CN114217124A - Fusion terminal with loop inspection function - Google Patents
Fusion terminal with loop inspection function Download PDFInfo
- Publication number
- CN114217124A CN114217124A CN202111337748.7A CN202111337748A CN114217124A CN 114217124 A CN114217124 A CN 114217124A CN 202111337748 A CN202111337748 A CN 202111337748A CN 114217124 A CN114217124 A CN 114217124A
- Authority
- CN
- China
- Prior art keywords
- module
- circuit
- layer
- output
- threshold
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R22/00—Arrangements for measuring time integral of electric power or current, e.g. electricity meters
- G01R22/06—Arrangements for measuring time integral of electric power or current, e.g. electricity meters by electronic methods
- G01R22/061—Details of electronic electricity meters
- G01R22/066—Arrangements for avoiding or indicating fraudulent use
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Current Or Voltage (AREA)
Abstract
The invention discloses a fusion terminal with a loop inspection function. Wherein, this terminal includes: the loop inspection module is configured to detect the circuit state of the secondary loop circuit and analyze the circuit state to obtain loop inspection related characteristic parameters so as to judge whether the ammeter is in a normal state or not; and the main control core module is configured to be connected with the loop inspection module through a USB interface, is used for processing the detection result of the loop inspection module and controls the display module to display the processed detection result. The invention solves the technical problem that the convergence terminal cannot accurately detect the electricity stealing behavior.
Description
Technical Field
The invention relates to the field of intelligent AI (intelligent input/output), in particular to a fusion terminal with a loop inspection function.
Background
Electricity is the most convenient and useful form of energy for modern people, without which the current social infrastructure will not be feasible. When the power consumption is continuously increased, the electricity price is continuously raised, and the condition that the electric power is stolen from a transmission line or the electricity is illegally taken appears. Moreover, the electricity stealing means has the characteristics of high technological content and strong concealment.
Most of the existing power inspection modes are field survey by inspection personnel, but the survey results are often unsatisfactory, and the power supply reliability and the power resource safety are threatened.
At present, except that a special transformer acquisition terminal and a concentrator in an electricity information acquisition system can carry out limited monitoring on an electric energy meter, equipment and a system for effectively monitoring electricity stealing behaviors with hidden means and high technological content are urgently needed.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The embodiment of the invention provides a convergence terminal with a loop inspection function, which at least solves the technical problem that the convergence terminal cannot accurately detect electricity stealing behaviors.
According to an aspect of the embodiments of the present invention, there is provided a convergence terminal having a loop polling function, including: the loop inspection module is configured to detect the circuit state of the secondary loop circuit and analyze the circuit state to obtain loop inspection related characteristic parameters so as to judge whether the ammeter is in a normal state or not; and the main control core module is configured to be connected with the loop inspection module through a USB interface, is used for processing the detection result of the loop inspection module and controls the display module to display the processed detection result.
According to another aspect of the embodiments of the present invention, there is also provided a loop inspection method, configured to detect a circuit state of a secondary loop circuit, and analyze a loop inspection related characteristic parameter based on the circuit state to determine whether the electric meter is in a normal state.
In the embodiment of the invention, the technical problem that the fusion terminal cannot accurately detect the electricity stealing behavior is solved by adopting a mode of detecting the circuit state of the secondary circuit and analyzing to obtain the circuit patrol related characteristic parameters based on the circuit state to judge whether the ammeter is in a normal state.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
fig. 1 is a schematic diagram of a convergence terminal having a loop inspection function according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of another converged terminal with loop polling functionality according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a loop inspection module according to an embodiment of the invention;
fig. 4 is a schematic structural diagram of a detection unit of the loop inspection module according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a resonance unit of the loop inspection module according to an embodiment of the present invention;
FIG. 6 is a flow chart of a method of electricity stealing behavior identification according to an embodiment of the invention;
FIG. 7 is a flow chart of another electricity stealing behavior identification method according to an embodiment of the invention;
FIG. 8 is a flow diagram of a method of constructing a neural network model in accordance with an embodiment of the present invention;
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
According to an embodiment of the present invention, a schematic diagram of a convergence terminal with a loop inspection function is provided, as shown in fig. 1, the convergence terminal includes a loop inspection module 10 and a main control core module 12.
The loop inspection module 10 is configured to detect a circuit state of a secondary loop circuit, and analyze the circuit state to obtain loop inspection related characteristic parameters so as to judge whether the ammeter is in a normal state;
and the main control core module 12 is configured to be connected with the loop inspection module through a USB interface, and is used for processing the detection result of the loop inspection module and controlling the display module to display the processed detection result.
In one exemplary embodiment, the loop inspection module includes: an analog signal processing section configured to detect a circuit state of the secondary loop circuit, the circuit state including a current state and/or a voltage state; a digital signal processing section configured to analyze the loop inspection related characteristic parameter based on the current state and the voltage state.
In one exemplary embodiment, the analog signal processing part includes: a detection unit configured to acquire a circuit state of the secondary loop circuit through a coil; a voltage applying unit configured to inject a voltage signal having a frequency greater than a first voltage threshold to the secondary loop circuit based on the circuit state; a resonance unit configured to resonate when the secondary loop circuit is injected with a voltage signal having a frequency greater than a first voltage threshold, and generate an oscillation signal based on the generated resonance signal, wherein the oscillation signal is used to identify whether there is a power stealing behavior.
In one exemplary embodiment, the resonance unit includes: a resonance circuit that resonates and generates a resonance signal when a voltage signal having a frequency greater than a first voltage threshold is injected into the secondary circuit; a self-oscillation circuit configured to generate an oscillation signal based on the generated resonance signal; a correction circuit configured to correct the oscillation signal and send the corrected oscillation signal to the digital processing section to identify whether there is a power stealing behavior.
In one exemplary embodiment, the correction circuit includes: a negative input end of the fourth comparator is connected with an output end of the fourth comparator, an output end of the fourth comparator is connected with a connection point of a resistor R31 and a resistor R32 which are connected in series through a capacitor C38, and an output end of the fourth comparator is further connected with one end of a capacitor C34; the positive input end of the fifth comparator is connected with the other end of the capacitor C34, the negative input end of the fifth comparator is connected with the connection point of the resistor R28 and the resistor R29 which are connected in series, the output end of the fifth comparator is connected with the other end of the resistor R28 through the resistor R29, and the output end of the fifth comparator is grounded through the capacitor C33.
In one exemplary embodiment, the digital signal processing part includes: a model building module configured to pre-build a neural network model for identifying electricity stealing behavior based on a neural network; an identification module configured to identify the electricity stealing behavior based on the constructed neural network model.
In an exemplary embodiment, the model building module is further configured to: correcting the weight value and the threshold value of the neural network model; and generating a neural network model based on the corrected weight and the threshold.
