CN112766702A - Distributed power station fault analysis method and system based on deep belief network - Google Patents
Distributed power station fault analysis method and system based on deep belief network Download PDFInfo
- Publication number
- CN112766702A CN112766702A CN202110043149.8A CN202110043149A CN112766702A CN 112766702 A CN112766702 A CN 112766702A CN 202110043149 A CN202110043149 A CN 202110043149A CN 112766702 A CN112766702 A CN 112766702A
- Authority
- CN
- China
- Prior art keywords
- power station
- data
- fault
- fault analysis
- operation 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses a distributed power station fault analysis method and system based on a deep belief network, which comprises the steps of collecting operation data of a master station and a substation when a photovoltaic power station fails; preprocessing the collected operation data; constructing a deep confidence network model, and performing fault waveform analysis on the preprocessed data by using the model; and generating an analysis report according to the result of the fault waveform analysis. The method has the advantages that the input data of the photovoltaic power station are preprocessed, interference factors during analysis are reduced, the accuracy of fault analysis is improved, and the power generation equipment with abnormal operation state can be timely and accurately positioned by utilizing the deep confidence network model, so that the cause of the power station fault can be accurately found.
Description
Technical Field
The invention relates to the technical field of fault analysis, in particular to a distributed power station fault analysis method and system based on a deep belief network.
Background
Through a big data mining and analyzing model and an algorithm model, processing, analyzing and managing photovoltaic power generation data in a unified manner, and discovering power generation equipment with abnormal operation state in time; and precisely positioning the sub-health root cause through multi-dimensional analysis. And carrying out statistical analysis on key service indicator data (KPI) of each station, and displaying the operation condition of each station in an visualized manner by adopting a plurality of charts. And a decision-making department makes a reasonable production plan according to the data, reduces the invalid operation condition of the equipment, provides a data basis for equipment maintenance, provides fault analysis of the photovoltaic power station, and enables an operation and maintenance team to know the overall condition of the power station in time and provide reference for expert decision-making.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The invention is provided in view of the problem of the existing photovoltaic power station fault analysis.
Therefore, the technical problem solved by the invention is as follows: when photovoltaic power generation fails, the power generation equipment with abnormal operation state cannot be timely and accurately positioned, and the cause of the power station failure is accurately found out.
In order to solve the technical problems, the invention provides the following technical scheme: collecting operation data of a master station and a substation when a photovoltaic power station fails; preprocessing the collected operation data; constructing a deep confidence network model, and performing fault waveform analysis on the preprocessed data by using the model; and generating an analysis report according to the result of the fault waveform analysis, and determining the fault position.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the operation data collection comprises the operation data of software and hardware equipment and channels of a main station system of the photovoltaic power station, the operation condition of a substation system, and the operation condition and operation fixed value of a station end relay protection device and a fault recording device.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the preprocessing of the collected operation data comprises the steps of carrying out data cleaning on the collected operation data, filling missing data and deleting repeated data, wherein the missing data filling adopts an adjacency algorithm, data filling is carried out by using information of k identical device data under the same type of fault condition, and the formula can be expressed as follows:
wherein: e is a filling value, δiUnder the normal condition of the ith equipmentOperating data of (a), thetaiIs the operation data of the ith equipment failure.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the constructing of the deep confidence network model comprises the steps that the deep confidence network model is composed of a plurality of restricted Boltzmann machines, and the restricted Boltzmann machines are trained from low to high.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the limited boltzmann machine comprises two layers which are an input layer and a hidden layer respectively, wherein the input layer inputs the preprocessed operation data, the output layer is a result of fault analysis on the operation data, and the input layer can be expressed as follows by a formula:
wherein: m is operation data, n is a fault analysis result, and p (m | n) is the probability of an input layer; the formula of its output layer can be expressed as follows:
wherein: p (n | m) is the probability of the output layer.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the training of the limited boltzmann machine comprises the steps that in the training process of the limited boltzmann machine, a probability distribution which can generate a training sample most needs to be obtained, and a decisive factor of the probability distribution is a weight value W, so that an optimal weight value needs to be obtained.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the weight values include that the weight values can be represented by a weight matrix, which can be represented as follows:
wherein: the omegai，jSetting a new sample as X ═ X (X) for the weight from the ith significant element to the jth hidden element, N is the number of significant elements, O is the number of hidden elements1,x2,…,xn) Each hidden element takes the following values:
