CN115811278B - Distributed photovoltaic fault detection method and device based on discrete rate - Google Patents
Distributed photovoltaic fault detection method and device based on discrete rate Download PDFInfo
- Publication number
- CN115811278B CN115811278B CN202310075736.4A CN202310075736A CN115811278B CN 115811278 B CN115811278 B CN 115811278B CN 202310075736 A CN202310075736 A CN 202310075736A CN 115811278 B CN115811278 B CN 115811278B
- Authority
- CN
- China
- Prior art keywords
- discrete
- rate
- distributed photovoltaic
- discrete rate
- data
- 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
Images
Classifications
-
- 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
- Photovoltaic Devices (AREA)
Abstract
The invention relates to an artificial intelligence technology, and discloses a distributed photovoltaic fault detection method and a distributed photovoltaic fault detection device based on a discrete rate, wherein the method comprises the following steps: acquiring real-time data of the distributed photovoltaic, and generating a string discrete rate of the distributed photovoltaic according to the real-time data and a preset discrete rate algorithm; determining the abnormal inverter of the distributed photovoltaic by using a preset discrete rate threshold value and a group string discrete rate; determining the electric quantity loss of the abnormal inverter according to the real-time data, and filtering the pseudo discrete rate variation of the group of string discrete rates according to the electric quantity loss to obtain the true discrete rate variation of the abnormal inverter; and classifying the true discrete rate variable quantity by using a support vector regression algorithm to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity, and determining the fault component of the distributed photovoltaic according to the classified discrete rate variable quantity. The method and the device can improve the accuracy of distributed photovoltaic fault detection.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a distributed photovoltaic fault detection method and device based on a discrete rate.
Background
With the rapid development of global economy, people have an increasing demand for various forms of energy. The development of photovoltaic power generation is seriously restricted by the progress of photovoltaic power generation technology, the increase of the grid-connected operation scale of photovoltaic power generation, the problems of optimization, improvement, operation cost and the like of a photovoltaic power station. The large grid-connected inverter plays a role in feeding electric energy to a power grid, the fault of any key component in a main circuit of the large grid-connected inverter can cause the whole photovoltaic power station to stop and even damage equipment, and the long stop time can reduce the power generation benefit of the power station. These faults of the photovoltaic power station seriously affect the normal operation of the photovoltaic power generation system. In order to prevent more serious accidents caused by faults, reduce the yield loss of a power station, detect the faults of photovoltaic power station equipment in time, contribute to stable and efficient operation of the photovoltaic power station and facilitate taking corresponding measures at the beginning of the faults of the photovoltaic power station equipment.
Today, equipment detection of photovoltaic power plant equipment comprises: infrared image with based on the electric signal detection, but infrared image detection method has the detection precision not high, and the shortcoming such as equipment cost is big, and the detection method based on the electric signal also has self limitation, consequently how to promote distributed photovoltaic fault detection accuracy, becomes the problem that urgently needs to be solved.
Disclosure of Invention
The invention provides a distributed photovoltaic fault detection method and device based on a discrete rate, and mainly aims to solve the problem of low accuracy rate in distributed photovoltaic fault detection.
In order to achieve the above object, the present invention provides a distributed photovoltaic fault detection method based on a discrete rate, including:
acquiring real-time data of distributed photovoltaics, and generating a string discrete rate of the distributed photovoltaics according to the real-time data and a preset discrete rate algorithm;
determining an abnormal inverter of the distributed photovoltaic by using a preset discrete rate threshold value and the group string discrete rate;
determining the electric quantity loss of the abnormal inverter according to the real-time data, and filtering the pseudo discrete rate variation of the group of discrete rates according to the electric quantity loss to obtain the true discrete rate variation of the abnormal inverter;
and classifying the true discrete rate variable quantity by utilizing a support vector regression algorithm to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity, and determining the fault component of the distributed photovoltaic according to the classified discrete rate variable quantity.
Optionally, the generating a string discrete rate of the distributed photovoltaic according to the real-time data and a preset discrete rate algorithm includes:
carrying out numerical value type conversion on the real-time data to obtain uniform format data of the real-time data;
carrying out standardization processing on the uniform format data to obtain standard data of the uniform format data;
and calculating the group string dispersion rate of the distributed photovoltaic by utilizing the standard data and a preset dispersion rate algorithm.
Optionally, the determining the abnormal inverter of the distributed photovoltaic by using the preset discrete rate threshold and the group string discrete rate includes:
performing difference processing on the group string discrete rate and a preset discrete rate threshold value to obtain a difference value between the group string discrete rate and the preset discrete rate threshold value;
and when the difference value is a positive value, the inverter corresponding to the group of string dispersion ratios is an abnormal inverter of the distributed photovoltaic.