In an exemplary embodiment, the convergence terminal further includes an interleaving module for acquiring data of the distribution transformer, a clock module for synchronizing clocks, a watchdog module for monitoring operation of the main control core module, a power supply module for supplying power, a communication isolation module for protecting the main control core module, a temperature measurement module for measuring temperature, and a display module for displaying.
Example 2
According to an embodiment of the present invention, an intelligent convergence terminal is provided, as shown in fig. 2, the intelligent convergence terminal includes: the loop inspection module 10, the main control core module 12, the power line carrier module 11, the USB interface 25, the Bluetooth module 13, the temperature measurement module 14, the watchdog module 15, the safety module 16, the power supply module 17, the handover module 18, the clock module 19, the communication isolation module 20, the Ethernet module 21, the 4G module 22, the display module 23, the human-computer interaction module 24, the RS-485 interface 201, the RS-232 interface 202 and the remote communication interface 203.
The main control core module is a key module of the intelligent fusion terminal and is responsible for management and control of each module of the intelligent fusion terminal. The master control CORE module can select a CORE board SCM701CORE, based on the design of the high-end master control chip SCM701, a 2GB memory and a 4GB flash memory are configured, rich peripheral interfaces are provided, the working master frequency reaches 1.2GHz, an intelligent CORE is carried on a hub OS and a container engine of an independently controllable safe operating system, and the edge computing capability is achieved.
A data acquisition chip of the alternating current acquisition module can select a multifunctional high-precision electricity larceny prevention three-phase electric energy special metering chip ATT7022EU to acquire data of the distribution transformer, 7 circuits of second-order sigma-delta ADC are integrated on the chip ATT7022EU, wherein 3 circuits are three-phase current sampling channels (A phase, B phase and C phase), 3 circuits are three-phase voltage sampling channels (A phase, B phase and C phase), and ESD protection circuits are arranged inside positive pins and negative pins of each channel.
The microcontroller of the alternating current collection module can adopt an STM32F103RCT6 microcontroller for processing and analyzing collected electric quantity data. The controller uses a high-performance cortex-M332-bit RISC core, the working frequency is 72MHz, 256K byte Flash storage is provided, two debugging modes of serial single-wire debugging (SWD) and JTAG interfaces are supported, the controller comprises 3 12-bit ADCs, 4 universal 16-bit timers and 2 PWM timers, and simultaneously comprises 2I 2C, 3 SPI, 2I 2S, 1 SDIO, 5 USART, a USB, a CAN and other standard communication interfaces, and the data storage, analysis and processing requirements of the intelligent fusion terminal CAN be met.
The clock module adopts an RX-8025T clock chip of Epson company to realize the real-time function of the intelligent convergence terminal.
The watchdog module is used for monitoring the internal operation state of the terminal and automatically resetting under the condition of program runaway or deadlock. The watchdog module chip can adopt an SP706EN chip, a main control core module feeds dogs at regular time, and when an external reset switch is pressed, the power supply voltage of the SP706EN chip is reduced to 4.4V by 5V or the core board program runs off and is deadlocked, the whole system starts power-on reset.
The power supply module is responsible for converting the external input voltage of the intelligent fusion terminal into the voltage required by each module, and the normal and stable work of each functional module is ensured.
In order to ensure the safety and reliability of the whole system of the intelligent convergence terminal and prevent the interference, influence and even damage of the fluctuation of current and voltage to each chip of the terminal, the communication isolation is realized by the communication isolation module to the RS-485 interface, the RS-232 interface and the remote signaling interface.
The temperature measurement module realizes temperature measurement, is internally provided with electric isolation, and can ensure that a stable measurement result is not interfered.
The safety module realizes safety certification of the intelligent fusion terminal, including data encryption, decryption, certification and the like, and guarantees safety and reliability of data storage and transmission processes.
The display module realizes the working state display and the function display of the intelligent fusion terminal, wherein the working state comprises a power supply state, an operation state, a communication state, a connection state with a main station and the like.
The USB interface module is connected with the loop inspection module.
The loop inspection module comprises an analog signal processing part and a digital signal processing part. The digital signal processing part utilizes an internal ADC in the STM32F405RGT6 to complete detection of three paths of analog output quantities, frequency and amplitude calculation is completed through wavelet transformation, and finally related characteristic parameters of routing inspection of the comprehensive analysis loop are analyzed. The loop inspection module will be described in detail below, and will not be described in detail here.
Example 3
In the present embodiment, a convergence terminal having a loop patrol function is provided, and the configuration of the convergence terminal is different from that of the convergence terminal in the foregoing embodiments in the loop patrol module.
Fig. 3 is a schematic structural diagram of a loop inspection module according to an embodiment of the present invention, and as shown in fig. 3, the loop inspection module includes an analog signal processing part 32 and a digital signal processing part 34. The analog signal processing section 32 includes a detection unit 321, a voltage application unit 322, a resonance unit 323, and a secondary loop circuit 324, wherein the secondary loop circuit 324 includes a sampling circuit and a current transformer.
The primary side of the current transformer is connected with a power supply, and the secondary side of the current transformer is connected with the sampling circuit to form a secondary loop circuit. The detection unit is connected with the digital signal processing part and detects the load current and/or voltage of the secondary circuit under the control of the digital signal processing part.
Fig. 4 is a circuit structure diagram of the detecting unit according to the embodiment of the invention, and as shown in fig. 4, the coil L1 is connected in parallel with the sampling resistor R41, and after being connected in parallel, one end of L1 and one end of R41 are grounded, and the other end is connected in series with the resistor R42. One end of the resistor R42 is connected to the positive input of the first comparator. One end of the resistor R36 is grounded, the other end of the resistor R37 is connected with the other end of the resistor R36, the connection point of the resistor R37 and the resistor R37 is connected with the negative input end of the first comparator, the other end of the resistor R37 is connected with the capacitor C43 in series, and the other end of the capacitor C43 is connected with the connection point of the resistors R41 and R42. The input end of the first comparator is connected to the positive input end of the second comparator, one end of the capacitor C39 is grounded, the other end of the capacitor C39 is connected in series with the resistor R38, and the other end of the resistor R38 is connected to the negative input end of the second comparator and is connected to the output end of the second comparator through the resistor R39. The negative output terminal of the second comparator is connected to the digital signal processing section via a resistor R40, and one end of the resistor R40 adjacent to the digital signal processing section is also connected to ground via a capacitor C41.
With the above circuit configuration, the load current of the secondary circuit can be detected. The secondary coil of the current transformer induces the load current of the secondary circuit, the sampling resistor R41 converts the current induced by the current transformer into corresponding voltage, the sampled voltage is processed to enable the sampled voltage to be more accurate and then input into the digital signal processing part, and the processed voltage is converted by the analog-to-digital converter integrated in the digital signal processing part.