the probability that the hidden element is in the on state is shown.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the process of training the restricted boltzmann machine further comprises the following steps of training each record A in a set, inputting the record A, calculating the probability of opening hidden layer neurons, extracting samples from the calculated probabilities for reconstruction, extracting the samples from the samples for visualization, calculating the probability of opening hidden layer neurons, and updating the weights according to the formula:
W′＝W+γ[p(n0＝1|m0)m0T-p(n1＝1|m1)m1T]
wherein: w' is the updated weight, γ is the failure probability, p (n)0＝1|m0)m0TTo initial turn-on probability, p (n)1＝1|m1)m1TIs the reconstructed probability.
The invention discloses a distributed power station fault analysis method based on a deep confidence network, which is a preferable scheme, wherein: the fault waveform analysis comprises the steps of connecting the limited Boltzmann machines in series to form the deep confidence network model, wherein a hidden layer in the upper limited Boltzmann machine is a shallow layer in the lower limited Boltzmann machine, the lower limited Boltzmann machine can be trained only by fully training the limited Boltzmann machine in the upper limited Boltzmann machine during training, real-time data of faults of the photovoltaic power station is input after the training is finished, system conversion efficiency, fault loss, power station indexes and fault rate comparison of the photovoltaic power station are analyzed through operation state data characteristics and upward and downward weights, fault analysis results are made into reports, and fault positions of the photovoltaic power station are judged through analysis conditions.
As a preferred scheme of the distributed power station fault analysis system based on the deep belief network, the system comprises: the operation data acquisition module is used for acquiring operation data of the photovoltaic power station and comprises a primary acquisition unit, a secondary acquisition unit and a tertiary acquisition unit; the data processing module is connected with the operation data acquisition module, and the operation data acquisition module transmits the acquired operation data to the data processing module in a wireless transmission mode to carry out preprocessing on the operation data; the fault analysis module is connected with the data processing module, receives data preprocessed by the data processing module by using an optical fiber and is used for carrying out fault analysis on the operation data; and the result display module is connected with the fault analysis module and used for generating and displaying the fault analysis result into a report. And the data storage module is connected with the operation data acquisition module and the fault analysis module and stores the acquired operation data and fault analysis results.
The invention has the beneficial effects that: the method has the advantages that the input data of the photovoltaic power station are preprocessed, interference factors during analysis are reduced, the accuracy of fault analysis is improved, and the power generation equipment with abnormal operation state can be timely and accurately positioned by utilizing the deep confidence network model, so that the cause of the power station fault can be accurately found.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
fig. 1 is a schematic flow chart of a distributed power station fault analysis method based on a deep belief network according to a first embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating deep confidence network model training of a distributed power station fault analysis method based on a deep confidence network according to a first embodiment of the present invention;
fig. 3 is a diagram illustrating a fault waveform analysis result of a distributed power station fault analysis method based on a deep belief network according to a first embodiment of the present invention;
fig. 4 is a schematic flow chart of a distributed power station fault analysis system based on a deep belief network according to a second embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, specific embodiments accompanied with figures are described in detail below, and it is apparent that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present invention, shall fall within the protection scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
The present invention will be described in detail with reference to the drawings, wherein the cross-sectional views illustrating the structure of the device are not enlarged partially in general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Meanwhile, in the description of the present invention, it should be noted that the terms "upper, lower, inner and outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and operate, and thus, cannot be construed as limiting the present invention. Furthermore, the terms first, second, or third are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected and connected" in the present invention are to be understood broadly, unless otherwise explicitly specified or limited, for example: can be fixedly connected, detachably connected or integrally connected; they may be mechanically, electrically, or directly connected, or indirectly connected through intervening media, or may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
Referring to fig. 1 to 3, a first embodiment of the present invention provides a distributed power station fault analysis method based on a deep belief network, including:
s1: and collecting the operation data of the main station and the substation when the photovoltaic power station fails. In which it is to be noted that,
collecting operation data comprises collecting operation data of soft and hard equipment and channels of a main station system of the photovoltaic power station, operation conditions of a substation system, and operation conditions and operation fixed values of a station end relay protection device and a fault recording device; the method comprises the working efficiency, the power generation loss, the equipment index, the number of the inverters of the transformer substation, the fault rate and the discrete rate of the string type equipment of each power station.
