CN113888353A - Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment - Google Patents

Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment Download PDF

Info

Publication number
CN113888353A
CN113888353A CN202111153747.7A CN202111153747A CN113888353A CN 113888353 A CN113888353 A CN 113888353A CN 202111153747 A CN202111153747 A CN 202111153747A CN 113888353 A CN113888353 A CN 113888353A
Authority
CN
China
Prior art keywords
power generation
photovoltaic power
fault
data
energy efficiency
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
Application number
CN202111153747.7A
Other languages
Chinese (zh)
Inventor
朱俊杰
吕亮
吴昊
祝金涛
武青
王海明
钱瑜波
李壮
方宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huaneng Clean Energy Research Institute
Clean Energy Branch of Huaneng Zhejiang Energy Development Co Ltd
Original Assignee
Huaneng Clean Energy Research Institute
Clean Energy Branch of Huaneng Zhejiang Energy Development Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huaneng Clean Energy Research Institute, Clean Energy Branch of Huaneng Zhejiang Energy Development Co Ltd filed Critical Huaneng Clean Energy Research Institute
Priority to CN202111153747.7A priority Critical patent/CN113888353A/en
Publication of CN113888353A publication Critical patent/CN113888353A/en
Pending legal-status Critical Current

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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/23—Updating
    • 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
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • 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
    • 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
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
    • 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
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Abstract

The application provides an energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment, wherein the method comprises the following steps: acquiring energy efficiency fault knowledge of distributed photovoltaic power generation equipment, and generating an ontology energy efficiency diagnosis knowledge base; acquiring historical operating data of each device, and determining a reference interval of detection data of each device under different working conditions; acquiring real-time data of each device, and performing anomaly detection on the real-time data based on the corresponding reference interval; if the abnormal data are detected, determining signs of the abnormal data, performing reasoning diagnosis through the corresponding probability relation between the signs and the fault modes preset in the body energy efficiency diagnosis knowledge base, and determining the abnormal modes and the faults of the distributed photovoltaic power generation equipment; and sending the diagnosis result and the maintenance strategy to a client of the operation and maintenance personnel. The method can quickly locate the failed distributed photovoltaic power generation equipment and specific faults, provide corresponding maintenance strategies, reduce the operation and maintenance cost and improve the operation and maintenance efficiency.

Description

Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment
Technical Field
The application relates to the technical field of photovoltaic power generation, in particular to an energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment.
Background
At present, power generation is generally carried out in areas with rich illumination resources through a distributed photovoltaic power generation technology, and grid connection or self-use are achieved through distributed photovoltaic power generation. Distributed photovoltaic power plants of this type are generally divided into areas for unified management, and are therefore distinguished from centralized photovoltaic power plants. The power generation components of the centralized photovoltaic power station are intensively arranged in a plane area and are generally open spaces without shelters, so that the environmental conditions of the power generation devices in the area are similar, and the environmental conditions of the power generation devices in the distributed photovoltaic power station area are different greatly, for example, the environmental conditions of mountain top power generation equipment and mountain foot power generation equipment in the area photovoltaic power station in the southwest mountain area are completely different. Therefore, the complexity of operation and maintenance work for such distributed photovoltaic power stations is high, however, the distributed photovoltaic power stations usually need to control the labor cost, and the development of an intelligent operation and maintenance system reduces the operation and maintenance cost.
In the related art, the monitoring and diagnosing method for the abnormal state of the centralized photovoltaic power station mainly comprises the following 3 steps: the 1 st method is to detect the fault by using sensors to detect the current, voltage or electric characteristics of the array and infrared characteristics of the photovoltaic string. The 2 nd method focuses on analyzing the power loss of the power station, and for example, a supervision and fault detection method is proposed based on the power loss analysis, which identifies a fault by analyzing the loss existing in the system and comparing the error deviation amounts of the direct current and the voltage. The 3 rd method determines a fault by comparing real-time operational data with model outputs obtained by historical data modeling.