Optionally, the determining, according to the real-time data, the power loss of the abnormal inverter includes:
extracting the characteristics of the real-time data to obtain the real-time characteristics of the real-time data, and determining electric quantity data in the real-time data according to the real-time characteristics;
and carrying out abnormity analysis on the electric quantity data to obtain the electric quantity loss of the abnormal inverter.
Optionally, the filtering, according to the power loss, the pseudo discrete rate variation of the group of discrete rates to obtain the true discrete rate variation of the abnormal inverter includes:
performing vacant branch filtering on the group of string discrete rates according to the electric quantity loss to obtain a first-stage discrete rate of the abnormal inverter;
performing weather factor filtering on the primary dispersion rate to obtain a secondary dispersion rate of the primary dispersion rate;
and carrying out building shadow filtering on the second-stage discrete rate to obtain a third-stage discrete rate of the second-stage discrete rate, and determining the third-stage discrete rate as the true discrete rate variation of the abnormal inverter.
Optionally, the classifying the proper discrete rate variation by using a support vector regression algorithm to obtain a classified discrete rate variation of the proper discrete rate variation includes:
constructing an initial support vector machine of the true discrete rate variable quantity by using an initial function of a support vector regression algorithm;
acquiring a training set of the initial support vector, and determining an input data matrix and a label matrix of the initial support vector machine according to the training set;
constructing a Lagrange equation of the initial support vector machine, and generating equation parameters of the Lagrange equation;
calculating target parameters of the initial support vector machine according to the equation parameters, generating a standard support vector machine of the true discrete rate variable quantity according to the target parameters, and classifying the true discrete rate variable quantity by using the standard support vector machine to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity.
Optionally, the constructing an initial support vector machine of the variance of the truth discrete rate by using an initial function of a support vector regression algorithm includes:
constructing a hyperplane of the true discrete rate variation by using an initial function of the support vector regression algorithm as follows:
wherein the content of the first and second substances,
is an initial function of a support vector regression algorithm, based on>
Is a transposition of the first linear parameter, is present>
Is the said true dispersion rate change amount>
Is a second linear parameter;
generating a plane norm of the hyperplane, and determining an objective function and a constraint condition of the true discrete rate variation according to the plane norm;
and generating an initial support vector machine of the true discrete rate variation according to the objective function and the constraint condition.
Optionally, the generating equation parameters of the lagrangian equation includes:
calculating the partial derivative of a first linear parameter of the Lagrange equation to obtain a first partial derivative function of the first linear parameter;
solving the partial derivative of a second linear parameter of the Lagrange equation to obtain a second partial derivative function of the second linear parameter;
determining a first optimization parameter and a second optimization parameter of the change amount of the truth discrete rate according to the first partial derivative function, the second partial derivative function and the constraint condition of the change amount of the truth discrete rate, and determining that the first optimization parameter and the second optimization parameter are equation parameters of the Lagrangian equation.
Optionally, the calculating target parameters of the initial support vector machine according to the equation parameters includes:
substituting the equation parameters into an initial function of the support vector regression algorithm to obtain a function solution of the initial function;
and determining the minimum value of the objective function of the true discrete rate variation by using a preset sequence minimum optimization algorithm, and generating the objective parameter of the initial support vector machine according to the minimum value.
In order to solve the above problem, the present invention further provides a distributed photovoltaic fault detection apparatus based on a discrete rate, the apparatus including:
the string discrete rate module is used for acquiring real-time data of the distributed photovoltaic, and generating the string discrete rate of the distributed photovoltaic according to the real-time data and a preset discrete rate algorithm, wherein the preset discrete rate algorithm is as follows:
wherein the content of the first and second substances,
is the string dispersion ratio of the distributed photovoltaic, is->
Is in the normative data->
The mean value of the attributes is determined,
is the total number of data in the attribute, and>
is a data flag in an attribute, is asserted>
Is attribute flag, is asserted>
Is in the normative data->
Standard deviation of attribute, <' >>
Is in the normative data->
First +of attribute>
A piece of data;
the abnormal inverter module is used for determining an abnormal inverter of the distributed photovoltaic by utilizing a preset discrete rate threshold value and the group string discrete rate;
the spurious filtering module is used for determining the electric quantity loss of the abnormal inverter according to the real-time data, and filtering the spurious discrete rate variation of the group of the discrete rates according to the electric quantity loss to obtain the true discrete rate variation of the abnormal inverter;
and the fault detection module is used for classifying the true discrete rate variable quantity by utilizing a support vector regression algorithm to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity, and determining the fault component of the distributed photovoltaic according to the classified discrete rate variable quantity.