In the present embodiment, only an example of sampling the load current is shown. In other embodiments, the voltage may be sampled in addition to the load current. In case of simultaneous sampling of current and voltage, the internally integrated analog-to-digital converter of the digital signal processing part may comprise a compensator for compensating the phase shift between the analog voltage and current signals in order to obtain voltage and current samples corresponding to voltage and current signals being substantially in phase. Compensation is achieved by sample time adjustment compared to the analog voltage and current signals to provide substantially in-phase voltage and current samples.
The voltage applying unit is connected to the digital signal processing part, and injects a high frequency voltage signal, which is a voltage signal having a voltage value greater than a first voltage threshold value and has the same frequency as the resonance frequency of the resonance unit, into the secondary loop circuit under the control of the digital signal processing part.
Example 4
In the present embodiment, there is provided a convergence terminal having a loop patrol function, which is different from the configuration of the convergence terminal in the foregoing embodiment 3 in the resonance unit in the loop patrol module.
As shown in fig. 3, the loop inspection module in this embodiment includes an analog signal processing section and a digital signal processing section. The analog signal processing part comprises a detection unit, a voltage applying unit, a resonance unit and a secondary loop circuit, wherein the secondary loop circuit comprises a sampling circuit and a current transformer.
The detection unit is used for detecting the current state and/or the voltage state of the secondary circuit. The voltage applying unit in this embodiment may include a voltage transformer and a voltage applying circuit. Under the control of the digital signal processing part, when the pressurizing circuit injects a high-frequency voltage signal into the secondary loop circuit, the frequency is controlled by the digital signal processing part. The voltage applying circuit drives and amplifies the voltage signal which is input by the digital signal processing part and used for controlling the frequency to obtain a high-frequency voltage signal, and then the generated high-frequency voltage signal is injected into the secondary circuit through the induction of the voltage transformer coil.
Fig. 5 is a circuit configuration diagram of a resonance unit according to an embodiment of the present invention. The resonance unit is connected with the digital signal processing part and is used for generating resonance when the secondary loop circuit injects a high-frequency voltage signal. As shown in fig. 5, a capacitor C35 is connected in series with the capacitor C36, the other end of the capacitor C36 is connected to the positive input terminal of the third comparator, one end of a coil L2 is connected to the capacitor C35, and the other end is connected to the capacitor C36, so that the coil L2 is connected in parallel with the capacitors C35 and C36 which are connected in series. The coil L2 and the capacitor C35 and the capacitor C36 connected in series constitute a resonance unit. One end of the inverter INV is connected to the input end of the C35, and the other end is connected to the output end of the third comparator, so that a self-oscillation circuit is formed. The positive input end of the third comparator is also connected with a resistor R34 connected to the ground. The other end of the resistor R26 which is grounded is connected with the resistor R27, the connection point of the resistor R26 and the resistor R27 is connected with the negative input end of the third comparator, and the other end of the resistor R27 is connected with the output end of the third comparator. The inverter INV is connected to one end of the capacitor C35, and the other end is connected to the output end of the first comparator, thereby forming a self-oscillating circuit. One end of the resistor R235 is connected to the output end of the third comparator, and the other end is connected to the connection point of the capacitor C35 and the capacitor C36. The input end of the third comparator is connected with the resistor R31, the other end of the resistor R31 is connected with the resistor R32, the other end of the resistor R32 is connected with the fourth comparator, and the input end of the fourth comparator is further connected with the ground through the capacitor C37. The negative input terminal of the fourth comparator is connected to the output terminal of the fourth comparator, and the output terminal of the fourth comparator is further connected to the connection point of the resistor R31 and the resistor R32 through the capacitor C38. The output end of the fourth comparator is also connected with a capacitor C34, and the other end of the capacitor C34 is connected with the positive input end of the fifth comparator. One end of the capacitor C32 is grounded, the other end of the capacitor C32 is connected with the resistor R28, the other end of the resistor R28 is connected to the output end of the fifth comparator through the resistor R29, and the connection point of the resistor R28 and the resistor R29 is connected with the negative input end of the fifth comparator. The output terminal of the fifth comparator is grounded via a capacitor C33, and the connection point of the fifth comparator and the capacitor C33 is connected with the digital signal processing part. The correction circuit constituted by the devices related to the fourth comparator and the fifth comparator is used to correct the output value from the oscillation circuit.
Specifically, when the secondary circuit operates, the coil L1 senses a load current of the secondary circuit, the sampling resistor converts the current sensed by the current transformer into a corresponding voltage, and the voltage is converted by an analog-to-digital converter integrated in the digital signal processing part after being processed. The digital signal processing section judges whether or not a load current of the secondary circuit is lower than a first current threshold.
The voltage applying unit injects a high frequency voltage signal to the secondary circuit if a load current of the secondary circuit is lower than a first current threshold. After that, the digital signal processing section judges whether or not the oscillation signal is detected from the resonance unit. When injecting high-frequency voltage signals into the secondary loop circuit, because the frequency of the injected high-frequency voltage signals is the same as the resonant frequency of the resonant unit, if the secondary loop circuit is not open-circuited, the resonant unit will be in a resonant state, at the moment, the high-frequency voltage signals are subjected to resonant amplification to obtain oscillation signals, and the amplitude of the oscillation signals is corrected by the correction circuit and then collected by the digital circuit part. Therefore, if the corrected oscillation signal is not detected, it is determined that the secondary loop circuit is open. If the oscillation signal is detected, the digital signal processing part further judges whether the amplitude of the corrected oscillation signal is higher than a preset voltage threshold. And if the amplitude of the corrected oscillation signal is higher than a preset voltage threshold, determining that the secondary loop circuit is short-circuited. And if the amplitude of the oscillation signal is not higher than the preset voltage threshold, determining that the secondary loop circuit works normally.
And starting the self-oscillation circuit if the load current of the secondary circuit is not lower than the first current threshold. The oscillation frequency of the self-oscillation circuit exhibits periodic increases and decreases with changes in current. The digital signal processing section judges whether the corrected minimum frequency of the self-oscillation circuit increases or decreases. When the secondary loop circuit is shorted, the minimum frequency may increase. Therefore, if the corrected minimum frequency increase is detected, it can be judged that the secondary loop circuit is shorted, i.e., electricity theft may occur in the electricity meter.
The digital signal processing part is used for realizing data acquisition and analyzing based on the acquired data. The digital signal processing part is used for providing a control signal with required frequency to the voltage applying unit, collecting the load current and/or load voltage of the secondary circuit, the minimum frequency of the resonant circuit and the oscillation signal when the resonant unit resonates, which are detected by the detecting unit, and processing the collected data to judge whether the electricity meter is stolen.
Through the structure, the correction circuit comprising the fourth comparator and the fifth comparator is arranged in the resonance unit, so that the oscillation signal output from the oscillation circuit is corrected, the amplitude of the oscillation signal acquired by the digital circuit part is more accurate, and whether the electric meter is in a state of being stolen or not can be judged more accurately.
