CN105184363A - Zinc oxide arrester valve disc power loss prediction system - Google Patents

Zinc oxide arrester valve disc power loss prediction system Download PDF

Info

Publication number
CN105184363A
CN105184363A CN201510537260.7A CN201510537260A CN105184363A CN 105184363 A CN105184363 A CN 105184363A CN 201510537260 A CN201510537260 A CN 201510537260A CN 105184363 A CN105184363 A CN 105184363A
Authority
CN
China
Prior art keywords
layer
current
prognoses
power attenuation
zinc
Prior art date
Application number
CN201510537260.7A
Other languages
Chinese (zh)
Inventor
陈波
Original Assignee
芜湖市凯鑫避雷器有限责任公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 芜湖市凯鑫避雷器有限责任公司 filed Critical 芜湖市凯鑫避雷器有限责任公司
Priority to CN201510537260.7A priority Critical patent/CN105184363A/en
Publication of CN105184363A publication Critical patent/CN105184363A/en

Links

Abstract

The invention relates to a zinc oxide arrester valve disc power loss prediction system, comprising a space-layer front-position data acquisition module, a network layer bus, and an upper monitoring diagnosis layer. The front-position data acquisition module is responsible for acquiring working voltage, resistive current, and temperature of a monitored circuit, and converts analog quantity to digital quantity, and digital waveform analysis processing is performed on the digital quantity, to obtain required resistive fundamental current characteristic quantity, and meanwhile a result obtained by processing is transmitted to the upper monitoring diagnosis layer. The upper monitoring diagnosis layer performs synthetic judgment on operation conditions, and according to output voltage of the front-position data acquisition module, chargeability of an arrester is obtained. The chargeability and the temperature are used as an input mode of a front-feed neural network input layer, and corresponding power loss is an output mode of an output layer. The system uses the temperature and the chargeability as input parameters of a neural network to predict power loss of a valve disc, so as to establish a rational and accurate power loss analogy model to analyze thermal characteristics of a voltage limiter.