S2: collected operational data is pre-processed. In which it is to be noted that,
the preprocessing of the collected operation data includes that in order to improve accuracy in fault analysis, the collected operation data needs to be subjected to data cleaning, missing data is filled, and repeated data is deleted, wherein the missing data is filled by adopting an adjacency algorithm, data filling is performed by using information of k identical device data under the same fault condition, and a formula can be expressed as follows:
wherein: e is a filling value, δiFor the operation data of the i-th plant in the normal condition, θiIs the operation data of the ith equipment failure.
S3: and constructing a deep confidence network model, and performing fault waveform analysis on the preprocessed data by using the model. In which it is to be noted that,
referring to fig. 2, constructing the deep confidence network model includes that the deep confidence network model is composed of a plurality of restricted boltzmann machines, and the restricted boltzmann machines are trained from low to high.
Furthermore, the limited boltzmann machine comprises two layers which are an input layer and a hidden layer in total, wherein the input layer inputs the preprocessed operation data, the output layer is a result of fault analysis on the operation data, and the input layer can be expressed as follows by a formula:
wherein: m is operation data, n is a fault analysis result, and p (m | n) is the probability of an input layer; the formula of its output layer can be expressed as follows:
wherein: p (n | m) is the probability of the output layer;
the training of the limited Boltzmann machine comprises the steps that in the training process of the limited Boltzmann machine, probability distribution which can generate training samples most needs to be obtained, and a decisive factor of the probability distribution is a weight value W, so that the optimal weight value needs to be obtained; the weight value includes that the weight value can be represented by a weight matrix, and the weight matrix can be represented as follows:
wherein: the omegai，jSetting a new sample as X ═ X (X) for the weight from the ith significant element to the jth hidden element, N is the number of significant elements, O is the number of hidden elements1,x2,…,xn) Each hidden element takes the following values:
the probability that the hidden element is in an open state is represented;
furthermore, the process of training the restricted boltzmann machine further includes, in the training process, inputting each record a in the training set, calculating the probability of the hidden layer neuron being turned on, extracting a sample from the calculated probability for reconstruction, extracting a sample from the sample, calculating the probability of the hidden layer neuron being turned on, and updating the weight according to the following formula:
W′＝W+γ[p(n0＝1|m0)m0T-p(n1＝1|m1)m1T]
wherein: w' is the updated weight, γ is the failure probability, p (n)0＝1|m0)m0TTo initial turn-on probability, p (n)1＝1|m1)m1TIs the reconstructed probability, wherein:
wherein the superscript is used for distinguishing different vectors, the subscript is used for distinguishing different dimensions of the same vector, and samples are extracted from the calculated sample probability:
n0＝P(n0|m0)
by n0Sample reconstruction was performed and a sample of the visualization was also taken:
m1＝P(m1|n0)
and (3) calculating the probability of the hidden layer neuron by using the reconstructed apparent layer neuron:
the fault waveform analysis comprises connecting the limited Boltzmann machines in series to form a deep confidence network model, wherein the hidden layer in the upper limited Boltzmann machine is the shallow layer in the next limited Boltzmann machine, during training, the lower layer can be trained only by fully training the limited Boltzmann machine in the upper layer, real-time data of faults of the photovoltaic power station is input after the training is finished, and through the data characteristics of the running state and the upward and downward weights, analyzing the system conversion efficiency, fault loss, power station index and fault rate comparison of the photovoltaic power station, storing the fault analysis result, inputting the result into a database to judge the fault position, the database is composed of historical analysis results and expert fault analysis and judgment theories, has high instantaneity and professionality, and analyzes fault positions with high accuracy by using the continuously updated historical analysis results and expert theoretical knowledge.