Example 5
In the present embodiment, there is provided a convergence terminal having a loop patrol function, which differs from the configuration of the convergence terminal in the foregoing embodiment 4 in the digital signal processing section.
As shown in fig. 3, the loop inspection module in this embodiment includes an analog signal processing section and a digital signal processing section. The analog signal processing part comprises a detection unit, a voltage applying unit, a resonance unit and a secondary loop circuit, wherein the secondary loop circuit comprises a sampling circuit and a current transformer.
The detection unit is used for detecting the current state and/or the voltage state of the secondary circuit. The resonance unit includes a resonance circuit and a self-oscillation circuit. Under the control of the digital signal processing part, when the pressurizing circuit injects a high-frequency voltage signal into the secondary loop circuit, the frequency is controlled by the digital signal processing part. The voltage applying circuit drives and amplifies the voltage signal which is input by the digital signal processing part and used for controlling the frequency to obtain a high-frequency voltage signal, and then the generated high-frequency voltage signal is injected into the secondary circuit through the induction of the voltage transformer coil. The resonance unit is connected with the digital signal processing part and is used for generating resonance when the secondary loop circuit injects a high-frequency voltage signal.
The digital signal processing section in the present embodiment executes a method of identifying a fraudulent use of electricity. The recognition of the anti-electricity-stealing behavior is realized through the pre-constructed electricity-stealing behavior recognition model. And extracting and analyzing relevant data and abnormal data of the electricity stealing behavior in a characteristic data input stage of the model. And after the sample data is preprocessed, training the recognition model by utilizing the sample data. And the sample data is used as input, the classification of the output samples is realized, and the electricity stealing identification is realized. And establishing a power stealing identification model, and then realizing real-time diagnosis by using the model.
The electricity stealing behavior recognition model in the embodiment is a model established based on a neural network system. Specifically, the method for identifying electricity stealing behavior is shown in fig. 6, and includes the following steps:
step S601, setting the weight and the threshold of the BP neural network.
The number of hidden layers of neurons was set to 2. The number of neurons in an input layer is n, the number of neurons in an implicit layer is l, the number of neurons in an output layer is m, the maximum training time is 5000, the preset index threshold value is epsilon-0.02, and the learning rate is eta.
The weight from the input layer to the hidden layer is wijThe weight from the hidden layer to the output layer is wjkThe threshold of the hidden layer is ajThe threshold value of the output layer is bkSetting each weight value wij、wjkAnd a threshold value aj、bkIs a random number between 0 and 1.
Step S602, providing a training sample.
Inputting historical electricity utilization data xiE.g. corrected oscillator signal, desired output being YkThe following iterations of steps S603 to S604 are performed for each input sample.
Step S603, calculating the output of the network hidden layer and the output of the output layer.
Calculating the actual output of the hidden layer of the BP neural network based on the corresponding output value of the previous input layer of the hidden layer, the corresponding weight values of the hidden layer and the previous input layer, the threshold value of the hidden layer and an activation function; and calculating the actual output of the BP neural network output layer based on the corresponding output value of the upper hidden layer of the output layer, the corresponding weight values of the output layer and the upper hidden layer and the threshold value of the output layer.
The specific calculation method is as follows:
g(x)＝1/(1+e-x)
wherein, i is 1 … n, j is 1 … l, k is 1 … m, n is the number of input layer neurons, l is the number of hidden layer neurons, m is the number of output layer neurons, w is the number of output layer neurons, and the likeijAs weights of the input layer to the hidden layer, wjkAs weights from hidden layer to output layer, ajThreshold for the hidden layer, bkFor the threshold of the output layer, g (x) is an activation function, g (x) is a Sigmoid function in the form of
HjFor the actual output of the hidden layer, OkIs the actual output of the output layer, xiIs the input of the BP neural network.
If the current layer network is the output layer, the Ok value is the last output value of the network.
And step S604, correcting the weight value and the threshold value.
Updating the weight from the hidden layer to the output layer:
updating the weight from the input layer to the hidden layer:
hidden layer to output layer threshold update:
input layer to hidden layer threshold update:
wherein, wijFor the weights of the input layer to the hidden layer,
is wijUpdated weight, wjkThe weights for the hidden layer to the output layer,
is wjkUpdated weight, ajIn order to imply the threshold of the layer,
is ajThe updated threshold value is set to a value that is less than the threshold value,bk is a threshold value of the output layer,
is b iskUpdated threshold, η is learning rate, HjFor the actual output of the hidden layer, OkIs the actual output of the output layer, xiAs input to a BP neural network, ek＝Yk-OkM is the number of neurons in the output layer, YkIs the desired output.
And step S605, calculating the electricity stealing index.
And step S606, judging whether the electricity stealing index meets the electricity stealing condition.
And judging whether the electricity stealing index meets the index requirement or not. And when E is less than epsilon, confirming that the index requirement is met, namely the electricity stealing condition of the electricity meter occurs.Wherein E is an electricity stealing index, YkAnd the Ok value is the output value of the output layer for the expected output, and the epsilon is 0.02 which is a preset electricity stealing index.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
Example 6
In the present embodiment, there is provided a convergence terminal having a loop patrol function, which differs from the configuration of the convergence terminal in the foregoing embodiment 4 in the digital signal processing section.
As shown in fig. 3, the loop inspection module in this embodiment includes an analog signal processing section and a digital signal processing section. The analog signal processing part comprises a detection unit, a voltage applying unit, a resonance unit and a secondary loop circuit, wherein the secondary loop circuit comprises a sampling circuit and a current transformer.
The detection unit is used for detecting the current state and/or the voltage state of the secondary circuit. The voltage applying unit in this embodiment may include a voltage transformer and a voltage applying circuit. The primary coil of the voltage transformer is a lead of a secondary circuit, one end of the secondary coil of the voltage transformer is connected with the output end of the voltage applying circuit, and the other end of the secondary coil of the voltage transformer is grounded. The resonance unit includes a resonance circuit and a self-oscillation circuit.
Under the control of the digital signal processing part, when the pressurizing circuit injects a high-frequency voltage signal into the secondary loop circuit, the frequency is controlled by the digital signal processing part. The voltage applying circuit drives and amplifies the voltage signal which is input by the digital signal processing part and used for controlling the frequency to obtain a high-frequency voltage signal, and then the generated high-frequency voltage signal is injected into the secondary circuit through the induction of the voltage transformer coil. The resonance unit is connected with the digital signal processing part and is used for generating resonance when the secondary loop circuit injects a high-frequency voltage signal.
The digital signal processing part in this embodiment is different from the above-described embodiment, and is identified by another electricity stealing identification method.