Description

Zinc-Oxide Arrester valve block power attenuation prognoses system
Technical field
The invention belongs to technical field of electric system protection, relate to a kind of Zinc-Oxide Arrester, be specifically related to a kind of Zinc-Oxide Arrester valve block power attenuation prognoses system.
Background technology
Zinc-Oxide Arrester is as the overvoltage protection of electrical equipment, and the quality of its performance plays a part very large to electrical equipment safe operation.Therefore, most important to the research of Zinc-Oxide Arrester operation conditions.The power loss characteristic of zinc oxide valve plate directly decides ultimate temperature rise and the stabilized operating temperature of voltage limiter.The energy that the power attenuation of zinc oxide valve plate and superpotential produce under different voltage, different temperatures effect is the endogenous pyrogen of voltage limiter.The power attenuation of zinc oxide valve plate is the complicated function of temperature and chargeability, adopts general method to be difficult to compare accurate simulation to it.
Summary of the invention
In order to overcome the deficiencies in the prior art, the invention provides a kind of Zinc-Oxide Arrester valve block power attenuation prognoses system, using temperature and chargeability as the input parameter of neural network, predicting the power attenuation of valve block.
Technical scheme of the present invention is: a kind of Zinc-Oxide Arrester valve block power attenuation prognoses system, comprise wall advance data acquisition module, network layer bus and top monitoring and diagnosis layer, advance data acquisition module is responsible for the working voltage gathering monitored circuit, current in resistance property and temperature, and convert analog quantity to digital quantity, digital waveform analyzing and processing is carried out to it and obtains desired resistive fundamental current characteristic quantity, pass to top monitoring and diagnosis layer processing the result obtained simultaneously, top monitoring and diagnosis layer carries out comprehensive descision to ruuning situation, output voltage according to described advance data acquisition module obtains lightning arrester chargeability, by chargeability, temperature is as the input pattern of feedforward neural network input layer, corresponding power attenuation is as the output mode of output layer, be net in the input value of jth node layer j=∑ W jio i, wherein, W jifor j, i two weight of interlayer, and the output valve of node j is O j=f (net j), f is the excitation function of node j: θ jfor the threshold value of node j.The increment of described connection weight is: Δ w ji=η o io j(1-o j) ∑ kδ kk kj.Described prognoses system also comprises temperature control module and baking oven, and described temperature control module comprises thermocouple, and thermocouple is placed in an oven, and described arrester valve piece is placed in baking oven.Described advance data acquisition module comprises simulating signal extraction unit and signal processing unit, and simulating signal extraction unit sends the signal collected to signal processing unit reprocessing.Described simulating signal extraction unit comprises sensor sample circuit, signal amplification circuit, Anti-aliasing Filter Circuits and electrical level polar translation circuit, the signal that sensor sample circuit obtains sensor amplifies, anti-interference, anti-aliasing filter and electrical level polar conversion, makes it to match with converter.Described Anti-aliasing Filter Circuits adopts second order Butterworth LPF.Described sensor sample circuit comprises current in resistance property sample circuit and lightning arrester working voltage sample circuit, and lightning arrester working voltage sample circuit and current in resistance property sample circuit be lead-in wire acquisition working voltage and current in resistance property under the secondary side and earth terminal of the voltage transformer (VT) be connected to bus respectively.Described current in resistance property sample circuit is counter sample circuit, resistance sampling circuit, or current transformer sample circuit.Described signal processing unit comprises A/D sampling unit, DSP processing unit and CAN interface unit, carry out frequency analysis obtain fundamental voltage and the monitoring feature such as resistive fundamental current, harmonic current amount by digitized voltage, current in resistance property signal after analog quantity is converted to digital quantity, be articulated in bus finally by controller, the monitoring feature amount that acquisition process obtains is uploaded to top monitoring and diagnosis layer.
The present invention has following good effect: prognoses system of the present invention, using temperature and chargeability as the input parameter of neural network, predicts the power attenuation of valve block, sets up rationally, thermal characteristics that accurate power attenuation analogy model analyzes voltage limiter.
Accompanying drawing explanation
Fig. 1 is the semilinear feed-forward network model of specific embodiment of the invention valve block power attenuation.
Embodiment
Contrast accompanying drawing below, by the description to embodiment, the specific embodiment of the present invention is as the effect of the mutual alignment between the shape of involved each component, structure, each several part and annexation, each several part and principle of work, manufacturing process and operation using method etc., be described in further detail, have more complete, accurate and deep understanding to help those skilled in the art to inventive concept of the present invention, technical scheme.
Arrester valve piece power attenuation prognoses system of the present invention comprises the Fieldbus Based layered distribution type structure of employing, and system forms by three layers, comprises wall advance data acquisition module, network layer bus and top monitoring and diagnosis layer.Realize interconnecting by means of only two shielded twisted pairs between advance data acquisition module and upper layer host.Data communication interface and the power input interface of advance data acquisition module all adopt Phototube Coupling, be equivalent to make connection be in suspended state, can ensure that the electrical grounding of each advance data acquisition module is completely independent, there is no electrical link each other, well solve anti-interference, shock proof problem.The operation characteristic amount of top monitoring and diagnosis layer to the monitored equipment of advance data acquisition module acquisition is from below analyzed and reprocessing, is predicted the power attenuation of arrester valve piece by the chargeability under different temperatures.