The method comprises the steps of generating a string discrete rate through real-time data of the distributed photovoltaic, performing initial abnormal judgment on the distributed photovoltaic by using the string discrete rate to determine an abnormal transformer, performing false discrete rate variable quantity filtering on the string discrete rate according to electric quantity loss, eliminating interference of a part of false factors, improving fault judgment precision, classifying the true discrete rate variable quantity by using a support vector regression algorithm, classifying possible fault conditions, obtaining a detailed fault classification result, and enabling the precision rate of distributed photovoltaic fault detection to be higher.
Drawings
Fig. 1 is a schematic flowchart of a distributed photovoltaic fault detection method based on a discrete rate according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating pseudo-discrete-rate change filtering according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a process of generating variance of discrete rate in classification according to an embodiment of the present invention;
fig. 4 is a functional block diagram of a distributed photovoltaic fault detection apparatus based on discrete rate according to an embodiment of the present invention;
the implementation, functional features and advantages of the present invention will be further described with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
The embodiment of the application provides a distributed photovoltaic fault detection method based on a discrete rate. The execution subject of the distributed photovoltaic fault detection method based on the discrete rate includes, but is not limited to, at least one of electronic devices, such as a server, a terminal, and the like, which can be configured to execute the method provided by the embodiments of the present application. In other words, the discrete-rate-based distributed photovoltaic fault detection method may be performed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides basic cloud computing services such as cloud service, a cloud database, cloud computing, cloud functions, cloud storage, web service, cloud communication, middleware service, domain name service, security service, content Delivery Network (CDN), and a big data and artificial intelligence platform.
Referring to fig. 1, a schematic flow chart of a distributed photovoltaic fault detection method based on a discrete rate according to an embodiment of the present invention is shown. In this embodiment, the distributed photovoltaic fault detection method based on the discrete rate includes:
s1, acquiring real-time data of distributed photovoltaics, and generating a string discrete rate of the distributed photovoltaics according to the real-time data and a preset discrete rate algorithm.
In the embodiment of the invention, the distributed photovoltaic is particularly a photovoltaic power generation facility which is built near a user site and has the characteristics of self-service use at a user side, net surfing of redundant electric quantity and balance adjustment in a power distribution system in an operation mode. The distributed photovoltaic power generation follows the principle of local conditions, cleanness, high efficiency, scattered layout and near utilization, makes full use of local solar energy resources, replaces and reduces fossil energy consumption, and particularly relates to a distributed power generation system which directly converts solar energy into electric energy by adopting photovoltaic components.
In detail, the real-time data includes, but is not limited to: the installation condition of the distributed photovoltaic system, the real-time power generation amount of the distributed photovoltaic system and the like.
In detail, in the operation and maintenance work of the photovoltaic power station, the group string dispersion rate is one of the examination indexes of the operation condition of the photovoltaic power station, when the photovoltaic power station works normally, the dispersion rate of the group string reflects the whole operation condition of each group string on the header box side or the group string inverter side, the lower the dispersion rate is, the more concentrated the power generation performance of each group string is, the better the consistency is, the higher the dispersion rate is, the lower the stability is, wherein the consistency condition of the power generation performance of the photovoltaic group string is an important index of the health state of the photovoltaic power station, the lower the dispersion rate is, the better the consistency of each branch current curve is, the more stable the power generation condition is, the higher the dispersion rate is, the worse the consistency is, and a fault unit generally exists.
In an embodiment of the present invention, the generating a string discrete rate of the distributed photovoltaic according to the real-time data and a preset discrete rate algorithm includes:
carrying out numerical value type conversion on the real-time data to obtain uniform format data of the real-time data;
carrying out standardization processing on the uniform format data to obtain standard data of the uniform format data;
and calculating the group string dispersion rate of the distributed photovoltaic by utilizing the standard data and a preset dispersion rate algorithm.
In detail, the performing of the numerical type conversion on the real-time data may select a conversion command according to the data type of the real-time data, and convert the real-time data into the numerical data by using the conversion command, for example: the string is converted to a numeric value using the "help triggering" command.
In detail, the standardization process not only does not "invalidate" the uniformly formatted data, but also improves the expressive power of the uniformly formatted data.
In detail, the preset discrete rate algorithm is as follows:
wherein the content of the first and second substances,
is the string dispersion ratio of the distributed photovoltaic, is->
Is in the normative data->
The mean value of the properties of the image,
is the total number of data in the attribute, and>
is a data flag in an attribute, is asserted>
Is attribute flag, is asserted>
Is in the normative data->
Standard deviation of attribute, <' >>
Is in the normative data->
First +of attribute>
And (4) data.
S2, determining the abnormal inverter of the distributed photovoltaic by using a preset discrete rate threshold and the group string discrete rate.
In the embodiment of the invention, the AC power dispersion rate of the inverter is used for measuring the difference degree of the AC output power of the total station inverter, and the smaller the dispersion rate is, the more concentrated the output power time series curve among the inverters is, the more stable the overall operation condition of the inverter is; if the dispersion rate is large, the individual inverter has a problem and needs to search for an abnormal inverter.
In an embodiment of the present invention, the determining the abnormal inverter of the distributed photovoltaic by using the preset discrete rate threshold and the group string discrete rate includes:
performing difference processing on the group string discrete rate and a preset discrete rate threshold value to obtain a difference value between the group string discrete rate and the preset discrete rate threshold value;
when the difference value is a positive value, the inverter corresponding to the group of string dispersion ratios is an abnormal inverter of the distributed photovoltaic.
And S3, determining the electric quantity loss of the abnormal inverter according to the real-time data, and filtering the pseudo discrete rate variation of the group of discrete rates according to the electric quantity loss to obtain the true discrete rate variation of the abnormal inverter.
In an embodiment of the present invention, the determining the power loss of the abnormal inverter according to the real-time data includes:
extracting the characteristics of the real-time data to obtain the real-time characteristics of the real-time data, and determining electric quantity data in the real-time data according to the real-time characteristics;
and carrying out abnormity analysis on the electric quantity data to obtain the electric quantity loss of the abnormal inverter.
In detail, the real-time features characterize influence factors of the string dispersion rate, wherein the influence factors include but are not limited to: the device comprises component attenuation, component deviation, dust and accumulated snow, shadow shielding, vacant branches and the like, wherein the component attenuation is normal phenomenon within a certain range; the component variation is typically a production process impact; the shadow mask includes surrounding obstructions, module spacing, sun altitude effects, and the like.
Further, among the above electric quantity influence factors, normal performance and fault conditions need to be distinguished, so as to reduce workload for operation and maintenance personnel and improve operation and maintenance efficiency, such as shadow problem: the front and back of the guideline are not large, the dispersion rate changes due to the fact that the variation of the sun altitude angle and the shadow generated on the assembly by surrounding buildings or trees or the shadow generated by other objects such as dust, accumulated snow, fallen leaves or bird droppings are attached, the dispersion rate changes, in addition, the installation problems comprise that 1-2 assemblies are rarely installed in part of the assembly, and spare branches are not connected to part of inverters, and the factors cause that the data curves are obviously different from normal data on a data platform.
In an embodiment of the present invention, referring to fig. 2, the performing a pseudo discrete rate variation filtering on the group of discrete rates according to the power loss to obtain a true discrete rate variation of the abnormal inverter includes:
s21, performing vacant branch filtering on the group of string discrete rates according to the electric quantity loss to obtain a first-stage discrete rate of the abnormal inverter;
s22, filtering weather factors of the primary dispersion rate to obtain a secondary dispersion rate of the primary dispersion rate;
s23, performing building shadow filtering on the second-level discrete rate to obtain a third-level discrete rate of the second-level discrete rate, and determining the third-level discrete rate as the truth discrete rate variation of the abnormal inverter.
In detail, whether the discrete rate of a certain inverter is low all the time is judged according to the electric quantity loss, and therefore whether spare branches exist is determined; screening peak clipping conditions according to the electric quantity loss, calculating and judging according to the peak-valley difference mean value variance, and adjusting the interval so as to determine the weather condition; and judging the dispersion rate of a certain transformer according to the determined certain weather condition, and determining the building shadow according to the incidence relation between the dispersion rate and the sun position.
And S4, classifying the true discrete rate variable quantity by utilizing a support vector regression algorithm to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity, and determining the fault component of the distributed photovoltaic according to the classified discrete rate variable quantity.
In the embodiment of the invention, the goal of the support vector regression algorithm is to find a hyperplane which explicitly classifies data points in an n-dimensional space, the data points on two sides of the hyperplane, which are closest to the hyperplane, are called support vectors, and these influence the position and direction of the hyperplane, so that a standard support vector machine is facilitated to be constructed.
In an embodiment of the present invention, as shown in fig. 3, the obtaining the variation of the classified discrete rate of the variation of the veracity discrete rate by classifying the variation of the veracity discrete rate by using a support vector regression algorithm includes:
s31, constructing an initial support vector machine of the true discrete rate variation by using an initial function of a support vector regression algorithm;
s32, obtaining a training set of the initial support vector, and determining an input data matrix and a label matrix of the initial support vector machine according to the training set;
s33, constructing a Lagrange equation of the initial support vector machine, and generating equation parameters of the Lagrange equation;
s34, calculating target parameters of the initial support vector machine according to the equation parameters, generating a standard support vector machine of the true discrete rate variable quantity according to the target parameters, and classifying the true discrete rate variable quantity by using the standard support vector machine to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity.
In detail, in machine learning, a support vector machine is a supervised learning model with associated learning algorithms for analyzing data for classification and regression analysis. In support vector regression, the straight line required to fit the data is called a hyperplane, and the original space of the input data is mapped to a higher dimensional feature space, usually by a nonlinear gaussian kernel function. In the feature space, the problem becomes the construction of an optimal linear plane to fit the data, wherein the optimal linear plane refers to a functional expression in a standard support vector machine that generates the variance of the true dispersion rate according to the target parameter.
In detail, the constructing an initial support vector machine of the variance of the truth discrete rate using an initial function of a support vector regression algorithm includes:
constructing a hyperplane of the true discrete rate variation by using an initial function of a support vector regression algorithm as follows:
wherein the content of the first and second substances,
is an initial function of a support vector regression algorithm, based on the evaluation of the function in question>
Is a transposition of the first linear parameter, is present>
Is the true variance value->
Is a second linear parameter;
generating a plane norm of the hyperplane, and determining a target function and a constraint condition of the true discrete rate variation according to the plane norm;
and generating an initial support vector machine of the true discrete rate variation according to the objective function and the constraint condition.
In detail, data is not lost in the interval band, and if and only if
The loss is calculated only if the absolute value of the difference from the expected value is greater than the tolerance error, and the model can be optimized by maximizing the width of the interval band and minimizing the total loss, that is, the loss is not calculated for all samples falling into the interval band, that is, only the support vector influences the function model, wherein the tolerance error is an empirical value set manually, the support vector is the sample decisively for the final first linear parameter and the second linear parameter, and finally the optimized model is obtained by minimizing the total loss and maximizing the interval.
In detail, the second linear parameter
Is a deviation or base, the first linear parameter->
Is the minimum vector norm, is the weight; the target function is established based on the support vector, the minimum value of the sum of Euclidean distances of the samples corresponding to the support vector is obtained, and the constraint condition refers to a mathematical form which needs to be satisfied by the samples located in the spacing band.
In detail, the generating equation parameters of the lagrangian equation includes:
calculating the partial derivative of a first linear parameter of the Lagrange equation to obtain a first partial derivative function of the first linear parameter;
solving the partial derivative of a second linear parameter of the Lagrange equation to obtain a second partial derivative function of the second linear parameter;
determining a first optimization parameter and a second optimization parameter of the variability of the truth discrete rate according to the first partial derivative function, the second partial derivative function and the constraint condition of the variability of the truth discrete rate, and determining that the first optimization parameter and the second optimization parameter are equation parameters of the Lagrangian equation.
In detail, the equation with constraint condition can be changed into the equation without constraint condition according to the Lagrangian equation.
In detail, the calculating target parameters of the initial support vector machine according to the equation parameters includes:
substituting the equation parameters into an initial function of the support vector regression algorithm to obtain a function solution of the initial function;
and determining the minimum value of the objective function of the true discrete rate variation by using a preset sequence minimum optimization algorithm, and generating the objective parameter of the initial support vector machine according to the minimum value.
In detail, the preset sequence minimum optimization algorithm decomposes the large quadratic programming problem into a series of quadratic programming sub-problems as small as possible; these small quadratic programming problems are then solved analytically in the inner loop, rather than numerically optimized, thereby reducing computation time.
In detail, the determining of the faulty component of the distributed photovoltaic according to the variation of the discrete rate in the classification may specify a component a with an output label of "0" and a component b with an output label of "1".
The method comprises the steps of generating a string discrete rate through real-time data of the distributed photovoltaic, carrying out primary abnormal judgment on the distributed photovoltaic by using the string discrete rate to determine an abnormal transformer, carrying out false discrete rate variable quantity filtration on the string discrete rate according to electric quantity loss, eliminating interference of part of false factors, improving fault judgment precision, and finally carrying out classification processing on the true discrete rate variable quantity by using a support vector regression algorithm to classify possible situations of faults to obtain a detailed fault classification result, so that the precision of distributed photovoltaic fault detection is higher.
Fig. 4 is a functional block diagram of a distributed photovoltaic fault detection apparatus based on discrete rate according to an embodiment of the present invention.
The distributed photovoltaic fault detection apparatus 100 based on the discrete rate according to the present invention may be installed in an electronic device. According to the realized function, the distributed photovoltaic fault detection device 100 based on the discrete rate may include a string discrete rate module 101, an abnormal inverter module 102, a spurious filtering module 103, and a fault detection module 104. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
the string discrete rate module 101 is configured to acquire real-time data of a distributed photovoltaic system, and generate a string discrete rate of the distributed photovoltaic system according to the real-time data and a preset discrete rate algorithm, where the preset discrete rate algorithm is:
wherein the content of the first and second substances,
is the string dispersion ratio of the distributed photovoltaic, is->
Is in the normative data->
The mean value of the attributes is determined,
is the total number of data in the attribute, and>
is a data identification in an attribute, and>
is attribute flag, is asserted>
Is in normative data>
Standard deviation of an attribute, <' > based on a predetermined criterion>
Is in the normative data->
First +of attribute>
A piece of data;
the abnormal inverter module 102 is configured to determine an abnormal inverter of the distributed photovoltaic by using a preset discrete rate threshold and the group string discrete rate;
the spurious filtering module 103 is configured to determine an electric quantity loss of the abnormal inverter according to the real-time data, and perform spurious discrete rate variation filtering on the group of discrete rates according to the electric quantity loss to obtain a true discrete rate variation of the abnormal inverter;
the fault detection module 104 is configured to classify the true discrete rate variation by using a support vector regression algorithm to obtain a classified discrete rate variation of the true discrete rate variation, and determine the fault component of the distributed photovoltaic according to the classified discrete rate variation.
In the embodiments provided in the present invention, it should be understood that the disclosed method and apparatus can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, artificial Intelligence (AI) is a theory, method, technique and application device that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus may also be implemented by one unit or means through software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the same, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (10)
1. A distributed photovoltaic fault detection method based on discrete rate is characterized by comprising the following steps:
acquiring real-time data of distributed photovoltaics, and generating a string discrete rate of the distributed photovoltaics according to the real-time data and a preset discrete rate algorithm, wherein the preset discrete rate algorithm is as follows:
wherein the content of the first and second substances,
is a string dispersion rate of the distributed photovoltaic, is greater than or equal to>
Is in the normative data->
Mean value of the attribute->
Is the total number of data in the attribute, and>
is a data flag in an attribute, is asserted>
Is attribute flag, is asserted>
Is in normative data>
Standard deviation of attribute, <' >>
Is in normative data>
The th->
A piece of data;
determining the abnormal inverter of the distributed photovoltaic by using a preset discrete rate threshold value and the group string discrete rate;
determining the electric quantity loss of the abnormal inverter according to the real-time data, and filtering the pseudo discrete rate variation of the group of discrete rates according to the electric quantity loss to obtain the true discrete rate variation of the abnormal inverter;
and classifying the true discrete rate variable quantity by utilizing a support vector regression algorithm to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity, and determining the fault component of the distributed photovoltaic according to the classified discrete rate variable quantity.
2. The discrete-rate-based distributed photovoltaic fault detection method of claim 1, wherein the generating the string discrete rate of the distributed photovoltaic according to the real-time data and a preset discrete-rate algorithm comprises:
carrying out numerical value type conversion on the real-time data to obtain uniform format data of the real-time data;
carrying out standardization processing on the uniform format data to obtain standard data of the uniform format data;
and calculating the group string dispersion rate of the distributed photovoltaic by utilizing the standard data and a preset dispersion rate algorithm.
3. The discrete-rate-based distributed photovoltaic fault detection method of claim 1, wherein the determining the abnormal inverter of the distributed photovoltaic by using the preset discrete-rate threshold and the group string discrete rate comprises:
performing difference processing on the group string discrete rate and a preset discrete rate threshold value to obtain a difference value between the group string discrete rate and the preset discrete rate threshold value;
when the difference value is a positive value, the inverter corresponding to the group of string dispersion ratios is an abnormal inverter of the distributed photovoltaic.
4. The discrete-rate-based distributed photovoltaic fault detection method of claim 1, wherein the determining the loss of power to the abnormal inverter from the real-time data comprises:
extracting the characteristics of the real-time data to obtain the real-time characteristics of the real-time data, and determining electric quantity data in the real-time data according to the real-time characteristics;
and carrying out abnormity analysis on the electric quantity data to obtain the electric quantity loss of the abnormal inverter.