However, the applicant finds that, because distributed photovoltaic power station power generation equipment is dispersed and the construction cost is relatively low, when the centralized photovoltaic power station abnormal state monitoring and diagnosing method is applied to the distributed photovoltaic power station power generation equipment, the former two methods lack suitable monitoring instruments and cannot be applied to the technical requirements of fault detection methods of distributed photovoltaic power generation with huge environmental condition differences, and the third intelligent method also lacks training samples, so that the photovoltaic power station abnormal state monitoring and diagnosing method in the prior art, for example, the most extensive early warning method of transverse comparison among a series of equipment such as detection of abnormal dispersion rate of combiner box group strings and abnormal inverter efficiency, is not suitable for operation and maintenance of the distributed photovoltaic power station.
Disclosure of Invention
The present application is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, a first objective of the present application is to provide an energy efficiency diagnosis method for distributed photovoltaic power generation equipment, which can quickly locate faulty generator set information and corresponding specific faults, provide corresponding health maintenance strategies, support operation and maintenance personnel to locate abnormal photovoltaic power generation equipment through a mobile end tool, query a generator set fault mode and maintenance measures, grasp operation conditions of all generator sets in a distributed power station area, save manpower, reduce equipment and data costs required for operation and maintenance, and improve operation and maintenance efficiency.
A second objective of the present application is to provide an energy efficiency diagnosis system for a distributed photovoltaic power generation device;
a third object of the present application is to propose a non-transitory computer-readable storage medium.
To achieve the above object, a first aspect of the present application is directed to an energy efficiency diagnosis method for a distributed photovoltaic power generation apparatus, the method including:
analyzing distributed photovoltaic power generation equipment to be diagnosed through failure modes, impact analysis FMEA and fault tree analysis FTA, acquiring energy efficiency fault knowledge of the distributed photovoltaic power generation equipment, and generating a body energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment;
acquiring historical operation data of each photovoltaic power generation device, and determining a reference interval of detection data of each photovoltaic power generation device under different working conditions according to the historical operation data;
acquiring real-time operation data of each photovoltaic power generation device, and performing anomaly detection on the real-time operation data based on the corresponding reference interval;
if the real-time operation data of any one photovoltaic power generation device is abnormal data, determining signs of the abnormal data, performing inference diagnosis on any one distributed photovoltaic power generation device through a corresponding probability relation between the signs and a fault mode preset in the body energy efficiency diagnosis knowledge base, and determining the abnormal mode and the fault mode of any one distributed photovoltaic power generation device;
and sending the diagnosis result and the maintenance strategy corresponding to the diagnosis result stored in the body energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance worker.
Optionally, in an embodiment of the present application, generating the ontology energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation device includes: converting the energy efficiency fault knowledge into a structured language recognizable by a computer; modeling through a preset open source tool based on the structured language to generate an energy efficiency diagnosis knowledge base of the distributed photovoltaic generator set based on the ontology; and acquiring historical fault data of each distributed photovoltaic power generation device, and performing prior probability assignment on fault modes corresponding to different symptoms based on Bayesian theorem to generate a corresponding probability relation between the preset symptoms and the fault modes.
Optionally, in an embodiment of the present application, determining a reference interval of detection data of each photovoltaic power generation device under different working conditions according to the historical operating data includes performing steady-state screening on the historical operating data, and screening out non-steady-state data in the historical operating data; the irradiance and the temperature are used as boundary conditions, and the working condition of each photovoltaic power generation device is divided through a K-mean clustering algorithm; removing fault data under each working condition through a multivariate Gaussian mixture model to generate a normal data training sample; and training a long-short term memory artificial neural network (LSTM) through the normal data training sample, and determining a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition according to a predicted value output by the long-short term memory artificial neural network (LSTM).