Advance data acquisition module primary responsibility gathers the working voltage of monitored lightning arrester, current in resistance property and temperature, and convert analog quantity to digital quantity, digital waveform analyzing and processing is carried out to it and obtains desired resistive fundamental current characteristic quantity, simultaneously process the result that obtains correct pass to top monitoring and diagnosis layer, top monitoring and diagnosis layer carries out comprehensive descision to ruuning situation.Advance data acquisition module is the sampling of settling signal and preposition simulation process link mainly, the precision of conditioning link and degree of stability are directly connected to the precision and stability of whole system, if the design of advance data acquisition module is improper, follow-up digital processing link adopts which type of device and algorithm to be all difficult to reach desirable effect, so the quality of advance data acquisition module design is a very important link.
Advance data acquisition module of the present invention comprises simulating signal extraction unit and signal processing unit, simulating signal extraction unit is the working voltage and current in resistance property that lead-in wire obtains under the secondary side and earth terminal of the voltage transformer (VT) be connected to bus respectively, employing high-accuracy voltage, current sensor convert voltage, current signal to low pressure small area analysis signal, then the signal obtained sensor amplifies, anti-interference, anti-aliasing filter and electrical level polar conversion, makes it to match with converter.Signal processing unit comprises A/D sampling unit, DSP processing unit and CAN interface unit, sampling holder and traffic pilot, carry out frequency analysis obtain fundamental voltage and the monitoring feature such as resistive fundamental current, harmonic current amount by digitized voltage, current in resistance property signal after analog quantity is converted to digital quantity, be articulated in bus finally by controller, the monitoring feature amount that acquisition process obtains is uploaded to top monitoring and diagnosis layer.Advance data acquisition module samples primarily of sensor, and isolation is amplified, and the circuit such as low-pass filtering and reversal forms.
Sensor sample circuit comprises current in resistance property sample circuit, lightning arrester working voltage sample circuit, the sampling current of current in resistance property sample circuit is the feeble signal of tens to a hundreds of microampere, such Weak absorption gets up very difficult, be easy to the interference receiving external electromagnetic signal, therefore, current in resistance property sample circuit of the present invention can be counter sample circuit, resistance sampling circuit, or current transformer sample circuit.Because the present invention is mainly analyzed the fundametal compoment in voltage, current signal and fractional harmonic component, and high frequency interference and most of higher hamonic wave are limited, therefore we need to design low-pass filter circuit, concrete, adopt second order Butterworth LPF.
Concrete, monitoring system also comprises temperature control module, and temperature control module connects the baking oven placing arrester valve piece, and temperature-controlling system comprises thermocouple, and thermocouple is placed in an oven, controls the temperature in baking oven.Whole supervisory system adopts independent voltage source, and lightning arrester is the lightning arrester be separated with high-tension line, therefore, needs to provide power supply.
Power attenuation is the important parameter judging Zinc-Oxide Arrester performance, artificial Neural Network Simulation model is set up to power loss characteristic, find out the exact relationship of power attenuation and running temperature and chargeability, above-mentioned power attenuation test macro is utilized to measure working voltage and the current in resistance property of lightning arrester, drawn the loss power of lightning arrester by working voltage and current in resistance property, then come weight and the threshold value of mediator's artificial neural networks model by loss power.
Feed-forward network model is by input layer, hidden layer and output layer composition, feedforward network is adopted to carry out Simulation of Complex function, then the input pattern of input layer is independent variable, the output mode of output layer is dependent variable, the neural network of hidden layer is intermediate variable, the difference of the desired output that the system output valve that feed-forward network model obtains through network calculations according to the input pattern of sample is corresponding with input pattern, each weight is regulated by learning process, the difference of the two is made to reach the requirement of precision, to the power attenuation P foundation artificial neural network as shown in Figure 1 of zinc oxide valve plate, by chargeability Q, temperature T is as the input pattern of input layer, corresponding power attenuation P is as the output mode of output layer, be net in the input value of jth node layer j=∑ W jio i, wherein, W jifor j, i two weight of interlayer, and the output valve of node j is O j=f (net j), f is the excitation function of node j: θ jfor the threshold value of node j.System output valve { the o of sample M kand desired output { t ksquare error be:
The increment of each weight in learning process can be obtained by square error E: in formula, η is pace of learning, and E is for exporting o kfunction, o kfor:
o k=f(net k)
net k=∑w kjo j
∂ E ∂ w k j = ∂ E ∂ net k ∂ net k ∂ w k j = ∂ E ∂ net k ∂ ∂ w k j ( w k j o j ) = o j ∂ E ∂ net k
Order then there is Δ w ji=η δ ko j.
δ k = - ∂ E ∂ o k ∂ o k ∂ net k = - ( t k - o k ) ∂ o k ∂ net k = ( t k - o k ) o k ( 1 - o k )
Therefore the arbitrary node for output layer has:
Δ w kj=η o j(t k-o k) o k(1-o k) set up.
Δw j i = ∂ E ∂ w j i = - η ∂ E ∂ net j ∂ net j ∂ w j i = - η ∂ E ∂ net j o i
Order then there is Δ w ji=η δ jo i
δ j = ∂ E ∂ o j ∂ o j ∂ net j = ∂ E ∂ o j [ - Σ ∂ E ∂ net k ∂ net k ∂ o j ] = o j ( 1 - o j ) Σ k δ k k k j
Therefore the increment of the connection weight of non-output layer can be obtained:
Δw j i = ηo i o j ( 1 - o j ) Σ k δ k k k j
Finally, test different temperatures, the ac power loss trial value of zinc oxide valve plate under different chargeability, the process above that substitutes into learns, and draws weight and the threshold value of the artificial nerve network model of valve block.
Above by reference to the accompanying drawings to invention has been exemplary description; obvious specific implementation of the present invention is not subject to the restrictions described above; as long as have employed the improvement of the various unsubstantialities that method of the present invention is conceived and technical scheme is carried out; or design of the present invention and technical scheme directly applied to other occasion, all within protection scope of the present invention without to improve.

