneously connected with the positive electrode of the power supply,
the other end of the resistor R9 is connected to both the active end of the varistor R10 and the inverting input of the amplifier U3A,
the output end of the amplifier U3A is connected with the other end of the resistor R11 and is used as the voltage signal output end of the detection circuit (1-4),
one stationary end of the varistor R10 is connected to power ground.
7. The live working helmet based on random forest algorithm according to claim 1, wherein the normalization processing is performed on the time domain features and the frequency domain features of the n signal segments according to the following formula:
wherein X is max Is the time domain feature, the frequency spectrum relative area, the variance, the maximum value or the maximum value of the range in the n signal segments, X min X is the time domain feature, the frequency spectrum relative area, the variance, the maximum value or the minimum value of the range in the n signal segments old Is the characteristic value of the digital signal extracted at the last moment.
8. The live working helmet based on random forest algorithm according to claim 1, wherein when the time period corresponding to the digital signal with the obstacle mark is continuous, the digital signals with continuous time are combined into a whole time interval, and the distance d=0.5 vt of the obstacle is calculated by using the time t corresponding to the maximum voltage value in the time interval.
9. The live working helmet based on random forest algorithm according to claim 1, wherein the emission driving circuit (2-6) is a MAX232 chip, the a/D conversion circuit (2-4) is a MAX120 chip, and the model number of the FPGA (2-2) is EP2C35F672.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010844921.1A CN111965652B (en) | 2020-08-20 | 2020-08-20 | Live working protective cap based on random forest algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010844921.1A CN111965652B (en) | 2020-08-20 | 2020-08-20 | Live working protective cap based on random forest algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111965652A CN111965652A (en) | 2020-11-20 |
| CN111965652B true CN111965652B (en) | 2024-03-12 |
Family
ID=73389639
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010844921.1A Active CN111965652B (en) | 2020-08-20 | 2020-08-20 | Live working protective cap based on random forest algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111965652B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112471670A (en) * | 2020-10-30 | 2021-03-12 | 国网天津市电力公司 | Intelligent insulating cap for live working |
| CN113432250B (en) * | 2021-06-08 | 2022-10-11 | Tcl空调器(中山)有限公司 | Air conditioner anti-pinch control method and device, air conditioner and readable storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB0030728D0 (en) * | 1999-12-27 | 2001-01-31 | Ando Electric | Probe signal outputting apparatus |
| US6202034B1 (en) * | 1999-01-27 | 2001-03-13 | Delphi Technologies, Inc. | Ultrasonic ranging system and method for improving accuracy thereof |
| WO2005106530A1 (en) * | 2004-04-28 | 2005-11-10 | Matsushita Electric Industrial Co., Ltd. | Ultrasonic range finder |
| CN101886744A (en) * | 2010-06-25 | 2010-11-17 | 东北大学 | Ultrasonic positioning device and positioning method of portable internal pipeline fault detection equipment |
| CN104483675A (en) * | 2014-12-18 | 2015-04-01 | 扬州天目光电科技有限公司 | Self-adaptive laser ranging device and self-adaptive laser ranging method |
| CN104851239A (en) * | 2015-06-09 | 2015-08-19 | 国网山东济南市历城区供电公司 | Alarm device and method for safety distance of high-voltage electrical equipment |
| JP2016027332A (en) * | 2014-07-01 | 2016-02-18 | 東日本旅客鉄道株式会社 | Non-contact electroscope and method for detecting electricity |
| JP2017173150A (en) * | 2016-03-24 | 2017-09-28 | カシオ計算機株式会社 | Obstacle detection unit, method, and program |
| CN110070031A (en) * | 2019-04-18 | 2019-07-30 | 哈尔滨工程大学 | A kind of sediment extracting echo characteristics of active sonar fusion method based on EMD and random forest |
-
2020