Optionally, in an embodiment of the present application, determining a reference interval corresponding to a boundary condition of each photovoltaic power generation device under each working condition according to a predicted value of a parameter output by the long-short term memory artificial neural network LSTM, where the step of determining the reference interval includes, for each parameter, obtaining multiple predicted values of the parameter output by the long-short term memory artificial neural network LSTM and corresponding actual values; calculating the difference value between each predicted parameter value and the actual value, and taking the predicted parameter value with the largest difference value with the actual value as a predicted target parameter value; and constructing the reference interval by taking the target parameter predicted value as a center and taking two times of the difference value between the target parameter predicted value and the actual value as an interval length.
Optionally, in an embodiment of the present application, after sending the diagnosis result and the maintenance policy corresponding to the diagnosis result and stored in the ontology energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance person, the method further includes: adding fault records on the mobile client of the operation and maintenance personnel according to the generated fault modes; and updating the posterior probability of the corresponding probability relation between the preset symptom and the fault mode according to the acquired fault record.
Optionally, in an embodiment of the present application, adding a fault record on the mobile client of the operation and maintenance staff includes receiving the fault mode, the cause of the fault, the symptom of the fault and the corresponding maintenance measures sent by the mobile client; the updating of the posterior probability of the corresponding probability relationship between the preset symptom and the fault mode according to the acquired fault record includes: judging whether the fault mode exists, and if the fault mode exists, updating the posterior probability; and if the fault mode does not exist, adding the fault mode to the ontology energy efficiency diagnosis knowledge base.
In order to achieve the above object, a second aspect of the present application provides an energy efficiency diagnosis system for a distributed photovoltaic power generation apparatus, including the following modules:
the generating module is used for analyzing the distributed photovoltaic power generation equipment to be diagnosed through failure mode and influence analysis FMEA and fault tree analysis FTA, obtaining energy efficiency fault knowledge of the distributed photovoltaic power generation equipment and generating a body energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment;
the determining module is used for acquiring historical operating data of each photovoltaic power generation device and determining a reference interval of detection data of each photovoltaic power generation device under different working conditions according to the historical operating data;
the abnormity detection module is used for acquiring real-time operation data of each photovoltaic power generation device and carrying out abnormity detection on the real-time operation data based on the corresponding reference interval;
the reasoning diagnosis module is used for determining the sign of the abnormal data if the real-time operation data of any photovoltaic power generation equipment is abnormal data, carrying out reasoning diagnosis on any distributed photovoltaic power generation equipment through the corresponding probability relation between the sign and the fault mode preset in the body energy efficiency diagnosis knowledge base, and determining the abnormal mode and the fault mode of any distributed photovoltaic power generation equipment;
and the sending module is used for sending the diagnosis result and the maintenance strategy corresponding to the diagnosis result stored in the body energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance worker.
Optionally, in an embodiment of the present application, the generating module is specifically configured to: converting the energy efficiency fault knowledge into a structured language recognizable by a computer; modeling through a preset open source tool based on the structured language to generate an energy efficiency diagnosis knowledge base of the distributed photovoltaic generator set based on the ontology; and acquiring historical fault data of each distributed photovoltaic power generation device, and performing prior probability assignment on fault modes corresponding to different symptoms based on Bayesian theorem to generate a corresponding probability relation between the preset symptoms and the fault modes.
Optionally, in an embodiment of the present application, the determining module is specifically configured to: performing steady-state screening on the historical operating data, and screening out unsteady-state data in the historical operating data; the irradiance and the temperature are used as boundary conditions, and the working condition of each photovoltaic power generation device is divided through a K-mean clustering algorithm; removing fault data under each working condition through a multivariate Gaussian mixture model to generate a normal data training sample; and training a long-short term memory artificial neural network (LSTM) through the normal data training sample, and determining a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition according to a predicted value output by the long-short term memory artificial neural network (LSTM).