Extracting features from smart meter data based on experience is difficult. However, feature extraction is a key factor in detecting the success or failure of a system. In this embodiment, a CNN network is used. In CNN, successive alternating convolutional and pooling layers aim to learn higher level features (e.g., trend indicators, sequence standard deviations, and linear slopes) step by step through historical power consumption data. Furthermore, there is a highly non-linear correlation between power consumption and these influencing factors. Since the activation function is designed on both the convolutional layer and the fully-connected layer, the CNN is able to model highly non-linear dependencies. In this embodiment, an activation function named "rectified Linear Unit" (ReLU) is used because of its sparsity and minimizes the gradient vanishing problem in the CNN-RF model.
The electricity stealing identification method executed by the digital signal processing section in the present embodiment is shown in fig. 7, and includes the following steps:
step S701, preprocessing the acquired data.
1) And eliminating abnormal data.
xi，tDefined as the i-th current collected during the t time interval,
is the average current value over the time interval t. In this embodiment, the digital signal processing section first rejects abnormal data using the following formula.
In that
Expressed as abnormal data, the data is eliminated, and an average value process, a mean square process, and a mean value supplemented with a correction factor are substituted for the abnormal data, where σ represents the mean square and W represents the weight.
2) Missing data is supplemented.
Due to various reasons such as storage problems and faults of the intelligent electric meter, the power consumption data sometimes have missing values. Through analysis of the raw data, two types of data loss are found: one is continuous deletion of a plurality of pieces of data, and the solution is to delete the data when the number of deletion values exceeds a preset threshold value, for example, 1O; the other is missing single data, and the missing value is recovered according to the following formula:
wherein NaN represents a deletion.
3) And (6) normalization processing.
Finally, the data needs to be normalized because neural networks are sensitive to different data. The normalization process can be performed using the following equation:
wherein, T represents the time period, T is a multiple of the time interval T, and n represents the number of the sample data collected in the time interval T.
Step S702, providing a training sample.
The data set was partitioned into a training set and a test set using a cross-validation method, where 80% was the training set and 20% was the test set. Given that the number of electricity stealing consumers significantly exceeds that of non-fraudulent consumers, an imbalance in the data set can have a significant negative impact on the performance of the supervised machine learning approach. To reduce this bias, the number of normal and abnormal samples in the training set is made equal.
Step S703, training a neural network model.
A neural network model is first established. The embodiment related to fig. 7 will explain the establishment of the neural network model in detail, and will not be described herein again.
The neural network model is trained using a forward propagation algorithm and a back propagation algorithm. Firstly, data in an input layer is transferred to a convolutional layer, a pooling layer and a full-link layer to obtain a predicted value. If the output value differs too much from the target value and exceeds a certain threshold, the back propagation phase starts updating the parameters. The difference between the actual output and the expected output of the neural network determines the direction of adjustment of the weights of the various layers in the network.
Given a sample set D { (x) of m samples in total1，y1)，...，xm，ym) Is first obtained by a feed forward process
. For all samples, the actual output yiAnd expected output
The mean of the difference of (a) may be expressed as:
where a denotes the learning rate, W denotes the connection weight between network layers, and b is the corresponding deviation.
Each parameter is then initialized with a random value generated with a mean of 0 and a variance of normal distribution and updated with a gradient descent method. And correcting the weight and the deviation, wherein a specific correction formula is as follows:
wherein, a represents the learning rate,
represents a connection weight between the ith neuron in the 1 st layer and the jth neuron in the 1+1 st layer,
indicates the deviation of the ith neuron in the l-th layer.
Then, a classifier is trained. The classification process of the classifier randomly generates a method of specific guide samples from the learned features for growing each tree.
Step S704, whether electricity is stolen is identified.
And inputting the collected data into a neural network model to determine whether electricity stealing behavior occurs.
The embodiment provides a method for identifying electricity stealing based on a neural network model. In the model, the electricity stealing behavior can be better identified through the designed automatic feature extractor and classifier. Furthermore, a fully connected layer is designed during the training phase, since having to optimize a large number of parameters increases the risk of overfitting. In addition, the problem of data imbalance is overcome.
Example 7
The architecture of the neural network model provided by the embodiment mainly comprises a feature extractor and a trainable classifier. The feature extractor is composed of a convolution layer, a down-sampling layer and a full-connection layer. The neural network model is a deep supervised learning architecture, typically comprising multiple layers, that can be trained using back propagation algorithms. It can also explore complex distributions in smart meter data by performing a stochastic gradient method. The classifier is composed of a combination of tree classifiers.
The process of building the neural network model proposed in this embodiment is shown in fig. 8, and may include the following steps:
s801, building a convolutional layer.
The main purpose of the convolutional layer is to learn the feature representation of the input data and reduce the influence of noise. The convolutional layer consists of several feature filters for computing different feature maps. In particular, to reduce network parameters and reduce complexity of the relevant layers in generating the feature maps, the weight of each core in each feature map is shared. In addition, each neuron in the convolutional layer connects to a local region of the previous layer, and then applies an activation function to the convolutional result. In this implementation, the output of the convolutional layer can be expressed as:
where δ is the activation function, and δ is the convolution operation,
and
are learnable parameters in the f-th feature filter,
is the collected data of the input of the f-th feature filter. Ft is the total number of feature process filters.
S802, constructing a down-sampling layer.
The downsampling layer is usually placed between two convolutional layers, which can reduce the number of parameters and achieve dimensionality reduction. In particular, each feature map of a pooling layer is connected to the feature map of its corresponding previous convolutional layer. Further, the size of the output feature map is reduced without reducing the number of output feature maps in the down-sampling layer. The downsampling operation mainly comprises maximum pooling and average pooling. The max pooling operation converts the widget to a single value by a maximum value, but the average pooling returns the average of the activations in the widget.
In this implementation, the output of the downsampling layer can be expressed as:
where M is a set of activation values in the pooling window, M is an activation index in the pooling window, 1 is a layer ID, Xi，mIs the input to the downsampling layer.
S803, constructing a full connection layer.
The feature map is flattened into a vector by applying a full-connected layer through the features extracted by sequential convolution and pooling, as follows:
wherein, WlIs the weight of layer 1,/l) Is the deviation of the first layer, XlIs the input to layer 1.
S804, constructing a classifier.
The classifier is used for the last output layer of the neural network model, and the classification is predicted according to the obtained characteristics. The classifier layer may be defined as:
yout(XL)sigm(Wrf·X1+bl)+Xlblδ
wherein, sigm is a sigmoid function, and the abnormal value is mapped into 0, and the normal value is mapped into 1. Parameter set W set in classifierrfIncluding the number of decision trees and the maximum depth of the trees, which are obtained by a trellis search algorithm.
The neural network model constructed for identifying electricity stealing behaviors provided by the embodiment has the following beneficial effects: hybrid models can automatically extract features, while the success of most other traditional classifiers relies mainly on retrieving good hand-designed features; the hybrid model combines the advantages of neural networks and classifiers.
Example 8
Embodiments of the present invention also provide a storage medium having a program stored thereon, which when executed, causes a computer to execute any of the methods in the above embodiments.