S4: and generating an analysis report according to the result of the fault waveform analysis, and determining the fault position.
The fault analysis result and the fault position are generated into an analysis report and displayed through the displayer, the displayer is connected with the fault analysis result in a wireless communication mode and can be transmitted to mobile equipment of workers, the workers can check the power station condition at any time, and the fault problem is timely processed.
In order to better verify and explain the technical effects adopted in the method, an inverter is selected for testing in the embodiment, and the real effect of the method is verified by a scientific demonstration means;
selecting operation state data of an inverter in a transformer substation, wherein basic operation parameters of the transformer are as follows: the power is 100kw, the current is 60.0A, the power grid frequency is 60.0HZ, the fault evaluation of the inverter is carried out by using the method, real-time running state data is input into a trained deep confidence network model for fault waveform analysis, the obtained analysis result refers to fig. 3, the one-day power and efficiency curve of the power converter can be seen, the condition that the inverter is overloaded at 10:20:00, the power grid frequency is too low, and the direct-current voltage is too high at 10:50:00 is obtained.
The traditional SPC analysis and assessment system is selected to be compared with the fault analysis method, the same photovoltaic power station is analyzed for 100 times by using the two methods, and 15 power station faults occur in the 100 times of analysis, wherein the traditional SPC analysis and assessment system is used for analyzing the operation condition of the photovoltaic power station by collecting two-dimensional and three-dimensional operation parameters of the power station, MATLB software is used for testing the fault analysis conditions of the two methods, and the obtained results are shown in the following table 1:
table 1: and comparing the fault analysis results.
| Number of failure reports | Fault location reporting accuracy | |
| The method of the invention | 14 | 86.7％ |
| SPC analysis evaluation System | 11 | 63.1％ |
As can be seen from table 1, the situation and the specific fault location of the photovoltaic power station at the time of the fault cannot be accurately analyzed by using the conventional SPC analysis and evaluation system, and the fault reporting times and the positioning accuracy of the invention are high, so that the invention has practicability.
Example 2
Referring to fig. 4, a second embodiment of the present invention, which is different from the first embodiment, provides a distributed power station fault analysis system based on a deep belief network, including: the system comprises an operation data acquisition module 100, a data processing module 200, a fault analysis module 300, a result display module 400 and a data storage module 500.
The operation data acquisition module 100 is used for acquiring operation data of a photovoltaic power station and comprises a primary acquisition unit 101, a secondary acquisition unit 102 and a tertiary acquisition unit 103; the data processing module 200 is connected with the operation data acquisition module 100, and the operation data acquisition module 100 transmits the acquired operation data to the data processing module 200 in a wireless transmission mode to carry out the preprocessing of the operation data; the fault analysis module 300 is connected with the data processing module 200, and receives data preprocessed by the data processing module 200 by using an optical fiber, so as to perform fault analysis on the operating data; the result display module 400 is connected to the fault analysis module 300, and is configured to generate and display a fault analysis result into a report; the data storage module 500 is connected to the operation data acquisition module 100 and the fault analysis module 300, and stores the acquired operation data and the fault analysis result.