Claims (9)

1. a Zinc-Oxide Arrester valve block power attenuation prognoses system, it is characterized in that, comprise wall advance data acquisition module, network layer bus and top monitoring and diagnosis layer, advance data acquisition module is responsible for the working voltage gathering monitored circuit, current in resistance property and temperature, and convert analog quantity to digital quantity, digital waveform analyzing and processing is carried out to it and obtains desired resistive fundamental current characteristic quantity, pass to top monitoring and diagnosis layer processing the result obtained simultaneously, top monitoring and diagnosis layer carries out comprehensive descision to ruuning situation, output voltage according to described advance data acquisition module obtains lightning arrester chargeability, by chargeability, temperature is as the input pattern of feedforward neural network input layer, corresponding power attenuation is as the output mode of output layer, be net in the input value of jth node layer j=∑ W jio i, wherein, W jifor j, i two weight of interlayer, and the output valve of node j is O j=f (net j), f is the excitation function of node j: θ jfor the threshold value of node j.
2. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 1, it is characterized in that, the increment of described connection weight is: Δ w ji=η o io j(1-o j) ∑ kδ kk kj.
3. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 2, it is characterized in that, described prognoses system also comprises temperature control module and baking oven, and described temperature control module comprises thermocouple, thermocouple is placed in an oven, and described arrester valve piece is placed in baking oven.
4. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 1, it is characterized in that, described advance data acquisition module comprises simulating signal extraction unit and signal processing unit, and simulating signal extraction unit sends the signal collected to signal processing unit reprocessing.
5. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 4, it is characterized in that, described simulating signal extraction unit comprises sensor sample circuit, signal amplification circuit, Anti-aliasing Filter Circuits and electrical level polar translation circuit, the signal that sensor sample circuit obtains sensor amplifies, anti-interference, anti-aliasing filter and electrical level polar conversion, makes it to match with converter.
6. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 5, is characterized in that, described Anti-aliasing Filter Circuits adopts second order Butterworth LPF.
7. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 5, it is characterized in that, described sensor sample circuit comprises current in resistance property sample circuit and lightning arrester working voltage sample circuit, and lightning arrester working voltage sample circuit and current in resistance property sample circuit be lead-in wire acquisition working voltage and current in resistance property under the secondary side and earth terminal of the voltage transformer (VT) be connected to bus respectively.
8. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 7, is characterized in that, described current in resistance property sample circuit is counter sample circuit, resistance sampling circuit, or current transformer sample circuit.
9. Zinc-Oxide Arrester valve block power attenuation prognoses system according to claim 4, it is characterized in that, described signal processing unit comprises A/D sampling unit, DSP processing unit and CAN interface unit, carry out frequency analysis obtain fundamental voltage and the monitoring feature such as resistive fundamental current, harmonic current amount by digitized voltage, current in resistance property signal after analog quantity is converted to digital quantity, be articulated in bus finally by controller, the monitoring feature amount that acquisition process obtains is uploaded to top monitoring and diagnosis layer.
CN201510537260.7A 2015-08-26 2015-08-26 Zinc oxide arrester valve disc power loss prediction system CN105184363A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510537260.7A CN105184363A (en) 2015-08-26 2015-08-26 Zinc oxide arrester valve disc power loss prediction system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510537260.7A CN105184363A (en) 2015-08-26 2015-08-26 Zinc oxide arrester valve disc power loss prediction system