Optionally, in this embodiment, the storage medium may include, but is not limited to: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The embodiment of the present application can also realize the following configuration:
the utility model provides a fuse terminal with function is patrolled and examined to return circuit, includes: the loop inspection module is configured to detect the circuit state of the secondary loop circuit and analyze the circuit state to obtain loop inspection related characteristic parameters so as to judge whether the ammeter is in a normal state or not; and the main control core module is configured to be connected with the loop inspection module through a USB interface, is used for processing the detection result of the loop inspection module and controls the display module to display the processed detection result.
2 the convergence terminal according to item 1, the loop inspection module comprises: an analog signal processing section configured to detect a circuit state of the secondary loop circuit, the circuit state including a current state and/or a voltage state; a digital signal processing section configured to analyze the loop inspection related characteristic parameter based on the current state and the voltage state.
3. The convergence terminal according to item 2, wherein the analog signal processing section comprises: a detection unit configured to acquire a circuit state of the secondary loop circuit through a coil; a voltage applying unit configured to inject a voltage signal having a frequency greater than a first voltage threshold to the secondary loop circuit based on the circuit state; a resonance unit configured to resonate when the secondary loop circuit is injected with a voltage signal having a frequency greater than a first voltage threshold, and generate an oscillation signal based on the generated resonance signal, wherein the oscillation signal is used to identify whether there is a power stealing behavior.
4. The convergence terminal of item 3, wherein the resonance unit comprises: a resonance circuit that resonates and generates a resonance signal when a voltage signal having a frequency greater than a first voltage threshold is injected into the secondary circuit; a self-oscillation circuit configured to generate an oscillation signal based on the generated resonance signal; a correction circuit configured to correct the oscillation signal and send the corrected oscillation signal to the digital processing section to identify whether there is a power stealing behavior.
5. The convergence terminal of item 4, wherein the correction circuit comprises: a negative input end of the fourth comparator is connected with an output end of the fourth comparator, an output end of the fourth comparator is connected with a connection point of a resistor R31 and a resistor R32 which are connected in series through a capacitor C38, and an output end of the fourth comparator is further connected with one end of a capacitor C34; the positive input end of the fifth comparator is connected with the other end of the capacitor C34, the negative input end of the fifth comparator is connected with the connection point of the resistor R28 and the resistor R29 which are connected in series, the output end of the fifth comparator is connected with the other end of the resistor R28 through the resistor R29, and the output end of the fifth comparator is grounded through the capacitor C33.
6. The convergence terminal according to item 2, wherein the digital signal processing section comprises: a model building module configured to pre-build a neural network model for identifying electricity stealing behavior based on a neural network; an identification module configured to identify the electricity stealing behavior based on the constructed neural network model.
7. The convergence terminal of item 6 wherein the model building module is further configured to: correcting the weight value and the threshold value of the neural network model; and generating a neural network model based on the corrected weight and the threshold.
8. The converged terminal of item 6, wherein the weight and threshold are modified based on the following formula:
updating the weight from the hidden layer to the output layer:
updating the weight from the input layer to the hidden layer:
hidden layer to output layer threshold update:
input layer to hidden layer threshold update:
wherein, wijFor the weights of the input layer to the hidden layer,
is wijUpdated weight, wjkThe weights for the hidden layer to the output layer,
is wjkUpdated weight, ajIn order to imply the threshold of the layer,
is ajThe updated threshold value is set to a value that is less than the threshold value,bk is a threshold value of the output layer,
is b iskUpdated threshold, η is learning rate, HjFor the actual output of the hidden layer, OkIs the actual output of the output layer, xiAs input to a BP neural network, ek＝Yk-OkM is the number of neurons in the output layer, YkIs the desired output.
9. The convergence terminal of item 8 wherein the neural network model is generated based on the following formula:
wherein, i is 1 … n, j is 1 … l, k is 1 … m, n is the number of input layer neurons, l is the number of hidden layer neurons, m is the number of output layer neurons, w is the number of output layer neurons, and the likeijAs weights of the input layer to the hidden layer, wjkAs weights from hidden layer to output layer, ajThreshold for the hidden layer, bkFor the threshold of the output layer, g (x) is an activation function, g (x) is a Sigmoid function in the form of
HjFor the actual output of the hidden layer, OkIs the actual output of the output layer, xiIs the input of the BP neural network.
10. The convergence terminal according to item 1, further comprising an acquisition module for acquiring data of the distribution transformer, a clock module for synchronizing clocks, a watchdog module for monitoring the operation of the master control core module, a power module for providing power, a communication isolation module for protecting the master control core module, a temperature measurement module for measuring temperature, and a display module for displaying.
The embodiment of the present application can also realize the following configuration:
1. a method of identifying electricity stealing behavior, comprising: acquiring a circuit state of a secondary circuit, and analyzing to obtain a circuit inspection related characteristic parameter based on the circuit state; and identifying the electricity stealing behavior based on the analyzed loop inspection related characteristic parameters.
2. The method of item 1, wherein identifying power stealing behavior based on the analyzed loop inspection related characteristic parameters comprises: inputting the loop inspection related characteristic parameters into a pre-established neural network model; based on the neural network model, electricity stealing behavior is identified according to the loop inspection related characteristic parameters.
3. The method of item 1, wherein the neural network model is created by: preprocessing a training sample; correcting the weight value and the threshold value of the neural network model; building the neural network model based on the corrected weights and thresholds and the processed training samples.
4. The method of item 3, wherein preprocessing the training samples comprises: removing abnormal data from the training samples; finding missing data from the training sample with the abnormal data removed, and supplementing the missing data; and carrying out normalization processing on the training samples supplemented with the missing data.
5. The method according to item 4, characterized in that the normalization process is carried out based on the following formula:
wherein T represents a time period, T is a multiple of a time interval T, n represents the number of sample data collected within the time interval T, xi，tRepresenting the i-th current, x, collected during the t time intervali，TRepresenting the ith current collected during the T time period.
6. An apparatus for identifying electricity stealing behavior, comprising: the detection module is configured to detect the circuit state of the secondary circuit and analyze the circuit state to obtain the relevant characteristic parameters of the loop inspection; an identification module configured to identify a power stealing behavior based on the analyzed loop inspection related characteristic parameter.
7. The apparatus of item 5, wherein the identification module is further configured to: inputting the loop inspection related characteristic parameters into a pre-established neural network model; based on the neural network model, electricity stealing behavior is identified according to the loop inspection related characteristic parameters.
8. The apparatus of item 7, further comprising a model building module configured to: preprocessing a training sample; correcting the weight value and the threshold value of the neural network model; building the neural network model based on the corrected weights and thresholds and the processed training samples.
9. The apparatus of item 8, wherein the model building module is further configured to: removing abnormal data from the training samples; finding missing data from the training sample with the abnormal data removed, and supplementing the missing data; and carrying out normalization processing on the training samples supplemented with the missing data.