Further, the primary collection unit 101, the secondary collection unit 102 and the tertiary collection unit 103 respectively collect operation conditions of a master station and a substation of the photovoltaic power station and communication conditions among the stations, and the fault analysis module 300 analyzes the working efficiency, the origin loss, the functional index and the dispersion rate of the string-type equipment of each power station in the power station, analyzes the stable operation condition of the power station equipment and performs positioning analysis on the fault equipment.
It should be understood that the system provided in the present embodiment, which relates to the connection relationship of the operation data acquisition module 100, the data processing module 200, the fault analysis module 300, the result display module 400 and the data storage module 500, may be, for example, a computer readable program, and is implemented by improving the program data interface of each module.
It should be recognized that embodiments of the present invention can be realized and implemented by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer readable memory. The methods may be implemented in a computer program using standard programming techniques, including a non-transitory computer-readable storage medium configured with the computer program, where the storage medium so configured causes a computer to operate in a specific and predefined manner, according to the methods and figures described in the detailed description. Each program may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language. Furthermore, the program can be run on a programmed application specific integrated circuit for this purpose.
Further, the operations of processes described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The processes described herein (or variations and/or combinations thereof) may be performed under the control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) collectively executed on one or more processors, by hardware, or combinations thereof. The computer program includes a plurality of instructions executable by one or more processors.
Further, the method may be implemented in any type of computing platform operatively connected to a suitable interface, including but not limited to a personal computer, mini computer, mainframe, workstation, networked or distributed computing environment, separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, and the like. Aspects of the invention may be embodied in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optically read and/or write storage medium, RAM, ROM, or the like, such that it may be read by a programmable computer, which when read by the storage medium or device, is operative to configure and operate the computer to perform the procedures described herein. Further, the machine-readable code, or portions thereof, may be transmitted over a wired or wireless network. The invention described herein includes these and other different types of non-transitory computer-readable storage media when such media include instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor. The invention also includes the computer itself when programmed according to the methods and techniques described herein. A computer program can be applied to input data to perform the functions described herein to transform the input data to generate output data that is stored to non-volatile memory. The output information may also be applied to one or more output devices, such as a display. In a preferred embodiment of the invention, the transformed data represents physical and tangible objects, including particular visual depictions of physical and tangible objects produced on a display.
As used in this application, the terms "component," "module," "system," and the like are intended to refer to a computer-related entity, either hardware, firmware, a combination of hardware and software, or software in execution. For example, a component may be, but is not limited to being: a process running on a processor, an object, an executable, a thread of execution, a program, and/or a computer. By way of example, both an application running on a computing device and the computing device can be a component. One or more components can reside within a process and/or thread of execution and a component can be localized on one computer and/or distributed between two or more computers. In addition, these components can execute from various computer readable media having various data structures thereon. The components may communicate by way of local and/or remote processes such as in accordance with a signal having one or more data packets (e.g., data from one component interacting with another component in a local system, distributed system, and/or across a network such as the internet with other systems by way of the signal).
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.
Claims (10)
1. A distributed power station fault analysis method based on a deep belief network is characterized in that: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
collecting operation data of a master station and a substation when a photovoltaic power station fails;
preprocessing the collected operation data;
constructing a depth confidence network model by using a restricted Boehmann machine, and performing fault waveform analysis on the preprocessed data by using the model;
and generating an analysis report according to the result of the fault waveform analysis, and determining the fault position.
2. The distributed power station fault analysis method based on the deep belief network of claim 1, characterized by: the collecting of the operational data may include,
and collecting the running data of the soft and hard equipment and the channel of the photovoltaic power station main station system, the running condition of the substation system, and the running conditions and running fixed values of the station end relay protection device and the fault recording device.
