CN113515885B - Intelligent health state diagnosis method for photovoltaic module - Google Patents
Intelligent health state diagnosis method for photovoltaic module Download PDFInfo
- Publication number
- CN113515885B CN113515885B CN202110439005.4A CN202110439005A CN113515885B CN 113515885 B CN113515885 B CN 113515885B CN 202110439005 A CN202110439005 A CN 202110439005A CN 113515885 B CN113515885 B CN 113515885B
- Authority
- CN
- China
- Prior art keywords
- photovoltaic module
- current
- health
- soh
- aging
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Photovoltaic Devices (AREA)
- Testing Of Individual Semiconductor Devices (AREA)
Abstract
The application discloses a photovoltaic module intelligent health state diagnosis method, which comprises the following steps: step one: sampling the output I-V characteristic of the photovoltaic module to be tested, and identifying the aging parameter by using an MQPSO algorithm; step two: mapping the aging parameters under any working environment to STC; step three: and (3) inputting the aging parameters into the PNN, and calculating the maximum likelihood estimated value of each category to obtain the current health condition of the photovoltaic module to be tested. According to the method, only the working I-V characteristic curve of the current photovoltaic module, the environment irradiation and the environment temperature are required to be obtained, so that the current health state SoH of the photovoltaic module can be embodied, and simulation and experiment results show that the method can effectively represent the current health state of the photovoltaic module, reduce the influence of the working environment on the health state of the module, and has a certain reference value for the fault early warning of the photovoltaic module and the establishment of a health management system.
Description
Technical Field
The application belongs to the field of diagnosis of photovoltaic modules, and particularly relates to an intelligent health state diagnosis method of a photovoltaic module.
Background
After the photovoltaic module is installed outdoors, the electrical characteristics of the photovoltaic module can be reduced along with the increase of the service time, and the power of the crystalline silicon solar cell is reduced by about 0.5% per year on average. The reasons of the problems are mainly that the photovoltaic module generates phenomena of yellowing, corrosion, hidden cracking, cable damage, welding spot aging, bypass diode failure and the like, and the Health State of the module (SoH) is reduced. Although this phenomenon does not temporarily cause damage to the system, it may be an early sign of a serious failure of the components, thereby reducing the operational life and economic efficiency of the photovoltaic power plant. Therefore, the detection of the SoH of the photovoltaic module is an important means for troubleshooting and maintaining the normal operation of the system.
At present, some detection methods related to the component SoH have been proposed at home and abroad, and mainly can be divided into two major categories, namely on-site on-line monitoring and time domain detection based on photovoltaic output characteristics. In-situ monitoring mainly utilizes infrared thermal imaging and electroluminescence techniques. Roderto Pierdicca and the like, a photovoltaic array health monitoring system is designed, a thermal infrared image is shot above an array by using an unmanned aerial vehicle, and then a hot spot in the array is identified by using an image processing algorithm, so that the health state of the array is diagnosed; frazao et al use consumer-grade cameras to observe electroluminescence of photovoltaic modules, and can accurately locate hidden cracks and breakage within the modules. Time domain detection methods based on photovoltaic output characteristics are more commonly studied. KUN DING et al used empirical mode decomposition and gaussian mixture model to detect health status of a 5*2 photovoltaic array: and carrying out Gaussian mixture clustering on the voltage and the current in a time window, examining the deviation condition of an experimental cluster center and a reference cluster center, and evaluating the health state of the photovoltaic array. Simona-Vasilicon Oprea et al propose an ultra-short term output power prediction model applied to a large photovoltaic system: and predicting the output power of the system in a sliding time window in the future by utilizing a big data technology and an artificial neural network to obtain the key performance index for evaluating the health state of the system. Besides, a plurality of sensor methods based on the operation parameters of the power station, a time domain reflection method based on high frequency signal injection, a parasitic capacitance estimation method, and the like, which evaluate the health condition of the photovoltaic module according to the deviation of the monitored and calculated values from the theoretical values, have been proposed by the scholars.
The various methods have the disadvantage of not being neglected: the infrared thermal imaging and electroluminescence technology based on field test has high reliability, but requires high experimental cost, and is difficult to popularize on a large scale; the time domain detection method based on the photovoltaic output characteristics is easy to be interfered by environmental factors; the multi-sensor method for detecting the operation parameters of the power station needs to collect numerous data including environment, operation and maintenance and working data, and needs many equipment assistance; the detection accuracy of the high frequency signal injection and parasitic capacitance estimation is low. More importantly, the methods can only evaluate the change of the health state of the photovoltaic module, cannot quantitatively judge the current SoH of the photovoltaic module, and are difficult to apply to the health management system of the photovoltaic power station.
Disclosure of Invention
The application aims to overcome the problems existing in the prior art and provides an intelligent health state diagnosis method for a photovoltaic module.
In order to achieve the technical purpose and the technical effect, the application is realized by the following technical scheme:
an intelligent health state diagnosis method for a photovoltaic module, comprising the following steps:
step one: sampling the output I-V characteristic of the photovoltaic module to be tested, and identifying the aging parameter by using an MQPSO algorithm;
step two: mapping the aging parameters under any working environment to STC;
step three: and (3) inputting the aging parameters into the PNN, and calculating the maximum likelihood estimated value of each category to obtain the current health condition of the photovoltaic module to be tested.
Further, in the first step, the aging parameter is identified specifically by calculating the root mean square error between the model I-V and the real I-V, where the formula is:
further, the aging parameter mapping formula in the second step is specifically as follows:
further, after the aging parameter is input to the PNN in the third step, the estimated amount of the multi-element PDF is estimated by the following formula:
when PNN is used for classifying tasks, the network model consists of four layers, namely an input layer, a mode layer, a summation layer and a decision layer;
the input layer transmits information to the mode layer, and calculates the maximum likelihood estimated value of each neuron according to the above;
the summation layer carries out summarization and averaging on all the neuron output information belonging to the same category, classifies the sample X according to Bayesian decision rules, and outputs the sample X by the decision layer.
Further, the current health status SoH of the photovoltaic module to be tested in the third step is calculated as follows:
when SoH takes on a value of-1, the component is represented to have serious faults; when the SoH value of the component is 1, the current working state of the component is very good; when SoH is at [0,1], the larger the value, the healthier; at [ -1,0], the smaller the value, the less healthy.
Further, select |soh|=0.5 as a boundary, divide the range of health SoH into three parts: health, sub-health and failure groups.
The beneficial effects of the application are as follows:
according to the method, the current health state SoH of the photovoltaic module can be reflected only by acquiring the working I-V characteristic curve of the current photovoltaic module, environmental irradiation and environmental temperature, the aging parameters of the photovoltaic module are firstly extracted through the I-V characteristic curve by utilizing an improved quantum particle swarm algorithm and are then mapped to the standard test conditions, finally the aging parameters are input into the probability neural network, the current SoH of the photovoltaic module is calculated, and simulation and experimental results show that the method can effectively represent the current health state of the photovoltaic module, reduce the influence of the working environment on the health state of the module, and has a certain reference value for the fault early warning of the photovoltaic module and the establishment of a health management system.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
FIG. 1 is a flowchart of the MPQSO algorithm in the present application;
fig. 2 is a diagram of a probabilistic neural network in the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
An intelligent health state diagnosis method for a photovoltaic module, comprising the following steps:
step one: sampling the output I-V characteristic of the photovoltaic module to be tested, and identifying the aging parameter by using an MQPSO algorithm;
step two: mapping the aging parameters under any working environment to STC;
step three: and (3) inputting the aging parameters into the PNN, and calculating the maximum likelihood estimated value of each category to obtain the current health condition of the photovoltaic module to be tested.
The following principle analysis:
the output of the photovoltaic module is nonlinear. In order to effectively evaluate the health of a component, it is necessary to first analyze the output characteristics of the component and select an appropriate aging factor.