5. The discrete-rate-based distributed photovoltaic fault detection method of claim 1, wherein the filtering the pseudo discrete-rate variation of the string of discrete rates according to the power loss to obtain the true discrete-rate variation of the abnormal inverter comprises:
performing vacant branch filtering on the group of string discrete rates according to the electric quantity loss to obtain a first-stage discrete rate of the abnormal inverter;
performing weather factor filtration on the primary dispersion rate to obtain a secondary dispersion rate of the primary dispersion rate;
and carrying out building shadow filtering on the second-level discrete rate to obtain a third-level discrete rate of the second-level discrete rate, and determining the third-level discrete rate as the true discrete rate variation of the abnormal inverter.
6. The discrete-rate-based distributed photovoltaic fault detection method according to claim 1, wherein the classifying the variance of the truth discrete-rate by using a support vector regression algorithm to obtain the classified variance of the truth discrete-rate variance comprises:
constructing an initial support vector machine of the true discrete rate variable quantity by using an initial function of a support vector regression algorithm;
acquiring a training set of the initial support vector, and determining an input data matrix and a label matrix of the initial support vector machine according to the training set;
constructing a Lagrange equation of the initial support vector machine, and generating equation parameters of the Lagrange equation;
calculating target parameters of the initial support vector machine according to the equation parameters, generating a standard support vector machine of the true dispersion rate variation according to the target parameters, and classifying the true dispersion rate variation by using the standard support vector machine to obtain the classified dispersion rate variation of the true dispersion rate variation.
7. The discrete-rate-based distributed photovoltaic fault detection method of claim 6, wherein the constructing an initial support vector machine of the variance of the truth discrete rate using an initial function of a support vector regression algorithm comprises:
constructing a hyperplane of the true discrete rate variation by using an initial function of the support vector regression algorithm as follows:
wherein the content of the first and second substances,
is an initial function of a support vector regression algorithm, based on the evaluation of the function in question>
Is a transposition of the first linear parameter, is present>
Is the true variance value->
Is a second linear parameter;
generating a plane norm of the hyperplane, and determining a target function and a constraint condition of the true discrete rate variation according to the plane norm;
and generating an initial support vector machine of the true discrete rate variation according to the objective function and the constraint condition.
8. The discrete-rate-based distributed photovoltaic fault detection method of claim 6, wherein the generating equation parameters of the Lagrangian equation comprises:
calculating the partial derivative of a first linear parameter of the Lagrange equation to obtain a first partial derivative function of the first linear parameter;
solving the partial derivative of a second linear parameter of the Lagrange equation to obtain a second partial derivative function of the second linear parameter;
determining a first optimization parameter and a second optimization parameter of the variability of the truth discrete rate according to the first partial derivative function, the second partial derivative function and the constraint condition of the variability of the truth discrete rate, and determining that the first optimization parameter and the second optimization parameter are equation parameters of the Lagrangian equation.
9. The discrete-rate-based distributed photovoltaic fault detection method of claim 6, wherein the calculating target parameters of the initial support vector machine according to the equation parameters comprises:
substituting the equation parameters into an initial function of the support vector regression algorithm to obtain a function solution of the initial function;
and determining the minimum value of the objective function of the true discrete rate variation by using a preset sequence minimum optimization algorithm, and generating the objective parameter of the initial support vector machine according to the minimum value.
10. A distributed photovoltaic fault detection apparatus based on discrete rate, the apparatus comprising:
the string discrete rate module is used for acquiring real-time data of the distributed photovoltaic, and generating the string discrete rate of the distributed photovoltaic according to the real-time data and a preset discrete rate algorithm, wherein the preset discrete rate algorithm is as follows:
wherein the content of the first and second substances,
is the string dispersion ratio of the distributed photovoltaic, is->
Is in the normative data->
Mean value of an attribute, <' > based on>
Is the total number of data in the attribute, and>
is a data flag in an attribute, is asserted>
Is attribute flag, is asserted>
Is in the normative data->
Standard deviation of attribute, <' >>
Is in the normative data->
The th->
A piece of data;
the abnormal inverter module is used for determining an abnormal inverter of the distributed photovoltaic by utilizing a preset discrete rate threshold value and the group string discrete rate;
the spurious filtering module is used for determining the electric quantity loss of the abnormal inverter according to the real-time data, and filtering the spurious discrete rate variation of the group of the discrete rates according to the electric quantity loss to obtain the true discrete rate variation of the abnormal inverter;