Publications (1)

Publication Number Publication Date
CN105184363A true CN105184363A (en) 2015-12-23

Family

ID=54906426

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510537260.7A CN105184363A (en) 2015-08-26 2015-08-26 Zinc oxide arrester valve disc power loss prediction system

Country Status (1)

Country Link
CN (1) CN105184363A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2475104Y (en) * 2001-04-18 2002-01-30 武汉大学 Zinc oxide lightning arrester
CN1877355A (en) * 2005-10-14 2006-12-13 安徽省电力科学研究院 Insulated on-line monitoring system checker of high-voltage electric equipment
EP2352269A1 (en) * 2008-12-30 2011-08-03 State Grid Corporation of China Service access method and system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2475104Y (en) * 2001-04-18 2002-01-30 武汉大学 Zinc oxide lightning arrester
CN1877355A (en) * 2005-10-14 2006-12-13 安徽省电力科学研究院 Insulated on-line monitoring system checker of high-voltage electric equipment
EP2352269A1 (en) * 2008-12-30 2011-08-03 State Grid Corporation of China Service access method and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
曹万磊: "分布式氧化锌避雷器在线监测系统研究", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技II辑》 *
李庆玲: "氧化锌避雷器运行状况研究", 《中国优秀硕士学位论文全文数据库工程科技II辑》 *

Similar Documents

Publication Publication Date Title
CN102819004B (en) Comprehensive detecting and analyzing platform for performance of digital electric energy metering systems of intelligent transformer substations
CN101975910B (en) Intelligent fault classification and location method for ultra-high voltage direct current transmission line
CN102841296B (en) Online monitoring system and method for partial discharge of intelligent switch cabinet based on ultra-high frequency detection
CN101408579B (en) Oil paper insulated electrothermic aging test and local discharge integrated detection device of transformer
CN105846780B (en) A kind of photovoltaic module method for diagnosing faults based on decision-tree model
CN103105550B (en) A kind of intelligent relay protection device detection method and system
EP3045930B1 (en) Testing system of gis electronic mutual inductor and method therefor
CN104267354B (en) A kind of peak power Forecasting Methodology of electrokinetic cell
Kim et al. Fast and reliable estimation of composite load model parameters using analytical similarity of parameter sensitivity
CN102288857B (en) Fault arc identification and detection method and detection protection device
CN103268519B (en) Based on the power-system short-term load forecasting method and the device that improve Lyapunov index
WO2015117304A1 (en) System for online monitoring of zinc oxide arrester and method thereof
CN107870306A (en) A kind of lithium battery charge state prediction algorithm based under deep neural network
CN201966909U (en) Integrated test system for low-voltage narrowband power line carrier communication
CN101408580B (en) Method for evaluating oil paper insulation ageing state based on local discharge characteristic parameter
CN106777984B (en) A method of photovoltaic array Working state analysis and fault diagnosis are realized based on density clustering algorithm
CN104796082A (en) System and method for diagnosing faults of photovoltaic power generation systems in online manner
CN104537415B (en) A kind of non-linear process industry failure predication and recognition methodss based on compressed sensing and DROS ELM
CN103901882B (en) A kind of system and method for train dynamics system on-line monitoring fault diagnosis
CN104090985A (en) Active disconnection optimum fracture surface searching method based on electrical distance
CN103698033B (en) A kind of coiling hot point of transformer temperature prediction assessment system for possessing adaptation function
US20190195727A1 (en) Gil fault on-line monitoring system based on vibration signals and support vector machine
CN105512448B (en) A kind of appraisal procedure of power distribution network health index
Sanchez-Pacheco et al. PLC-based PV plants smart monitoring system: field measurements and uncertainty estimation
CN102608452A (en) High speed train equipment state and electric energy quality monitoring system and method

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20151223

RJ01 Rejection of invention patent application after publication