During use of the photovoltaic module, the health condition may be inevitably lowered. Research has found that the causes of the decline in component health generally include three categories: optical degradation, electrical aging, and other factors.
Optical degradation causes yellowing, delamination and glass breakage of the front plate, resulting in reduced light transmittance of the assembly; the electrical aging comprises welding spot aging, transmission cable damage, battery piece damage, short circuit and the like, so that the parasitic series resistance of the component is increased; other factors relate to potential induced decay, bypass diode aging, etc., resulting in an increase in parasitic shunt resistance. These phenomena reduce the output power of the module, leaving the photovoltaic module "healthy".
The photovoltaic module is generally formed by connecting a plurality of single batteries in series, and in an ideal state, the characteristics of each single battery in the module are consistent, the output current is the same as the current of each single battery, and the output voltage is the sum of the voltages of the single batteries. The output current equation of the photovoltaic module can be obtained by the method
Wherein U is the output voltage of the component; i is the output current of the component; i0 is the diode reverse saturation current; rs is equivalent series resistance; rp is the equivalent parallel resistance; a can be calculated by the following formula:
wherein q is an electron charge constant, 1.6x10-19C; t is absolute temperature in Kelvin; k is Boltzmann constant, 1.38 x 10-23J/K, n is diode management ideal factor.
The quantum particle swarm is a PSO algorithm model proposed from the quantum mechanics perspective. In QPSO, a wave function is used for describing the state of the particles, a probability density function of the particles at a certain point in space is obtained through a Schrodinger equation, and then a position equation of the particles is obtained through a Monte Carlo simulation mode. Assume that in an N-dimensional search space, a population X (t) = { X1 (t), X2 (t), …, xm (t) } of M groups of particles representing potential problem solutions, wherein the position of each particle Xi (t) at time t is determined by formula (3)
Wherein u (t) is a random number subject to uniform distribution on [0,1], p is the attractor position, L is a convergent iterative strategy, determined by equations (4) - (5), respectively
L i,j (t)＝α·|C j (t)-X i,j (t)| (5)
Wherein P (t) is the individual optimal position, G (t) is the global optimal position, and C (t) is the average best position in the population
The α in equation (5), called the contraction-expansion coefficient, is the only control parameter in the overall algorithm except for the population size and the number of iterations. Generally, a linear reduction strategy is adopted, and the value is taken according to the formula (7)
Linearly decreasing from m to n with the number of iterations, typically taking m=1, n=0.5; k is the maximum number of iterations. The particle evolution equation for the final QPSO is:
randomly selecting one dimension in the global optimum position G (t), assuming the d dimension, and applying Gaussian disturbance on the one dimension to finish the variation operation of the global optimum
Wherein G is d For the selected dimension in the global optimum position, it is a random number subject to gaussian distribution, with mean μ=0. The standard deviation sigma, also known as elite learning rate, is calculated according to equation (10) following some neural network training schemes with sigma decreasing linearly with iteration number
Where a and b represent the upper and lower limits of σ, studies indicate that taking a=1, b=0.1, there is better performance in the most test functions.
The MQPSO algorithm flow chart is shown in fig. 1: after the perturbation is applied to G (t) according to equation (9) (10), re-evaluating the adaptation value of G (t), the new position being taken by G (t) if and only if the adaptation value of G (t) is better than the adaptation value of the current position; otherwise the new position will replace the particle with the worst adaptation value in the current generation of particles.
The main problem in modeling a block of photovoltaic modules is how to determine the module electrical parameters in equation (1) so that the simulation output approximates as closely as possible the true module I-V characteristics. From an optimization perspective, this can be equivalent to a minimization problem, considering that the difference between the real I-V and the model I-V characteristics is minimized by a specific set of identification parameters, the most intuitive way to measure this difference is to calculate the root mean square error (Root Mean Squared Error, RMSE) of the model I-V and the real I-V
In the middle ofAnd->Respectively representing the ith current and voltage measurement value in N measurements, x represents the set of parameters to be identified,/v>Representing the output current calculated by the simulation model according to equation (1)
Where x= { Iph, I0, a, rs, rp } represents five parameters in the single diode model, including the photovoltaic module burn-in factors Iph, rs, rp analyzed above. And (3) carrying out QPSO parameter identification according to the I-V characteristics measured in the real environment, so that the key aging factors of the component can be extracted.
The output of the component depends on the irradiation G and the temperature T, and the parameter identification results of the photovoltaic component in different working environments are different. Such differences can affect the accuracy of subsequent health assessment. According to the method, the identification results of different irradiation and temperatures are mapped to STC, so that the accuracy of an evaluation algorithm is improved, and the calculated amount is reduced. The photovoltaic module characteristic parameters Iph, I0, a, rs, rp change along with G and T to satisfy formulas (13) - (17).
The photo-generated current Iph is affected by both the irradiation G and the temperature T: the photo-generated current value of any working environment can be calculated according to the formula (13) in direct proportion to the irradiation G and in linear relation to the temperature T
Where ki is the temperature coefficient of the component; gstc and Tstc are the irradiation and working temperatures under standard test conditions, 1000W/m2 and 25℃respectively.
Diode reverse saturation current I 0 Has a very close relationship with the battery operating temperature, calculated according to formula (14)
Where ε is the band gap width of the cell and the band gap width at single crystal silicon STC is 1.12eV, depending on the cell material and temperature.
Diode idealities are determined by semiconductor materials and their fabrication techniques, and no study has been made to show that as a component ages, its idealities change. Only the change in thermal voltage is considered here.
According to simulation and experimental results, the resistance is regarded as being changed along with temperature and irradiation, and is calculated according to formulas (16) and (17)
Then the functions in the formulas (13) - (17) are inverted and substituted into all known constants to obtain a 5-parameter characteristic mapping formula from any working environment of the formulas (18) - (22) to the standard test condition
PNN is a feed-forward neural network developed from radial basis functions whose theoretical basis is based on Bayes decision theory and the Probability Density Function (PDF) estimation method of Parzen window, and estimates the estimated quantity g (x) of the multiple PDF by equation (23). In practical application, the classifier of the PNN is very good in performance, PNN training is easy, and the classifier has high classification precision.
Where N is the total number of training samples, x is the vector of random variables, and xi is the training vector. Is a smoothing parameter, which is the only parameter to be determined in the PNN. Delta=0.9 was chosen in the present application.
When PNN is used for classification tasks, the network model is composed of four layers, namely an input layer, a mode layer, a summation layer and a decision layer, as shown in fig. 2. The input layer delivers information to the pattern layer and calculates maximum likelihood estimates for each neuron according to equation (23).
The summation layer carries out summarization and averaging on all neuron output information belonging to the same category, and classifies the samples X according to a Bayesian decision rule:
C(X)＝argmax(f i (X)) (25)
where C (X) represents the final classification of sample X, which is output by the decision layer.
Photovoltaic module health is a fuzzy concept. Current research classifies photovoltaic modules into two major categories, health and failure, but the module is either a transient process from health to failure, and the state of health is the state of change that describes this transition. Due to this ambiguous definition, classification modeling like PNN has drawbacks, and in order to overcome this problem, the concept of photovoltaic module SoH is proposed.
From equation (24), PNN can derive maximum likelihood estimates fi (x) for the samples under different categories. The following SoH calculation mode is now defined
Where f (health) and f (fault) are each class estimates of formula (24). The structure of the PNN network needs to be changed properly for calculating SoH, the decision layer is deleted from the network, the maximum likelihood estimated values of the summation layer for various types are directly output, and then the calculation is carried out through the step (25).
As can be seen from the expression, the essence of SoH is the degree of difference between health and failure, the larger the value is, the more obvious the difference is, and when the value is 0, it is proved that the component cannot be distinguished whether it is healthy or failed, and is in a typical sub-health state.