- 2020-08-20 CN CN202010844921.1A patent/CN111965652B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6202034B1 (en) * | 1999-01-27 | 2001-03-13 | Delphi Technologies, Inc. | Ultrasonic ranging system and method for improving accuracy thereof |
| GB0030728D0 (en) * | 1999-12-27 | 2001-01-31 | Ando Electric | Probe signal outputting apparatus |
| WO2005106530A1 (en) * | 2004-04-28 | 2005-11-10 | Matsushita Electric Industrial Co., Ltd. | Ultrasonic range finder |
| CN101886744A (en) * | 2010-06-25 | 2010-11-17 | 东北大学 | Ultrasonic positioning device and positioning method of portable internal pipeline fault detection equipment |
| JP2016027332A (en) * | 2014-07-01 | 2016-02-18 | 東日本旅客鉄道株式会社 | Non-contact electroscope and method for detecting electricity |
| CN104483675A (en) * | 2014-12-18 | 2015-04-01 | 扬州天目光电科技有限公司 | Self-adaptive laser ranging device and self-adaptive laser ranging method |
| CN104851239A (en) * | 2015-06-09 | 2015-08-19 | 国网山东济南市历城区供电公司 | Alarm device and method for safety distance of high-voltage electrical equipment |
| JP2017173150A (en) * | 2016-03-24 | 2017-09-28 | カシオ計算機株式会社 | Obstacle detection unit, method, and program |
| CN110070031A (en) * | 2019-04-18 | 2019-07-30 | 哈尔滨工程大学 | A kind of sediment extracting echo characteristics of active sonar fusion method based on EMD and random forest |
Non-Patent Citations (3)
| Title |
|---|
| Range Measurement using Ultrasound FMCW Signals;Masanori Kunita et al.;《2008 IEEE Ultrasonics Symposium》;第1-4页 * |
| 基于ZigBee的临近带电体安全距离报警装置研制;朱明柱 等;《通信电源技术》;第34卷(第01期);第70-72页 * |
| 大量程超声波回波测距***的研究;刘先中 等;《仪器仪表学报》;第25卷(第4期);第1-4页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111965652A (en) | 2020-11-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111965652B (en) | Live working protective cap based on random forest algorithm | |
| CN101655538B (en) | Generator local discharge on-line monitoring device and monitoring method thereof | |
| CN104076195B (en) | Non-contact electricity verification device based on vertical field intensity difference distribution curve matching | |
| CN105911499B (en) | Ultrasonic wave shelf depreciation metering system and method under site environment | |
| CN106124939A (en) | Distributed high tension cable partial discharge monitoring and alignment system | |
| CN103675482A (en) | Complex electrical field measuring system | |
| CN208847788U (en) | A kind of non-contact type high-voltage detecting device based on field strength distribution | |
| CN103823165A (en) | Insulator pollution flashover pre-warning method and system based on leakage currents | |
| CN104155568B (en) | A kind of transmission line of lightning strike lightning conducter accurate positioning method | |
| CN104625332B (en) | Stability evaluation system and method based on carbon dioxide welding input end electric signal | |
| CN110361606A (en) | Leakage conductor on-line monitoring system | |
| CN112782540A (en) | High-voltage cable on-line monitoring and fault point positioning device | |
| CN110084986B (en) | Perimeter security method and device | |
| CN117353465B (en) | Sensor-based operation supervision system for variable-frequency power distribution cabinet | |
| CN104655941A (en) | Electromagnetic radiation detection method and device and electronic equipment | |
| CN206959908U (en) | High-tension switch cabinet busbar electric current, temperature integrated measurer | |
| CN115508614B (en) | Airborne non-contact type high-voltage electricity testing method and system | |
| CN204008823U (en) | Contactless charged identification electroscope | |
| CN105277795A (en) | Grounding impedance monitoring collecting analyzer | |
| CN213750105U (en) | Real-time thunder and lightning early warning device | |
| CN114459532A (en) | Passive wireless partial discharge and temperature and humidity composite sensing monitoring system | |
| CN111913182A (en) | Transformer substation operation inspection robot and live area isolation method thereof | |
| CN206945831U (en) | A kind of mechanical arm safe distance monitoring device | |
| CN206420992U (en) | A kind of offline inspection positioner of high-tension cable potential problems | |
| CN206132970U (en) | Ultrasonic wave partial discharge measurement system under site environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |