CN111030302A - Intelligent electricity safety detector - Google Patents
Intelligent electricity safety detector Download PDFInfo
- Publication number
- CN111030302A CN111030302A CN201911321481.5A CN201911321481A CN111030302A CN 111030302 A CN111030302 A CN 111030302A CN 201911321481 A CN201911321481 A CN 201911321481A CN 111030302 A CN111030302 A CN 111030302A
- Authority
- CN
- China
- Prior art keywords
- module
- central processing
- data
- current
- data acquisition
- 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.)
- Pending
Links
- 230000005611 electricity Effects 0.000 title claims description 18
- 238000012545 processing Methods 0.000 claims abstract description 52
- 238000004891 communication Methods 0.000 claims abstract description 46
- 238000003860 storage Methods 0.000 claims abstract description 21
- 230000003993 interaction Effects 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims description 28
- 238000012706 support-vector machine Methods 0.000 claims description 19
- 239000013598 vector Substances 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000011161 development Methods 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000013480 data collection Methods 0.000 claims description 4
- 238000010891 electric arc Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 7
- 238000001514 detection method Methods 0.000 description 7
- 238000005070 sampling Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000001052 transient effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
Abstract
Intelligence power consumption detector relates to intelligence power consumption and surveys the field, and this detector includes: the device comprises a power supply module, a central processing module, a data acquisition module, a display module, a communication module and a storage module; the power module, the data acquisition module, the display module, the communication module and the storage module are respectively connected with the central processing module, and the central processing module, the data acquisition module, the display module, the communication module and the storage module are respectively powered by the power module; the human-computer interaction module in the display module sends an instruction to the central processing module, the central processing module controls the data acquisition module to acquire numerical values of current, temperature and voltage and then transmits the numerical values to the central processing module, the central processing module processes the numerical values and then respectively transmits the numerical values to the display module, the communication module and the storage module, the human-computer interaction module displays and controls the data type and size acquired by the data acquisition module, the communication module outputs the data to an external terminal, and the storage module stores the data.
Description
Technical Field
The invention relates to the field of intelligent electricity utilization detection, in particular to an intelligent electricity utilization detector.
Background
In recent years, electrical fires frequently occur in China, and serious casualties and property loss are caused frequently. According to statistics, 52.4 thousands of electric fires occur in China from 2011 to 2016, so that 3261 people die and 2063 people are injured, and the direct economic loss is more than 92 billion yuan, which accounts for more than 30% of the total fire and casualty loss in China; wherein the fire 17 from the very big electric fire accounts for 70% of the total number of the very big electric fires. The accidents expose the outstanding problems in the aspects of production quality, circulation and sale of electric products, electrical design and construction of construction projects, use of the electric products and lines thereof, maintenance and management and the like.
According to the statistics of the technical appraisal center of the fire-fighting bureau electric fire cause of the ministry of public security, the electric fire is mostly caused by the direct or indirect cause of the electric circuit. One end of public safety is connected with thousands of households, and the other end of public safety is connected with economic and social development, and is a stable wind vane for the society.
In the actual condition, many production and operation units electric line are old, the circuit hidden danger is many and the disguise is strong, and most of small and micro enterprises lack professional electrician, and the naked eye can't discover electric hidden danger directly perceivedly, and a series of difficult problems such as various hidden dangers are difficult to in time to investigate to traditional detection means for the monitoring of electric fire and early warning are difficult to put in place.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an intelligent electricity utilization detector, which uses wavelet analysis and load current as characteristic vectors and is based on a training recognition algorithm of a support vector machine, and the electricity utilization safety is solved in advance.
The technical scheme adopted by the invention for solving the technical problem is as follows:
an intelligent electricity usage detector, comprising: the device comprises a power supply module, a central processing module, a data acquisition module, a display module, a communication module and a storage module; the power module, the data acquisition module, the display module, the communication module and the storage module are respectively connected with the central processing module, and the central processing module, the data acquisition module, the display module, the communication module and the storage module are respectively powered by the power module; the human-computer interaction module in the display module sends an instruction to the central processing module, the central processing module controls the data acquisition module to acquire numerical values of current, temperature and voltage and then transmits the numerical values to the central processing module, the central processing module processes the numerical values and then respectively transmits the numerical values to the display module, the communication module and the storage module, the human-computer interaction module displays and controls the data type and size acquired by the data acquisition module, the communication module outputs the data to an external terminal, and the storage module stores the data.
The central processing module obtains current change data under different load conditions acquired by the data acquisition module, and performs wavelet analysis on the current value to obtain a fourth scale detail component; in the central processing module, constructing a feature vector of a fourth scale detail component and a current change value and substituting the feature vector into a support vector machine module, training a sample and developing an arc fault recognizer, substituting actual features into the trained support vector machine module in actual application, and judging whether the obtained actual numerical value is an arc fault numerical value or not; the communication module updates the training sample at any time and is used in different environments.
Preferably, the central processing module comprises a dual-core processor with a DSP core and an ARM core.
Preferably, the support vector machine module collects different current values through the data acquisition module in the training stage under the condition of different loads, calculates wavelet components and load currents, takes the wavelet components and the load currents as characteristic vectors as samples of training data, obtains a training model through learning, and stores the training model; in the engineering stage, different current values are acquired through a data acquisition module according to the actual load current condition, the wavelet component and the load current are calculated, and the data are substituted into a training model for judgment to obtain whether the electric arc is a fault numerical value.
Preferably, the data acquisition module comprises a current acquisition module, a temperature acquisition module and a voltage acquisition module; the current acquisition module, the temperature acquisition module and the voltage acquisition module are respectively connected with the central processing module.
Preferably, the human-computer interaction module comprises the HMI development module and a field HMI module; the HMI development module is embedded into the field HMI component, and an operator controls data collected by the data collection module to be transmitted to the central processing module through an instruction sent by the field HMI component and then displayed by the display module after processing.
Preferably, the communication module includes: a Web network communication interface, an Ethernet communication interface, an RS485 communication interface, an RS232 communication interface, a USB communication interface and a Bluetooth communication interface.
Preferably, the feature vector of the support vector machine module is overcurrent, voltage, temperature, short circuit, ground fault or fault arc.
Preferably, the device further comprises an AD conversion module; the signal that the data acquisition module gathered passes through AD conversion module, converts analog signal into digital signal, transmits to central processing module.
The invention has the beneficial effects that: the invention monitors real-time parameters and operation conditions of the electric terminal, processes and analyzes data sampling, starts an alarm function if the data exceeds an early warning limit value, transmits real-time data, an alarm state, a fault state and a position to a user end, and directly transmits serious conditions to a fire department. The invention adopts a machine learning method of a support vector machine, a high-speed sampling and conversion technology and a rapid fault arc detection algorithm to realize the learning training of different load conditions of the electricity safety, and the rapid detection is carried out according to a model obtained by learning in the practical application. When harmful voltage disturbance occurs each time or impact current exceeding a set value occurs, the detector can accurately capture and judge. The pre-alarm is given out when necessary, so that the electricity safety is not processed after the fact, but the problem is solved in advance.
Drawings
FIG. 1 is a schematic diagram of an intelligent electricity consumption detector according to the present invention.
FIG. 2 is a flow chart of the support vector machine module of the present invention during a training phase.
FIG. 3 is a flow chart of the support vector machine module of the present invention during the engineering phase.
Fig. 4AD9253 functional block diagram.
Fig. 5AD9253 data output timing.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
As shown in fig. 1, the intelligent electricity consumption detector includes: the device comprises a power supply module, a central processing module, a data acquisition module, a display module, a communication module and a storage module; the power module, the data acquisition module, the display module, the communication module and the storage module are respectively connected with the central processing module, and the central processing module, the data acquisition module, the display module, the communication module and the storage module are respectively powered by the power module, so that the normal working state is ensured. The field worker sends an instruction to the central processing module through the man-machine interaction module in the display module, the central processing module controls the data acquisition module to acquire numerical values of current, temperature and voltage and then transmits the numerical values to the central processing module, the numerical values are processed by the central processing module and then respectively transmitted to the display module, the communication module and the storage module, the man-machine interaction module displays the data type and size acquired by the data acquisition module, the communication module outputs the data to an external terminal, and the storage module stores the data. The central processing module comprises a DSP core and an ARM core dual-core processor.
In this embodiment, the ARM core is an ARMCortex-A8 series microprocessor, American TI. The data acquisition module comprises a current acquisition module, a temperature acquisition module and a voltage acquisition module; the current acquisition module, the temperature acquisition module and the voltage acquisition module are respectively connected with the central processing module. The human-computer interaction module comprises the HMI development module and a field HMI module; the HMI development module is embedded into the field HMI component, and an operator controls data collected by the data collection module to be transmitted to the central processing module through an instruction sent by the field HMI component and then displayed by the display module after processing. The communication module includes: a Web network communication interface, an Ethernet communication interface, an RS485 communication interface, an RS232 communication interface, a USB communication interface and a Bluetooth communication interface.
The line currents of various typical loads exhibit various characteristics. When the load is operating in a steady state, as is the case in "normal" conditions, the current in the circuit remains relatively stable and exhibits good symmetry characteristics and some periodicity. When a load can be characterized as a transient process, such as starting the load, dimming the light, adjusting the speed, and plugging in or out, the current in the circuit initially changes monotonically and then returns to a steady state. However, if there is a fault arc in the circuit, the periodicity of the current will be lost and the amplitude will become unstable, sometimes possibly zero. This is because the inductive and capacitive reactances of fault arcs are generally unstable and they may change occasionally due to external conditions. The overall current characteristic can be used to overcome the effects of crosstalk, and the current variation should be quantitatively described to accurately extract arc fault features. Current feature extraction current integration can be used to observe changes in current.
Fault arcing is obviously a random process. The line current of a fault arc is often severely distorted or varied in an indeterminate manner. Changes in the current cycle integral can also be observed when the load is operating under normal conditions, especially during transient processes. To avoid false determinations during transients, several current cycle current changes are introduced. Depending on the geometric meaning of the definite integral, the definite integral can be used to represent the total area of the graph. The region contains positive and negative regions, such as the current wave region of the current time period. The current time is 20 milliseconds. The load current wave of an arc fault is typically asymmetric, while the load current wave of a normal state is symmetric. Thus, a deterministic integral over an AC current period can be used to represent a current change. The current change can be calculated as the sum of the integrals over a certain time for several present periods, expressed as:
wherein T isAC10 is half of the AC current period, i (t) represents the line current, and K8 represents an integration time of 80 ms. The current change threshold between the fault arc and the normal state is unequal at different loads. The high-frequency energy of the fault arc is obviously increased, and the current period integral changes randomly. However, high frequency signals are susceptible to crosstalk and using only current variations may lead to false identifications under different loads.
The central processing module obtains current change data under different load conditions acquired by the data acquisition module, and performs wavelet analysis on a current value to obtain a fourth scale detail component; a support vector machine module is arranged in the central processing module, and the fourth scale detail component y is used for1And a current variation value y2Constructing a characteristic vector and substituting the characteristic vector into a support vector machine module to form a training sample and develop an arc fault recognizer, and substituting the actual numerical value characteristics into the trained support vector machine module to judge whether the actual numerical value is an arc fault numerical value or not in actual application; the communication module updates the training samples in real time so as to be used under different environments.
The non-linear function f (y) is used to map the feature vector y from the observation space to a higher dimensional feature space. An optimal recognition between a faulty arc and a normal state can be obtained in a high-dimensional feature space. The constrained optimization problem can be described as:
where U is the weight vector, N is the number of samples, g is the penalty factor, bn(N ═ 1.. times, N) is a weighting factor, which is a deviation term, ξnAnd (N ═ 1.. times, N) are error variables and output data. Designed in view of all the derivations and discussions aboveA flow chart of the arc fault detection algorithm is shown in fig. 2 and 3. min is the minimum of the equation and s.t. is subject to. . Constraining
In the training stage, the support vector machine module collects different current values through the data acquisition module under the condition of different loads, calculates wavelet components and load currents, and takes the wavelet components and the load currents of typical loads as characteristic vectors as sample sets (y, z) of training data, wherein '0' represents a normal state, and '1' represents an arc fault state. An arc fault recognizer is developed through learning training samples to obtain a training model, and then the training model is stored, and g is determined to be 3.26 through a ten-fold cross validation method; in the engineering stage, different current values are acquired through a data acquisition module according to the actual load current condition, the wavelet component and the load current are calculated, and the data are substituted into a training model for judgment to obtain whether the electric arc is a fault numerical value.
In this embodiment, the eigenvector of the support vector machine module is overcurrent, voltage, temperature, short circuit, ground fault or fault arc.
In the application and implementation of the scheme, the AD9253TCPZ-125EP chip of ADI company is used as the analog-to-digital converter of the AD conversion module, and the chip works at-55-125 ℃ and can meet the working environment temperature requirement of the application. The AD9253TCPZ-125EP is internally integrated with 4 paths of 14-bit differential AD converters which can work simultaneously so as to ensure the synchronization of the AD conversion sampling time, the maximum sampling rate of 125MHz and the minimum sampling rate of 10MHz, support the synchronization of multiple chips and the maximum analog bandwidth of 650 MHz.
The functional block diagram of the AD9253TCPZ-125EP is shown in figure 4. The AD9253 outputs data using LVDS differential signals, and the timing of the data output signals is shown in FIG. 5. The AD9253 works under the drive of a synchronous 10MHz clock signal from a load controller, and the synchronous control signal ensures the sampling time synchronization among multiple AD. According to different load conditions in application, a corresponding training model is learned through a support vector machine and is led into a central processing module and a storage module, wavelet characteristics and load current characteristics obtained through real-time sampling calculation are brought into the support vector machine to be distinguished in real time, and rapid detection of fault arcs in power utilization safety is achieved. Meanwhile, the training model can be updated in a background in real time and parameters can be updated by combining the support of the communication module, so that better detection precision can be obtained.
When the central processing module detects the fault arc, the information can be transmitted to the computer client or the mobile client through the communication module, so that an expert outside a field can know the fault condition in time and make corresponding judgment and timely processing.
Claims (8)
1. Intelligent electricity detector, its characterized in that, this detector includes: the device comprises a power supply module, a central processing module, a data acquisition module, a display module, a communication module and a storage module; the power module, the data acquisition module, the display module, the communication module and the storage module are respectively connected with the central processing module, and the central processing module, the data acquisition module, the display module, the communication module and the storage module are respectively powered by the power module; sending an instruction to a central processing module through a human-computer interaction module in a display module, controlling a data acquisition module to acquire numerical values of current, temperature and voltage by the central processing module, transmitting the numerical values to the central processing module, processing the numerical values by the central processing module, transmitting the numerical values to the display module, a communication module and a storage module respectively, displaying the type and the size of data acquired by the data acquisition module by the human-computer interaction module, outputting the data to an external terminal by the communication module, and storing the data by the storage module; the central processing module obtains current change data under different load conditions acquired by the data acquisition module, and performs wavelet analysis on the current value to obtain a fourth scale detail component; in the central processing module, constructing a feature vector of a fourth scale detail component and a current change value and substituting the feature vector into a support vector machine module for training to obtain a training model and develop an arc fault recognizer, and substituting an actual feature value into the trained support vector machine module in actual application to judge whether the obtained actual value is an arc fault value; the communication module updates the training sample at any time and is used in different environments.
2. The intelligent power consumption detector of claim 1, wherein the central processing module comprises a DSP core and an ARM core dual-core processor.
3. The intelligent electricity utilization detector according to claim 1, wherein the support vector machine module collects different current values through the data collection module under different loads in a training stage, calculates wavelet components and load currents, takes the wavelet components and the load currents as characteristic vectors as samples of training data, obtains a training model through learning, and stores the training model; in the engineering stage, different current values are acquired through a data acquisition module according to the actual load current condition, the wavelet component and the load current are calculated, and the data are substituted into a training model for judgment to obtain whether the electric arc is a fault numerical value.
4. The intelligent electricity consumption detector of claim 1, wherein the data acquisition module comprises a current acquisition module, a temperature acquisition module and a voltage acquisition module; the current acquisition module, the temperature acquisition module and the voltage acquisition module are respectively connected with the central processing module.
5. The intelligent electricity usage detector of claim 1, wherein the human-machine interaction module includes the HMI development module and a field HMI module; the HMI development module is embedded into the field HMI component, and an operator controls data collected by the data collection module to be transmitted to the central processing module through an instruction sent by the field HMI component and then displayed by the display module after processing.
6. The intelligent electricity usage detector of claim 1, wherein the communication module comprises: a Web network communication interface, an Ethernet communication interface, an RS485 communication interface, an RS232 communication interface, a USB communication interface and a Bluetooth communication interface.