3. The distributed power station fault analysis method based on the deep belief network of claim 2, characterized by: the pre-processing of the collected operational data includes,
performing data cleaning on the collected operation data, filling missing data and deleting repeated data, wherein the missing data is filled by adopting an adjacency algorithm, and the data is filled by using information of k identical device data under the same type of fault condition, and the formula can be expressed as:
wherein: e is a filling value, δiFor the operation data of the i-th plant in the normal condition, θiIs the operation data of the ith equipment failure.
4. The distributed power station fault analysis method based on the deep belief network as claimed in any of claims 1 to 3, characterized in that: the building of the deep belief network model includes,
the deep confidence network model consists of a plurality of limited Boltzmann machines, and the limited Boltzmann machines are trained from low to high.
5. The distributed power station fault analysis method based on the deep belief network of claim 4 characterized in that: the limited boltzmann machine includes,
the limited boltzmann machine has two layers, namely an input layer and a hidden layer, wherein the input layer inputs the preprocessed operation data, the output layer is a result of fault analysis on the operation data, and the input layer can be expressed as follows by a formula:
wherein: m is operation data, n is a fault analysis result, and p (m | n) is the probability of an input layer; the formula of its output layer can be expressed as follows:
wherein: p (n | m) is the probability of the output layer.
6. The distributed power station fault analysis method based on the deep belief network of claim 5, characterized by: the training of the restricted boltzmann machine includes,
in the process of training the limited boltzmann machine, a probability distribution which can generate a training sample most needs to be obtained, and a decisive factor of the probability distribution is a weight value W, so that an optimal weight value needs to be obtained.
7. The distributed power station fault analysis method based on the deep belief network of claim 6, characterized by: the weight values include, for example,
the weight values may be represented by a weight matrix, which may be represented as follows:
wherein: the omegai，jSetting a new sample as X ═ X (X) for the weight from the ith significant element to the jth hidden element, N is the number of significant elements, O is the number of hidden elements1,x2,…,xn) Each hidden element takes the following values:
the probability that the hidden element is in the on state is shown.
8. The distributed power station fault analysis method based on the deep belief network of claim 6, characterized by: the process of training the restricted boltzmann machine further comprises,
in the training process, each record A in a training set is input, the started probability of hidden layer neurons is calculated, samples are extracted from the calculated probability for reconstruction, the apparent layer samples are extracted again, the started probability of the hidden layer neurons is calculated, and therefore weight updating is conducted, and the updating formula is as follows:
W′＝W+γ[p(n0＝1|m0)m0T-p(n1＝1|m1)m1T]
wherein: w' is the updated weight, γ is the failure probability, p (n)0＝1|m0)m0TTo initial turn-on probability, p (n)1＝1|m1)m1TIs the reconstructed probability.
9. The distributed power station fault analysis method based on the deep belief network as claimed in any of claims 5 to 8, characterized by: the performing of the fault waveform analysis may include,
and connecting the limited Boltzmann machines in series to form the deep confidence network model, wherein a hidden layer in the upper limited Boltzmann machine is a shallow layer in the next limited Boltzmann machine, the lower limited Boltzmann machine can be trained only by fully training the limited Boltzmann machine in the upper limited Boltzmann machine during training, real-time data of faults of the photovoltaic power station is input after the training is finished, the system conversion efficiency, the fault loss, the power station index and the fault rate of the photovoltaic power station are compared and analyzed through the running state data characteristics and the upward and downward weights, the fault analysis result is made into a report, and the fault position of the photovoltaic power station is judged through the analysis condition.
10. A distributed power station fault analysis system based on a deep belief network is characterized by comprising,
the operation data acquisition module (100) is used for acquiring operation data of a photovoltaic power station and comprises a primary acquisition unit (101), a secondary acquisition unit (102) and a tertiary acquisition unit (103);
the data processing module (200) is connected with the operation data acquisition module (100), and the operation data acquisition module (100) transmits acquired operation data to the data processing module (200) in a wireless transmission mode to carry out preprocessing on the operation data;
the fault analysis module (300) is connected with the data processing module (200), receives data preprocessed by the data processing module (200) by using an optical fiber and is used for carrying out fault analysis on the operation data;
and the result display module (400) is connected to the fault analysis module (300) and is used for generating and displaying a report of the fault analysis result.