10. The apparatus according to item 8, characterized in that the normalization process is performed based on the following formula:
wherein T represents a time period, T is a multiple of a time interval T, n represents the number of sample data collected within the time interval T, xi，tRepresenting the i-th current, x, collected during the t time intervali，TRepresenting the ith current collected during the T time period.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (10)
1. The utility model provides a fuse terminal with function is patrolled and examined to return circuit which characterized in that includes:
the loop inspection module is configured to detect the circuit state of the secondary loop circuit and analyze the circuit state to obtain loop inspection related characteristic parameters so as to judge whether the ammeter is in a normal state or not;
and the main control core module is configured to be connected with the loop inspection module through a USB interface, is used for processing the detection result of the loop inspection module and controls the display module to display the processed detection result.
2. The convergence terminal of claim 1, wherein the loop inspection module comprises:
an analog signal processing section configured to detect a circuit state of the secondary loop circuit, the circuit state including a current state and/or a voltage state;
a digital signal processing section configured to analyze the loop inspection related characteristic parameter based on the current state and the voltage state.
3. The convergence terminal of claim 2 wherein the analog signal processing section comprises:
a detection unit configured to acquire a circuit state of the secondary loop circuit through a coil;
a voltage applying unit configured to inject a voltage signal having a frequency greater than a first voltage threshold to the secondary loop circuit based on the circuit state;
a resonance unit configured to resonate when the secondary loop circuit is injected with a voltage signal having a frequency greater than a first voltage threshold, and generate an oscillation signal based on the generated resonance signal, wherein the oscillation signal is used to identify whether there is a power stealing behavior.
4. The convergence terminal of claim 3 wherein the resonating unit comprises:
a resonance circuit that resonates and generates a resonance signal when a voltage signal having a frequency greater than a first voltage threshold is injected into the secondary circuit;
a self-oscillation circuit configured to generate an oscillation signal based on the generated resonance signal;
a correction circuit configured to correct the oscillation signal and send the corrected oscillation signal to the digital processing section to identify whether there is a power stealing behavior.
5. The convergence terminal of claim 4 wherein the correction circuit comprises:
a negative input end of the fourth comparator is connected with an output end of the fourth comparator, an output end of the fourth comparator is connected with a connection point of a resistor R31 and a resistor R32 which are connected in series through a capacitor C38, and an output end of the fourth comparator is further connected with one end of a capacitor C34;
the positive input end of the fifth comparator is connected with the other end of the capacitor C34, the negative input end of the fifth comparator is connected with the connection point of the resistor R28 and the resistor R29 which are connected in series, the output end of the fifth comparator is connected with the other end of the resistor R28 through the resistor R29, and the output end of the fifth comparator is grounded through the capacitor C33.
6. The convergence terminal of claim 2 wherein the digital signal processing section comprises:
a model building module configured to pre-build a neural network model for identifying electricity stealing behavior based on a neural network;
an identification module configured to identify the electricity stealing behavior based on the constructed neural network model.
7. The convergence terminal of claim 6 wherein the model building module is further configured to:
correcting the weight value and the threshold value of the neural network model;
and generating a neural network model based on the corrected weight and the threshold.
8. The converged terminal of claim 6, wherein the weights and thresholds are modified based on the following formula:
updating the weight from the hidden layer to the output layer:
updating the weight from the input layer to the hidden layer:
hidden layer to output layer threshold update:
input layer to hidden layer threshold update:
wherein, wijFor the weights of the input layer to the hidden layer,
is wijUpdated weight, wjkThe weights for the hidden layer to the output layer,
is wjkUpdated weight, ajIn order to imply the threshold of the layer,
is ajUpdated threshold, bkIs the threshold value of the output layer,
is b iskUpdated threshold, η is learning rate, HjFor the actual output of the hidden layer, OkIs the actual output of the output layer, xiAs input to a BP neural network, ek＝Yk-OkM is the number of neurons in the output layer, YkIs the desired output.
9. The convergence terminal of claim 8 wherein the neural network model is generated based on the following formula:
wherein, i is 1 … n, j is 1 … l, k is 1 … m, n is the number of input layer neurons, l is the number of hidden layer neurons, m is the number of output layer neurons, w is the number of output layer neurons, and the likeijAs weights of the input layer to the hidden layer, wjkAs weights from hidden layer to output layer, ajThreshold for the hidden layer, bkFor the threshold of the output layer, g (x) is an activation function, g (x) is a Sigmoid function in the form of
HjFor the actual output of the hidden layer,Okis the actual output of the output layer, xiIs the input of the BP neural network.
10. The convergence terminal of claim 1 further comprising an interleaving module for collecting data of a distribution transformer, a clock module for synchronizing clocks, a watchdog module for monitoring operation of the master control core module, a power module for providing power, a communication isolation module for protecting the master control core module, a temperature measurement module for measuring temperature, and a display module for displaying.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111337748.7A CN114217124B (en) | 2021-11-09 | 2021-11-09 | Fusion terminal with loop inspection function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111337748.7A CN114217124B (en) | 2021-11-09 | 2021-11-09 | Fusion terminal with loop inspection function |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114217124A true CN114217124A (en) | 2022-03-22 |
| CN114217124B CN114217124B (en) | 2024-05-03 |
Family
ID=80697018
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111337748.7A Active CN114217124B (en) | 2021-11-09 | 2021-11-09 | Fusion terminal with loop inspection function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114217124B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002100962A (en) * | 2000-09-21 | 2002-04-05 | Texas Instr Japan Ltd | Frequency-characteristic adjusting circuit |
| CN1567710A (en) * | 2003-07-10 | 2005-01-19 | 立积电子股份有限公司 | An oscillator characteristic curve correction system and device |
| CN204287315U (en) * | 2014-11-24 | 2015-04-22 | 山东科技大学 | Based on the power monitoring system of STM32F103 microcontroller |