The technical scheme provided by the embodiment of the application at least has the following beneficial effects: the method comprises the steps of forming fault knowledge of abnormal energy efficiency level of the photovoltaic generator set by an FMEA (failure mode analysis and data analysis) and expressing relevant fault knowledge of the unstructured photovoltaic generator set by a structured language which can be identified by a computer based on a body concept to form an open body knowledge base, carrying out energy efficiency diagnosis and health maintenance on real-time data by a distributed photovoltaic generator set energy efficiency diagnosis and health maintenance process of 'abnormity detection-determination symptom-diagnosis inference-positioning fault-health maintenance decision-mobile end user reading', carrying out abnormity detection according to a reference interval determined by the self environmental characteristics of each device, inputting corresponding symptoms of abnormal information into the energy efficiency diagnosis knowledge base, carrying out inference diagnosis work based on a Bayesian model, and thus quickly locating the faulted generator set information and corresponding specific faults, and provide a corresponding health maintenance policy. And moreover, through a mobile client interface, operation and maintenance personnel can access related unit information through mobile terminal equipment according to the number of the generator set, inquire a unit fault mode and maintenance measures, master the operation conditions of all the generator sets in the distributed power station area, make a maintenance plan in advance, reduce the operation and maintenance cost and improve the operation and maintenance efficiency.
In order to implement the foregoing embodiments, an embodiment of the third aspect of the present application further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the energy efficiency diagnosis method for the distributed photovoltaic power generation apparatus in the foregoing embodiments.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of an energy efficiency diagnosis method for a distributed photovoltaic power generation apparatus according to an embodiment of the present application;
fig. 2 is a flowchart of a method for determining a reference interval according to an embodiment of the present application;
fig. 3 is a schematic flowchart of an energy efficiency diagnosis method for a specific distributed photovoltaic power generation apparatus according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of an energy efficiency diagnosis system of a distributed photovoltaic power generation device according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
It should be noted that, in the operation and maintenance work in the later stage of the distributed photovoltaic system, the intelligent operation and maintenance system in the related art only stops making an early warning when positioning an abnormality of one or more parameters, and reminds the operation and maintenance personnel to perform maintenance. The intelligent operation and maintenance system cannot meet the actual requirements of distributed photovoltaic operation and maintenance, and therefore the energy efficiency diagnosis method capable of intelligently locating to a specific fault mode and providing a corresponding maintenance strategy for operation and maintenance personnel is provided.
The method comprises the steps of generating power by using a photovoltaic power station, and generating power by using a photovoltaic power station. According to the method and the device, respective operation parameter reference intervals are determined according to differences of generator sets in the distributed photovoltaic power station region, and respective energy efficiency diagnosis knowledge bases are established. In addition, the current new energy intelligent operation and maintenance system stops state monitoring and fault early warning, and the method for fault diagnosis and health maintenance is further provided. The traditional fault diagnosis method is generally divided into several types based on data, knowledge, signals, models and the like, and the method combines the data and the knowledge according to the particularity of the distributed photovoltaic generator set to complete the real-time diagnosis process of 'abnormity detection-symptom determination-diagnosis inference-fault positioning-health maintenance decision' of the distributed photovoltaic generator set.
The following describes an energy efficiency diagnosis method and system for a distributed photovoltaic power generation device, which are provided by the embodiments of the present invention, with reference to the accompanying drawings.
Fig. 1 is a flowchart of an energy efficiency diagnosis method for a distributed photovoltaic power generation apparatus according to an embodiment of the present application, and as shown in fig. 1, the method includes the following steps:
step 101, analyzing distributed photovoltaic power generation equipment to be diagnosed through failure mode and impact analysis FMEA and fault tree analysis FTA, obtaining energy efficiency fault knowledge of the distributed photovoltaic power generation equipment, and generating a body energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment.
The distributed photovoltaic power generation equipment to be diagnosed can be each generator set in the distributed photovoltaic power station, can also be each device in the generator set, and can be specifically set according to the operation and maintenance requirements. Failure Mode and impact Analysis (FMEA for short) and Fault Tree Analysis (FTA for short) are two methods for analyzing a complex system to find out potential failures.