7. The intelligent electricity usage detector of claim 1, wherein the support vector machine module's eigenvector is overcurrent, voltage, temperature, short circuit, ground fault, or fault arc.
8. The intelligent electricity consumption detector of claim 1, further comprising an AD conversion module; the signal that the data acquisition module gathered passes through AD conversion module, converts analog signal into digital signal, transmits to central processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911321481.5A CN111030302A (en) | 2019-12-20 | 2019-12-20 | Intelligent electricity safety detector |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911321481.5A CN111030302A (en) | 2019-12-20 | 2019-12-20 | Intelligent electricity safety detector |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111030302A true CN111030302A (en) | 2020-04-17 |
Family
ID=70212208
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911321481.5A Pending CN111030302A (en) | 2019-12-20 | 2019-12-20 | Intelligent electricity safety detector |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111030302A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202257832U (en) * | 2011-09-23 | 2012-05-30 | 宁波习羽电子发展有限公司 | Detector for electric fire protection |
| CN103278734A (en) * | 2013-06-26 | 2013-09-04 | 浙江爱德电子有限公司 | Arc fault detection device and detection method thereof |
| CN103353570A (en) * | 2013-06-09 | 2013-10-16 | 福州大学 | Method and system for identifying arc fault based on load terminal voltage detection |
| CN109406949A (en) * | 2018-12-14 | 2019-03-01 | 国网山东省电力公司电力科学研究院 | Power distribution network incipient fault detection method and device based on support vector machines |
-
2019
- 2019-12-20 CN CN201911321481.5A patent/CN111030302A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202257832U (en) * | 2011-09-23 | 2012-05-30 | 宁波习羽电子发展有限公司 | Detector for electric fire protection |
| CN103353570A (en) * | 2013-06-09 | 2013-10-16 | 福州大学 | Method and system for identifying arc fault based on load terminal voltage detection |
| CN103278734A (en) * | 2013-06-26 | 2013-09-04 | 浙江爱德电子有限公司 | Arc fault detection device and detection method thereof |
| CN109406949A (en) * | 2018-12-14 | 2019-03-01 | 国网山东省电力公司电力科学研究院 | Power distribution network incipient fault detection method and device based on support vector machines |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN201876970U (en) | Intelligent measuring and controlling device for electrical fires | |
| CN107991525B (en) | Green industry park electricity supply and utilization efficiency monitoring and fault responsibility identification method and system | |
| CN102255330A (en) | Device and method for detecting island of micro power grid system | |
| CN103259333B (en) | Intelligent alarm and comprehensive analysis of fault system | |
| CN103440726A (en) | Electrical fire warning information processing method and system | |
| CN103944265A (en) | Interactive and smart electric energy distribution and consumption monitoring device and method | |
| CN105391168A (en) | Transformer load real-time control method | |
| CN204497848U (en) | A kind of based on man-machine interface LV Distribution Monitoring System | |
| CN213659613U (en) | Electrical fire early warning device based on deep neural network | |
| CN104391184B (en) | Transformer phase loss detection and warning device and method | |
| CN103578336A (en) | Intelligent self-learning system for electromechanical equipment | |
| CN111030302A (en) | Intelligent electricity safety detector | |
| CN110954784A (en) | Intelligent electricity safety detector | |
| CN101846715A (en) | Low-current ground fault line selection device and multi-platform combined line selection method | |
| CN205453302U (en) | Electric power monitoring device | |
| CN115980519A (en) | Early warning method and system for potential safety hazard of electric bicycle during charging | |
| CN108429243A (en) | Disconnection device and hydroelectric system | |
| CN107247199A (en) | A kind of remote power quality detecting system based on ARM | |
| CN204241561U (en) | A kind of online energy monitor end device | |
| CN210039082U (en) | Intelligent electricity utilization safety detection equipment based on EnOcean communication technology | |
| CN103795141A (en) | DC reclosing switch | |
| CN106469902A (en) | 10KV phase-loss low-voltage intelligent tripping protection device | |
| CN112731059A (en) | Low-voltage line intelligent monitoring device and monitoring system thereof | |
| CN202405862U (en) | Automatic power transfer switch controller | |
| CN112394222A (en) | Load comparison method and system for line loss monitoring |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200417 |
|
| RJ01 | Rejection of invention patent application after publication |