| CN106645931A (en) * | 2016-11-29 | 2017-05-10 | 国网四川省电力公司电力科学研究院 | Current transformer secondary circuit monitoring module and method, and specific transformer acquiring terminal |
| CN106779069A (en) * | 2016-12-08 | 2017-05-31 | 国家电网公司 | A kind of abnormal electricity consumption detection method based on neutral net |
| WO2017126273A1 (en) * | 2016-01-18 | 2017-07-27 | 東京電力ホールディングス株式会社 | Power-theft detection apparatus and program |
| CN108107248A (en) * | 2017-12-12 | 2018-06-01 | 宁波三星医疗电气股份有限公司 | A kind of stealing recognition methods based on neutral net |
| CN108595905A (en) * | 2017-10-25 | 2018-09-28 | 中国石油化工股份有限公司 | A kind of erosion failure quantitative forecasting technique based on BP neural network model |
| US10414357B1 (en) * | 2019-06-03 | 2019-09-17 | Wild Energy, Inc. | System to selectively provide power to recreational vehicles with a SaaS application accessed via mobile devices |
| US20200228018A1 (en) * | 2019-01-14 | 2020-07-16 | Texas Instruments Incorporated | Methods and apparatus to calibrate a power converter |
| CN111525697A (en) * | 2020-05-09 | 2020-08-11 | 西安交通大学 | Method and system for preventing electricity theft in medium and low voltage distribution network based on current monitoring and line topology analysis |
| CN112649642A (en) * | 2020-12-14 | 2021-04-13 | 广东电网有限责任公司广州供电局 | Electricity stealing position judging method, device, equipment and storage medium |
| CN113255880A (en) * | 2021-04-09 | 2021-08-13 | 中国电力科学研究院有限公司 | Method and system for judging electricity stealing data based on improved neural network model |
| CN113393103A (en) * | 2021-06-03 | 2021-09-14 | 西南科技大学 | Anti-electricity-stealing system based on genetic algorithm optimization BP neural network |
| CN114186611A (en) * | 2021-11-09 | 2022-03-15 | 河南九域腾龙信息工程有限公司 | Method and device for identifying electricity stealing behavior |
-
2021
- 2021-11-09 CN CN202111337748.7A patent/CN114217124B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002100962A (en) * | 2000-09-21 | 2002-04-05 | Texas Instr Japan Ltd | Frequency-characteristic adjusting circuit |
| CN1567710A (en) * | 2003-07-10 | 2005-01-19 | 立积电子股份有限公司 | An oscillator characteristic curve correction system and device |
| CN204287315U (en) * | 2014-11-24 | 2015-04-22 | 山东科技大学 | Based on the power monitoring system of STM32F103 microcontroller |
| WO2017126273A1 (en) * | 2016-01-18 | 2017-07-27 | 東京電力ホールディングス株式会社 | Power-theft detection apparatus and program |
| CN106645931A (en) * | 2016-11-29 | 2017-05-10 | 国网四川省电力公司电力科学研究院 | Current transformer secondary circuit monitoring module and method, and specific transformer acquiring terminal |
| CN106779069A (en) * | 2016-12-08 | 2017-05-31 | 国家电网公司 | A kind of abnormal electricity consumption detection method based on neutral net |
| CN108595905A (en) * | 2017-10-25 | 2018-09-28 | 中国石油化工股份有限公司 | A kind of erosion failure quantitative forecasting technique based on BP neural network model |
| CN108107248A (en) * | 2017-12-12 | 2018-06-01 | 宁波三星医疗电气股份有限公司 | A kind of stealing recognition methods based on neutral net |
| US20200228018A1 (en) * | 2019-01-14 | 2020-07-16 | Texas Instruments Incorporated | Methods and apparatus to calibrate a power converter |
| US10414357B1 (en) * | 2019-06-03 | 2019-09-17 | Wild Energy, Inc. | System to selectively provide power to recreational vehicles with a SaaS application accessed via mobile devices |
| CN111525697A (en) * | 2020-05-09 | 2020-08-11 | 西安交通大学 | Method and system for preventing electricity theft in medium and low voltage distribution network based on current monitoring and line topology analysis |
| CN112649642A (en) * | 2020-12-14 | 2021-04-13 | 广东电网有限责任公司广州供电局 | Electricity stealing position judging method, device, equipment and storage medium |
| CN113255880A (en) * | 2021-04-09 | 2021-08-13 | 中国电力科学研究院有限公司 | Method and system for judging electricity stealing data based on improved neural network model |
| CN113393103A (en) * | 2021-06-03 | 2021-09-14 | 西南科技大学 | Anti-electricity-stealing system based on genetic algorithm optimization BP neural network |
| CN114186611A (en) * | 2021-11-09 | 2022-03-15 | 河南九域腾龙信息工程有限公司 | Method and device for identifying electricity stealing behavior |
Non-Patent Citations (3)
| Title |
|---|
| FEI, JUNTAO: "Double Hidden Layer Output Feedback Neural Adaptive Global Sliding Mode Control of Active Power Filte", IEEE TRANSACTIONS ON POWER ELECTRONICS, vol. 35, no. 3, pages 3069 - 3084, XP011761748, DOI: 10.1109/TPEL.2019.2925154 * |
| 李锋: "基于双隐层量子线路循环单元神经网络的状态退化趋势预测", 机械工程学报, vol. 55, no. 6, pages 83 - 92 * |
| 黄明娟: "基于GPRS的电力远程抄表和数据分析系统的设计实现", 中国优秀硕士学位论文全文数据库工程科技Ⅱ辑, 15 March 2017 (2017-03-15), pages 042 - 2958 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114217124B (en) | 2024-05-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110108914B (en) | An intelligent decision-making method, system, equipment and medium for anti-stealing electricity | |
| CN106338406B (en) | The on-line monitoring of train traction electric drive system and fault early warning system and method | |
| CN111160424B (en) | A method and system for NFC device fingerprint authentication based on CNN image recognition | |
| CN113221086B (en) | Offline face authentication method and device, electronic equipment and storage medium | |
| CN111144522A (en) | Power grid NFC equipment fingerprint authentication method based on hardware intrinsic difference | |
| CN114186611A (en) | Method and device for identifying electricity stealing behavior | |
| CN117614060A (en) | Wireless charging method | |
| CN113538037A (en) | Method, system, equipment and storage medium for monitoring charging event of battery car | |
| CN114184870A (en) | Non-invasive load identification method and device | |
| CN114217124B (en) | Fusion terminal with loop inspection function | |
| CN114862821A (en) | Automatic monitoring method, system and equipment for relay protection hard pressing plate | |
| CN119066547A (en) | A classification and status diagnosis method for DC equipment | |
| Yuan et al. | A data-driven framework for power system event type identification via safe semi-supervised techniques | |
| Pan et al. | Study on intelligent anti–electricity stealing early-warning technology based on convolutional neural networks | |
| CN117746327A (en) | PCS fault detection method and related device for energy storage converter | |
| CN116628620A (en) | Non-invasive load identification calculation method | |
| Lu et al. | Time series power anomaly detection based on Light Gradient Boosting Machine | |
| CN114550466B (en) | A parking space state detection method, device and electronic equipment | |
| An et al. | A ground fault detection method of substation DC system based on particle filter | |
| CN115273350A (en) | Intelligent tail cabinet verification management method and system based on RFID detection | |
| CN113419140B (en) | Distribution network line fault diagnosis method and device considering human-computer-object cooperation | |
| Du et al. | Fault diagnosis method of automation equipment in independent and controllable substation based on deep reinforcement learning | |
| CN119291337B (en) | Remote diagnosis method for electric bicycle rechargeable battery and charger faults | |
| Nadas et al. | Towards continuous subject identification using wearable devices and deep cnns | |
| Zhang et al. | Transformer Fault diagnosis model and method based on DBNI in photoelectric sensors diagnosis system |
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 |