In the embodiment of the application, mechanism analysis and equipment analysis are performed on the distributed photovoltaic generator set through an FMEA (failure mode analysis) and FTA (fiber to the array) analysis method, and common related fault knowledge causing abnormal energy efficiency level of photovoltaic power generation equipment is combed out, wherein the related fault knowledge is the energy efficiency fault knowledge of the distributed photovoltaic power generation equipment. Because the FMEA analysis method and the FTA analysis method are complementary in analysis mode, the energy efficiency fault knowledge of the distributed photovoltaic power generation equipment can be analyzed more fully and comprehensively through the two methods.
Further, since the energy efficiency fault knowledge of the generator set obtained by the analysis method is unstructured, in order to facilitate subsequent diagnosis, in the embodiment of the application, an ontology energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment is generated according to the obtained energy efficiency fault knowledge based on an ontology concept.
The energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment is a term set for semantically representing knowledge of a fault evolution causal chain of the distributed photovoltaic power generation equipment, and comprises various energy efficiency fault knowledge, such as a fault mode, fault symptoms, fault reasons, fault influence, maintenance measures and the like.
In specific implementation, as a possible implementation manner, energy efficiency fault knowledge can be converted into a structured language recognizable by a computer, modeling is performed through a preset open source tool based on the structured language to generate an energy efficiency diagnosis knowledge base of the distributed photovoltaic generator set based on the body, meanwhile, historical fault data of each distributed photovoltaic generator set can be obtained, prior probability assignment is performed on fault modes corresponding to different signs based on bayesian theorem, and a corresponding probability relation between the signs and the fault modes is generated.
Specifically, the preset open source tool may be "prot g", and other tools may be selected as needed. In the embodiment, the related failure knowledge of the unstructured photovoltaic generator set is expressed through a structured language which can be recognized by a computer, and an open source tool prot g is used for modeling to form a distributed photovoltaic generator set energy efficiency diagnosis knowledge base based on an ontology. When an open source tool prot g is used for ontology modeling, the prior probability assignment work of a symptom-fault mode can be completed by using data attributes according to the historical fault occurrence case of the distributed photovoltaic power generation equipment and combining the Bayes principle.
102, obtaining historical operation data of each photovoltaic power generation device, and determining a reference interval of detection data of each photovoltaic power generation device under different working conditions according to the historical operation data.
The historical operation data may be historical operation data of each photovoltaic power generation device in an operation state within a preset time, for example, the past year. In the embodiment of the application, the reference model can be trained according to data in the historical operating data under the normal operating state, namely, the operating parameter operating reference interval is monitored under the conditions of different irradiation and different temperatures in different working conditions of each device. And the real-time data can be conveniently and subsequently subjected to abnormity detection according to the reference interval.
In an embodiment of the present application, in order to more clearly illustrate a manner of determining a base interval, a specific method for determining a reference interval is further provided, and fig. 2 is a flowchart of the method for determining a reference interval provided in the embodiment of the present application, as shown in fig. 2, the method includes:
step 201, performing steady-state screening on the historical operating data, and screening out unsteady-state data in the historical operating data.
The unsteady data is data generated by the photovoltaic power generation equipment in an abnormal operation state caused by emergency such as shielding or coverage. In the embodiment of the application, steady-state screening is carried out on the operation data of the distributed photovoltaic generator set in the past year, and unstable-state data caused by emergency conditions such as shielding and covering are eliminated.
It should be noted that, because there are great differences between distributed photovoltaic devices in a distributed photovoltaic power station, for example, the environments in which the distributed photovoltaic devices are located are different, and for the differences of the devices, in an embodiment of the present application, a knowledge base for energy efficiency fault diagnosis of each distributed photovoltaic device may also be established, and by mining knowledge of historical operating data of the distributed photovoltaic devices, it is convenient to subsequently determine a reference interval for the current distributed photovoltaic device more accurately, and accuracy of diagnosis is improved.
Step 202, dividing the working condition of each photovoltaic power generation device by a K-mean clustering algorithm by taking the irradiance and the temperature as boundary conditions.