The data storage module (500) is connected to the operation data acquisition module (100) and the fault analysis module (300) and stores the acquired operation data and fault analysis results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110043149.8A CN112766702A (en) | 2021-01-13 | 2021-01-13 | Distributed power station fault analysis method and system based on deep belief network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110043149.8A CN112766702A (en) | 2021-01-13 | 2021-01-13 | Distributed power station fault analysis method and system based on deep belief network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112766702A true CN112766702A (en) | 2021-05-07 |
Family
ID=75700137
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110043149.8A Pending CN112766702A (en) | 2021-01-13 | 2021-01-13 | Distributed power station fault analysis method and system based on deep belief network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112766702A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107991878A (en) * | 2017-12-20 | 2018-05-04 | 湘潭大学 | A kind of chaos tracing control method based on depth belief network |
| CN108089099A (en) * | 2017-12-18 | 2018-05-29 | 广东电网有限责任公司佛山供电局 | The diagnostic method of distribution network failure based on depth confidence network |
| CN108732528A (en) * | 2018-05-28 | 2018-11-02 | 国网福建省电力有限公司电力科学研究院 | A kind of digitalized electrical energy meter method for diagnosing faults based on depth confidence network |
| CN109358230A (en) * | 2018-10-29 | 2019-02-19 | 国网甘肃省电力公司电力科学研究院 | A kind of micro-capacitance sensor is fallen into a trap and the Intelligent electric-energy metering method of m-Acetyl chlorophosphonazo |
| CN109596942A (en) * | 2018-12-06 | 2019-04-09 | 华北电力大学 | A kind of voltage sag reason recognition methods based on depth confidence network |
| CN109615109A (en) * | 2018-12-29 | 2019-04-12 | 大唐河南清洁能源有限责任公司 | Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database |
| CN109765559A (en) * | 2019-02-02 | 2019-05-17 | 中国水利水电科学研究院 | A kind of mountain area disastrous rainstorm recognition methods based on Doppler radar and deep learning |
| CN110763997A (en) * | 2019-11-04 | 2020-02-07 | 华北电力大学（保定） | Early fault early warning method for synchronous motor stator |
| CN110874601A (en) * | 2018-08-29 | 2020-03-10 | 阿里巴巴集团控股有限公司 | Method for identifying running state of equipment, and state identification model training method and device |
| CN111259834A (en) * | 2020-01-20 | 2020-06-09 | 武汉大学 | Transformer fault diagnosis method and system based on integrated deep belief network |
| CN111553297A (en) * | 2020-05-06 | 2020-08-18 | 东华大学 | Method and system for diagnosing production fault of polyester filament based on 2D-CNN and DBN |
| CN112116029A (en) * | 2020-09-25 | 2020-12-22 | 天津工业大学 | Intelligent fault diagnosis method for gearbox with multi-scale structure and characteristic fusion |
-
2021
- 2021-01-13 CN CN202110043149.8A patent/CN112766702A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108089099A (en) * | 2017-12-18 | 2018-05-29 | 广东电网有限责任公司佛山供电局 | The diagnostic method of distribution network failure based on depth confidence network |
| CN107991878A (en) * | 2017-12-20 | 2018-05-04 | 湘潭大学 | A kind of chaos tracing control method based on depth belief network |
| CN108732528A (en) * | 2018-05-28 | 2018-11-02 | 国网福建省电力有限公司电力科学研究院 | A kind of digitalized electrical energy meter method for diagnosing faults based on depth confidence network |
| CN110874601A (en) * | 2018-08-29 | 2020-03-10 | 阿里巴巴集团控股有限公司 | Method for identifying running state of equipment, and state identification model training method and device |
| CN109358230A (en) * | 2018-10-29 | 2019-02-19 | 国网甘肃省电力公司电力科学研究院 | A kind of micro-capacitance sensor is fallen into a trap and the Intelligent electric-energy metering method of m-Acetyl chlorophosphonazo |
| CN109596942A (en) * | 2018-12-06 | 2019-04-09 | 华北电力大学 | A kind of voltage sag reason recognition methods based on depth confidence network |