and the fault detection module is used for classifying the true discrete rate variable quantity by utilizing a support vector regression algorithm to obtain the classified discrete rate variable quantity of the true discrete rate variable quantity, and determining the fault component of the distributed photovoltaic according to the classified discrete rate variable quantity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310075736.4A CN115811278B (en) | 2023-02-07 | 2023-02-07 | Distributed photovoltaic fault detection method and device based on discrete rate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310075736.4A CN115811278B (en) | 2023-02-07 | 2023-02-07 | Distributed photovoltaic fault detection method and device based on discrete rate |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115811278A CN115811278A (en) | 2023-03-17 |
| CN115811278B true CN115811278B (en) | 2023-04-18 |
Family
ID=85487579
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310075736.4A Active CN115811278B (en) | 2023-02-07 | 2023-02-07 | Distributed photovoltaic fault detection method and device based on discrete rate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115811278B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112821865A (en) * | 2020-12-30 | 2021-05-18 | 南京南瑞继保工程技术有限公司 | Rapid positioning method for low-efficiency equipment of photovoltaic power station |
| CN114139744A (en) * | 2021-11-29 | 2022-03-04 | 新奥数能科技有限公司 | Abnormal photovoltaic group string branch identification method and device, electronic equipment and storage medium |
| WO2022220746A1 (en) * | 2021-04-15 | 2022-10-20 | Envision Digital International Pte. Ltd. | Method, device, and system for monitoring photovoltaic power station |
| CN115588144A (en) * | 2022-10-25 | 2023-01-10 | 山东海量信息技术研究院 | Real-time attitude capturing method, device and equipment based on Gaussian dynamic threshold screening |
-
2023
- 2023-02-07 CN CN202310075736.4A patent/CN115811278B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112821865A (en) * | 2020-12-30 | 2021-05-18 | 南京南瑞继保工程技术有限公司 | Rapid positioning method for low-efficiency equipment of photovoltaic power station |
| WO2022220746A1 (en) * | 2021-04-15 | 2022-10-20 | Envision Digital International Pte. Ltd. | Method, device, and system for monitoring photovoltaic power station |
| CN114139744A (en) * | 2021-11-29 | 2022-03-04 | 新奥数能科技有限公司 | Abnormal photovoltaic group string branch identification method and device, electronic equipment and storage medium |
| CN115588144A (en) * | 2022-10-25 | 2023-01-10 | 山东海量信息技术研究院 | Real-time attitude capturing method, device and equipment based on Gaussian dynamic threshold screening |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115811278A (en) | 2023-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109146093B (en) | Power equipment field investigation method based on learning | |
| KR102159692B1 (en) | solar photovoltatic power generation forecasting apparatus and method based on big data analysis | |
| Kang et al. | Big data analytics in China's electric power industry: modern information, communication technologies, and millions of smart meters | |
| CN112749904B (en) | Power distribution network fault risk early warning method and system based on deep learning | |
| CN108376982B (en) | Load phase sequence identification method and device | |
| Yang et al. | Day-ahead forecasting of photovoltaic output power with similar cloud space fusion based on incomplete historical data mining | |
| CN110750524A (en) | Method and system for determining fault characteristics of active power distribution network | |
| WO2011101189A2 (en) | Energy generating system and control thereof | |
| CN111488896B (en) | Distribution line time-varying fault probability calculation method based on multi-source data mining | |
| Jia et al. | Defect prediction of relay protection systems based on LSSVM-BNDT | |
| CN114399081A (en) | Photovoltaic power generation power prediction method based on weather classification | |
| Yang et al. | Sub-minute probabilistic solar forecasting for real-time stochastic simulations | |
| CN116308883A (en) | Regional power grid data overall management system based on big data | |
| CN116307641A (en) | Digital power plant-oriented resource collaborative scheduling management method and system | |
| CN115617784A (en) | Data processing system and processing method for informationized power distribution | |
| CN115811278B (en) | Distributed photovoltaic fault detection method and device based on discrete rate | |
| CN117674119A (en) | Power grid operation risk assessment method, device, computer equipment and storage medium | |
| CN115905319B (en) | Automatic identification method and system for abnormal electricity fees of massive users | |
| CN110717623A (en) | Photovoltaic power generation power prediction method, device and equipment integrating multiple weather conditions | |
| CN115908082A (en) | Enterprise pollution discharge monitoring method and device based on electricity utilization characteristic indexes | |
| CN116151799A (en) | BP neural network-based distribution line multi-working-condition fault rate rapid assessment method | |
| Zhang et al. | Fault Prediction Method for Distribution Network Outage Based on Feature Selection and Ensemble Learning | |
| CN113610285A (en) | Power prediction method for distributed wind power | |
| CN115598459A (en) | Power failure prediction method for 10kV feeder line fault of power distribution network | |
| Moradzadeh et al. | Image processing-based data integrity attack detection in dynamic line rating forecasting applications |
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 |