Specifically, the irradiance and the temperature are used as boundary conditions, working condition division is carried out based on a K-means clustering algorithm, different working conditions of operation of the photovoltaic power generation equipment are determined, and as an example, the optimal clustering number K can be determined through a Silhouette criterion. The reason for dividing the working conditions is that historical data under each working condition can be collected, reference intervals can be generated conveniently according to different working conditions, fault diagnosis can be carried out on subsequent combined working conditions, and diagnosis accuracy is improved.
And 203, eliminating fault data under each working condition through the multivariate Gaussian mixture model to generate a normal data training sample.
The multivariate Gaussian mixture model is a probability density estimation method based on half parameters by combining a parameter estimation method and a nonparametric estimation method. When the number of the sub-models is enough, the multivariate Gaussian mixture model can approach to any continuous distribution with higher precision, normal data in the middle position of a target in probability distribution is reserved, and abnormal data are removed.
The rejected fault data can be data in various abnormal operating states such as operating data of the photovoltaic power generation equipment during the fault period.
And 204, training the long-short term memory artificial neural network LSTM through the normal data training sample, and determining a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition according to the predicted value output by the long-short term memory artificial neural network LSTM.
Specifically, the long-term and short-term memory artificial neural network LSTM is trained by taking the data in the normal operation state acquired in the above steps as training data. Because the parameter types in the training data correspond to the parameter types in the actual operation data, after the long-short term memory artificial neural network LSTM is trained, a plurality of predicted values output by the long-short term memory artificial neural network LSTM can be obtained for each parameter to be detected, an actual detected actual value is obtained, and a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition is determined according to the predicted values and the actual values.
As an example, for each parameter, a plurality of parameter predicted values output by the long-short term memory artificial neural network LSTM and corresponding actual values are obtained, then a difference between each parameter predicted value and the actual value is calculated, the difference values are compared, sorting is performed according to the difference values, the parameter predicted value with the largest difference value with the actual value is taken as a target parameter predicted value, the target parameter predicted value is taken as a center, and a reference interval is constructed by taking two times of the difference value between the target parameter predicted value and the actual value as an interval length, namely the reference interval is (target parameter predicted value-maximum difference value, target parameter predicted value + maximum difference value).
Therefore, the reference interval corresponding to the full boundary condition under each working condition is determined.
And 103, acquiring real-time operation data of each photovoltaic power generation device, and performing abnormity detection on the real-time operation data based on the corresponding reference interval.
Specifically, the collected real-time operation data of each photovoltaic power generation device is compared with a corresponding reference interval, whether the operation data exceed the reference interval or not is detected in real time, and if the real-time operation data are not in the reference interval, the data can be judged to be abnormal data.
And 104, if the real-time operation data of any photovoltaic power generation equipment is abnormal data, determining signs of the abnormal data, performing inference diagnosis on any distributed photovoltaic power generation equipment through the corresponding probability relation between the signs and the fault modes preset in the body energy efficiency diagnosis knowledge base, and determining the abnormal mode and the fault mode of any distributed photovoltaic power generation equipment.
The sign of the abnormal data is evidence for identifying the occurrence of the failure mode and the cause, for example, the sign may be "the output voltage of the generator set drops" or "the output voltage of the generator set fluctuates greatly" or the like.
In the embodiment of the application, the signs of the abnormal data are input into the body energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment, and the distributed photovoltaic power generation equipment which generates the abnormal data is subjected to reasoning diagnosis through the corresponding probability relation between the signs and the fault modes which is preset in the body energy efficiency diagnosis knowledge base in the steps. The method comprises the steps of determining a fault mode corresponding to an assigned symptom according to the input symptom through the prior probability of the symptom-fault mode after assignment, so as to realize inference diagnosis work through a Bayesian model.
It should be noted that, in the embodiment of the present application, from the perspective of energy efficiency, a relevant operation abnormal mode causing the energy efficiency abnormality of the power generation device and the above-mentioned failure mode may also be analyzed, where the abnormal mode is a mode in which the photovoltaic power generation device is abnormal but no significant failure occurs, for example, the voltage output by the photovoltaic power generation device does not reach the set voltage, and the like. As one possible implementation manner, according to a pre-assigned prior probability of a "symptom-abnormal mode", an abnormal mode with the maximum probability corresponding to the current symptom may be determined as an abnormal mode for making a diagnosis and drawing a conclusion. The specific implementation is not limited herein.