According to such definition rules, the range of SoH should be between [ -1,1], when SoH takes on a value of-1, it represents that the component has failed seriously; when the component SoH takes a value of 1, it represents that the current working state of the component is very good. When SoH is at [0,1], the larger the value, the healthier; at [ -1,0], the smaller the value, the less healthy. For better qualitative analysis, the application selects |soh|=0.5 as a boundary line, and divides the range of SoH into three parts: health, sub-health and failure groups.
The foregoing has shown and described the basic principles, principal features and advantages of the application. It will be understood by those skilled in the art that the present application is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present application, and various changes and modifications may be made without departing from the spirit and scope of the application, which is defined in the appended claims.
Claims (2)
1. A photovoltaic module intelligent health state diagnosis method is characterized in that: the diagnostic method comprises the steps of:
step one: sampling the output I-V characteristic of the photovoltaic module to be tested, and identifying the aging parameter by using an MQPSO algorithm;
step two: mapping the aging parameters under any working environment to STC;
step three: inputting the aging parameters into the PNN, and calculating the maximum likelihood estimation value of each category to obtain the current health condition of the photovoltaic module to be tested;
in the first step, the aging parameters are identified specifically by calculating the root mean square error of the model I-V and the real I-V, and the formula is as follows:
iim and Vim respectively represent the ith current and voltage measured values in N times of measurement, x represents a set of parameters to be identified, and Isiim represents the output current calculated by the simulation model;
the aging parameter mapping formula in the second step is specifically as follows:
photo-generated current is I ph The irradiation is G, the temperature is T, k i The reverse saturation current of the diode is I as the temperature coefficient of the component 0 Epsilon is the band gap width of the cell,
in the third step, after the aging parameters are input into the PNN, estimating the estimation amount of the multi-element PDF according to the following formula:
where N is the total number of training samples, x is the vector of random variables, x i Is a training vector;
when PNN is used for classifying tasks, the network model consists of four layers, namely an input layer, a mode layer, a summation layer and a decision layer;
the input layer transmits information to the mode layer, and calculates the maximum likelihood estimated value of each neuron according to the above;
the summation layer carries out summarization and averaging on all the neuron output information belonging to the same category, classifies the sample X according to a Bayesian decision rule, and outputs the sample X by the decision layer;
in the third step, the current health status SoH of the photovoltaic module to be tested is calculated as follows:
wherein f (health) and f (fault) are estimated values of each class;
when SoH takes on a value of-1, the component is represented to have serious faults; when the SoH value of the component is 1, the current working state of the component is very good; when SoH is at [0,1], the larger the value, the healthier; at [ -1,0], the smaller the value, the less healthy.
2. The intelligent health state diagnosis method for the photovoltaic module according to claim 1, wherein the method comprises the following steps of: selecting |soh|=0.5 as a boundary line, dividing the range of health status SoH into three parts: health, sub-health and failure groups.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110439005.4A CN113515885B (en) | 2021-04-23 | 2021-04-23 | Intelligent health state diagnosis method for photovoltaic module |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110439005.4A CN113515885B (en) | 2021-04-23 | 2021-04-23 | Intelligent health state diagnosis method for photovoltaic module |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113515885A CN113515885A (en) | 2021-10-19 |