| CN109615109A (en) * | 2018-12-29 | 2019-04-12 | 大唐河南清洁能源有限责任公司 | Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database |
| CN109765559A (en) * | 2019-02-02 | 2019-05-17 | 中国水利水电科学研究院 | A kind of mountain area disastrous rainstorm recognition methods based on Doppler radar and deep learning |
| CN110763997A (en) * | 2019-11-04 | 2020-02-07 | 华北电力大学（保定） | Early fault early warning method for synchronous motor stator |
| CN111259834A (en) * | 2020-01-20 | 2020-06-09 | 武汉大学 | Transformer fault diagnosis method and system based on integrated deep belief network |
| CN111553297A (en) * | 2020-05-06 | 2020-08-18 | 东华大学 | Method and system for diagnosing production fault of polyester filament based on 2D-CNN and DBN |
| CN112116029A (en) * | 2020-09-25 | 2020-12-22 | 天津工业大学 | Intelligent fault diagnosis method for gearbox with multi-scale structure and characteristic fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112382064B (en) | Power Internet of things fault early warning method and system based on digital twin technology | |
| CN108830745B (en) | Power grid cascading failure diagnosis, early warning and evaluation system based on monitoring information | |
| CN112859822B (en) | Equipment health analysis and fault diagnosis method and system based on artificial intelligence | |
| CN112803592B (en) | Intelligent fault early warning method and system suitable for distributed power station | |
| CN108008332A (en) | A kind of new energy Remote testing device method for diagnosing faults based on data mining | |
| CN104318482A (en) | Comprehensive assessment system and method of smart distribution network | |
| CN104700321A (en) | Analytical method of state running tendency of transmission and distribution equipment | |
| CN109451532A (en) | A kind of check method and device of base station location | |
| CN105956788A (en) | Dynamic management control method for cost of power transmission and transformation project | |
| CN103729804A (en) | On-line decision support method for responding to power quality early warning | |
| CN107784392A (en) | A kind of the defects of transmission line of electricity based on machine learning Forecasting Methodology and device | |
| CN107784393A (en) | A kind of the defects of transmission line of electricity Forecasting Methodology and device | |
| CN104281525A (en) | Defect data analytical method and method for shortening software testing programs by using same | |
| CN110865924A (en) | Health degree diagnosis method and health diagnosis framework for internal server of power information system | |
| CN116865258A (en) | Hierarchical distributed power supply intelligent power grid construction method | |
| CN115438726A (en) | Device life and fault type prediction method and system based on digital twin technology | |
| CN114035468A (en) | Predictive monitoring method and system for fan overhaul process based on XGboost algorithm | |
| CN109345121A (en) | A kind of method and apparatus for Operation Event of Nuclear Power Plants analysis | |
| CN115222284A (en) | Digital twin system maturity evaluation method based on maturity model | |
| CN114510869A (en) | Principal component analysis method and photovoltaic equipment power generation amount loss prediction method of Resnet network | |
| CN117172556B (en) | Construction risk early warning method and system for bridge engineering | |
| CN110033102A (en) | A kind of huge hydroelectric power plant has the intelligent diagnosing method and expert system of learning functionality | |
| CN116821610B (en) | Method for optimizing wind power generation efficiency by utilizing big data | |
| CN114091213A (en) | Power communication network planning scheme evaluation system | |
| CN112766702A (en) | Distributed power station fault analysis method and system based on deep belief network |
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 |