It should be further noted that, in practical applications, different faults may have the same sign, that is, the same sign may correspond to different fault modes, and in order to further improve accuracy of energy efficiency diagnosis, in an embodiment of the present application, a working condition may also be used as supplementary information to assist in reasoning and diagnosing, that is, a fault with a low fault occurrence probability under the working condition is eliminated according to the working condition determined in the above step, so as to obtain a more accurate diagnosis result.
And 105, sending the diagnosis result and the maintenance strategy corresponding to the diagnosis result stored in the ontology energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance worker.
Specifically, the ontology knowledge base and the acquired data are stored on the server, and in addition, a mobile client interface can be developed, so that operation and maintenance personnel can access information of related equipment through the mobile terminal equipment according to the serial number of the photovoltaic power generation equipment, inquire a unit fault mode and maintenance measures, and master the operation conditions of all photovoltaic power generation equipment in a distributed power station area. And moreover, the health maintenance strategy corresponding to the current failure mode stored in the ontology knowledge base can be transmitted to the mobile client, so that operation and maintenance personnel can conveniently look up the maintenance strategy for maintenance.
In an embodiment of the present application, a fault record may be added to the mobile terminal of the operation and maintenance staff according to the currently determined fault occurrence mode, and the posterior probability of the corresponding probability relationship between the symptom and the fault pattern stored in the ontology knowledge base is updated according to the currently occurring fault, so as to update the relevant knowledge in the knowledge base according to the historical fault occurrence condition, and further improve the accuracy of diagnosis in the subsequent diagnosis process.
As one possible implementation manner, a fault record is first added to the mobile client of the operation and maintenance personnel according to the occurred fault mode, wherein the added fault record includes, but is not limited to, the occurred fault mode, the reason of the fault occurrence, the fault symptom, and the corresponding maintenance measure. During specific implementation, operation and maintenance personnel can input the fault record information in modes of checking or filling in and the like through the mobile terminal according to instructions, can automatically analyze and arrange the fault record through the application installed on the mobile terminal, and sends the fault record to the background server of the energy efficiency diagnosis system of the distributed photovoltaic power generation equipment
And then updating the posterior probability of the corresponding probability relation between the preset symptom and the fault mode according to the acquired fault record. In specific implementation, a background algorithm of the energy efficiency diagnosis system updates the posterior probability according to the current fault, in one embodiment of the application, whether the fault mode exists is judged firstly, if the fault mode exists, the posterior probability is updated, and if the fault mode does not exist, the fault mode is added to the body energy efficiency diagnosis knowledge base.
Therefore, the final diagnosis result and the corresponding health maintenance strategy in the ontology knowledge base are transmitted to the mobile client, operation and maintenance personnel can directly check the operation state of the designated unit to make a related maintenance plan, and the knowledge base is continuously updated according to the fault occurrence record, so that the posterior probability can be updated according to the current fault occurrence record, and the diagnosis accuracy is improved.
To sum up, the energy efficiency diagnosis method for distributed photovoltaic power generation equipment in the embodiment of the present application forms failure knowledge of abnormal energy efficiency level of a photovoltaic power generation unit by using an FMEA and FTA analysis method, then expresses relevant failure knowledge of an unstructured photovoltaic power generation unit by using a structured language recognizable by a computer based on a body concept to form an open body knowledge base, performs energy efficiency diagnosis on real-time data, performs abnormality detection according to a reference interval determined by the environmental characteristics of each piece of equipment by using a distributed photovoltaic power generation unit energy efficiency diagnosis and health maintenance flow of "abnormality detection-symptom determination-diagnosis inference-positioning failure-health maintenance decision-mobile terminal user reading", inputs abnormality information corresponding to the symptom into the energy efficiency diagnosis knowledge base, and performs inference diagnosis work based on a bayesian model, therefore, the information of the generator set with the fault and the corresponding specific fault can be quickly positioned, and a corresponding health maintenance strategy is provided. And moreover, through a mobile client interface, operation and maintenance personnel can access related unit information through mobile terminal equipment according to the number of the generator set, inquire a unit fault mode and maintenance measures, master the operation conditions of all the generator sets in the distributed power station area, make a maintenance plan in advance, reduce the operation and maintenance cost and improve the operation and maintenance efficiency.