| CN113515885B true CN113515885B (en) | 2023-09-12 |
Family
ID=78061704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110439005.4A Active CN113515885B (en) | 2021-04-23 | 2021-04-23 | Intelligent health state diagnosis method for photovoltaic module |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113515885B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20020059302A (en) * | 2002-06-03 | 2002-07-12 | 윤승룡 | Photovoltaic ion toothbrush with character drawing and electroluminescent display. |
| JP2013000991A (en) * | 2011-06-17 | 2013-01-07 | Roland Dg Corp | Inkjet printer |
| CN103676817A (en) * | 2013-06-13 | 2014-03-26 | 奇瑞汽车股份有限公司 | Communication base station photoelectric complementary power supply system and operation method thereof |
| CN104734362A (en) * | 2015-04-07 | 2015-06-24 | 上海理工大学 | Photovoltaic inverter health status monitoring and alarm system |
| CN104767482A (en) * | 2014-01-02 | 2015-07-08 | 上海岩芯电子科技有限公司 | Method for diagnosing aging and short circuit of photovoltaic module online |
| CN104951607A (en) * | 2015-06-15 | 2015-09-30 | 中国建筑设计咨询有限公司 | Method for wind-induced vibration calculation of photovoltaic supporting system based on fluid-solid coupling simulation |
| CN105023070A (en) * | 2015-08-12 | 2015-11-04 | 河海大学常州校区 | Output power prediction method of photovoltaic system |
| CN105186588A (en) * | 2014-02-24 | 2015-12-23 | 通用电气能源能量变换技术有限公司 | Battery energy storage system with arc flash protection, energy conversion system and protection method |
| CN106021806A (en) * | 2016-06-06 | 2016-10-12 | 福州大学 | Photovoltaic string fault diagnosis method based on kernel function limit learning machine |
| CN107579707A (en) * | 2017-10-13 | 2018-01-12 | 江苏大学 | A kind of diagnosing failure of photovoltaic array method based on parameter identification |
| CN108009730A (en) * | 2017-12-05 | 2018-05-08 | 河海大学常州校区 | A kind of photovoltaic power station system health status analysis method |
| CN108075728A (en) * | 2016-11-15 | 2018-05-25 | 上海英孚特电子技术有限公司 | A kind of photovoltaic system DC side arc fault type identification and protective device |
| CN209085657U (en) * | 2017-08-02 | 2019-07-09 | 强力物联网投资组合2016有限公司 | For data gathering system related or industrial environment with chemical production technology |
| CN110148068A (en) * | 2019-05-23 | 2019-08-20 | 福州大学 | One kind being based on meteorological data similarity analysis and LSTM neural fusion photovoltaic plant ultra-short term power forecasting method |
| CN111610410A (en) * | 2020-05-27 | 2020-09-01 | 上海岩芯电子科技有限公司 | SSTDR technology-based photovoltaic cable sub-health detection and positioning method |
-
2021
- 2021-04-23 CN CN202110439005.4A patent/CN113515885B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20020059302A (en) * | 2002-06-03 | 2002-07-12 | 윤승룡 | Photovoltaic ion toothbrush with character drawing and electroluminescent display. |
| JP2013000991A (en) * | 2011-06-17 | 2013-01-07 | Roland Dg Corp | Inkjet printer |
| CN103676817A (en) * | 2013-06-13 | 2014-03-26 | 奇瑞汽车股份有限公司 | Communication base station photoelectric complementary power supply system and operation method thereof |
| CN104767482A (en) * | 2014-01-02 | 2015-07-08 | 上海岩芯电子科技有限公司 | Method for diagnosing aging and short circuit of photovoltaic module online |
| CN105186588A (en) * | 2014-02-24 | 2015-12-23 | 通用电气能源能量变换技术有限公司 | Battery energy storage system with arc flash protection, energy conversion system and protection method |
| CN104734362A (en) * | 2015-04-07 | 2015-06-24 | 上海理工大学 | Photovoltaic inverter health status monitoring and alarm system |
| CN104951607A (en) * | 2015-06-15 | 2015-09-30 | 中国建筑设计咨询有限公司 | Method for wind-induced vibration calculation of photovoltaic supporting system based on fluid-solid coupling simulation |
| CN105023070A (en) * | 2015-08-12 | 2015-11-04 | 河海大学常州校区 | Output power prediction method of photovoltaic system |
| CN106021806A (en) * | 2016-06-06 | 2016-10-12 | 福州大学 | Photovoltaic string fault diagnosis method based on kernel function limit learning machine |
| CN108075728A (en) * | 2016-11-15 | 2018-05-25 | 上海英孚特电子技术有限公司 | A kind of photovoltaic system DC side arc fault type identification and protective device |
| CN209085657U (en) * | 2017-08-02 | 2019-07-09 | 强力物联网投资组合2016有限公司 | For data gathering system related or industrial environment with chemical production technology |
| CN110073301A (en) * | 2017-08-02 | 2019-07-30 | 强力物联网投资组合2016有限公司 | The detection method and system under data collection environment in industrial Internet of Things with large data sets |
| CN107579707A (en) * | 2017-10-13 | 2018-01-12 | 江苏大学 | A kind of diagnosing failure of photovoltaic array method based on parameter identification |
| CN108009730A (en) * | 2017-12-05 | 2018-05-08 | 河海大学常州校区 | A kind of photovoltaic power station system health status analysis method |
| CN110148068A (en) * | 2019-05-23 | 2019-08-20 | 福州大学 | One kind being based on meteorological data similarity analysis and LSTM neural fusion photovoltaic plant ultra-short term power forecasting method |
| CN111610410A (en) * | 2020-05-27 | 2020-09-01 | 上海岩芯电子科技有限公司 | SSTDR technology-based photovoltaic cable sub-health detection and positioning method |
Non-Patent Citations (1)
| Title |
|---|
| 基于希尔伯特黄变换和灰色关联度的光伏***健康状态研究;丁坤 等;《kezaishegn》;第37卷(第3期);342-348 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113515885A (en) | 2021-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110336534B (en) | Fault diagnosis method based on photovoltaic array electrical parameter time series feature extraction | |
| CN109034191B (en) | ELM-based one-dimensional telemetry data abnormal interpretation method | |
| Van Gompel et al. | Satellite based fault diagnosis of photovoltaic systems using recurrent neural networks | |
| CN110879377B (en) | Metering device fault tracing method based on deep belief network | |
| CN109672406B (en) | Photovoltaic power generation array fault diagnosis and classification method based on sparse representation and SVM | |
| CN105846780A (en) | Decision tree model-based photovoltaic assembly fault diagnosis method | |
| Zaki et al. | Deep‐learning–based method for faults classification of PV system | |
| CN103103570B (en) | Based on the aluminium cell condition diagnostic method of pivot similarity measure | |
| CN112085621B (en) | Distributed photovoltaic power station fault early warning algorithm based on K-Means-HMM model | |
| CN114004137A (en) | Multi-source meteorological data fusion and pretreatment method | |
| CN108092623A (en) | A kind of photovoltaic array multisensor fault detecting and positioning method | |
| CN109617526A (en) | A method of photovoltaic power generation array fault diagnosis and classification based on wavelet multiresolution analysis and SVM | |
| Mustafa et al. | Fault identification for photovoltaic systems using a multi-output deep learning approach | |
| CN113378449A (en) | Photovoltaic module health state diagnosis method based on fuzzy C-means clustering | |
| CN116610998A (en) | Switch cabinet fault diagnosis method and system based on multi-mode data fusion | |
| Sebbane et al. | Performance study of artificial neural network training algorithms for classifying pv field defects | |
| Zhu et al. | Photovoltaic failure diagnosis using sequential probabilistic neural network model | |
| Liu et al. | Online diagnosis and prediction of power battery voltage comprehensive faults for electric vehicles based on multi-parameter characterization and improved K-means method | |
| CN113515885B (en) | Intelligent health state diagnosis method for photovoltaic module | |
| CN117554821A (en) | Energy storage battery on-line monitoring management method and system based on deep learning network | |
| Fagiani et al. | Computational Intelligence in Smart water and gas grids: An up-to-date overview | |
| Gong et al. | Photovoltaic arrays fault diagnosis based on an improved dilated convolutional neural network with feature-enhancement | |
| Benninger et al. | Anomaly detection by comparing photovoltaic systems with machine learning methods | |
| Bi et al. | A fault diagnosis algorithm for wind turbine blades based on bp neural network | |
| CN115310653A (en) | Photovoltaic array residual life prediction method based on performance degradation data |
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 |