In order to more clearly describe the energy efficiency diagnosis method of the distributed photovoltaic power generation apparatus according to the embodiment of the present application, a detailed description is given below with respect to an embodiment of energy efficiency diagnosis of a specific distributed photovoltaic power generation apparatus. Fig. 3 is a schematic flowchart of a specific energy efficiency diagnosis method for a distributed photovoltaic power generation device according to an embodiment of the present application.
As shown in fig. 3, when energy efficiency diagnosis is performed by the method, firstly, for real-time operation data of each power generation device obtained by field communication in a distributed photovoltaic power generation station, mechanism analysis and device analysis are performed on the distributed photovoltaic power generation devices by means of an FMEA and FTA analysis method, and common related fault knowledge causing abnormal energy efficiency level of a photovoltaic power generation unit is combed. The method comprises the steps of expressing relevant fault knowledge of an unstructured photovoltaic generator set through a structured language which can be recognized by a computer by means of an ontology concept, utilizing a source-opening tool prot g to model, forming a distributed photovoltaic generator set energy efficiency diagnosis knowledge base based on an ontology, and utilizing data attributes to complete prior probability assignment work of a symptom-fault mode according to historical fault occurrence cases and by combining a Bayesian principle when the source-opening tool prot g is utilized to model the ontology. And the operation and maintenance staff can add fault records on the mobile terminal according to the subsequent fault mode, and the background algorithm updates the posterior probability according to the current fault.
Then, aiming at each generator set, a reference model is trained according to historical operating data under the normal operating state of the generator set in the past year, irradiance is used as a reference interval for determining each monitoring operating parameter under the conditions of different irradiation and different temperatures, and fault data is used for checking to ensure the accuracy of the reference interval. And carrying out steady-state screening on the operation data of the distributed photovoltaic generator set in the past year, and eliminating unstable-state data caused by emergency conditions such as shielding, covering and the like. And then removing operation data during the fault period to form a normal data training sample, during specific implementation, dividing the working conditions based on a K-means clustering algorithm by taking irradiance and temperature as boundary conditions, determining the optimal clustering number K according to a silouette criterion, and removing abnormal data from historical data under each working condition by adopting a local optimal algorithm multi-Gaussian mixture model.
It should be noted that the multivariate gaussian mixture model is a probability density estimation method based on half-parameters by combining a parameter estimation method and a non-parameter estimation method. When the number of the sub-models is large enough, the multivariate Gaussian mixture model can approach any continuous distribution with higher precision, and the probability distribution of the multivariate Gaussian mixture model is shown as the following formula:
Figure BDA0003287999350000101
wherein X is L-dimensional parameter data column vector, and X is [ X ]1x2,...,xL]T(ii) a k is the number of sub-models in the multivariate Gaussian mixture model; omegakIs the weight coefficient of the kth sub-model, and
Figure BDA0003287999350000102
φ(X|θk) The Gaussian probability density function of the kth sub-model is represented, and the specific calculation formula is as follows:
Figure BDA0003287999350000103
the multivariate Gaussian mixture model adopts an Expectation Maximization (EM) algorithm for parameter estimation, and the EM algorithm is an iterative algorithm suitable for parameter estimation of a probability model containing hidden variables. Wherein, sigmakAnd mukValues are estimated by the EM algorithm for the covariance matrix and the mean. The objective function of the EM algorithm is:
Figure BDA0003287999350000111
the iterative process of the EM algorithm comprises the following steps: selecting initial values of mu and sigma, calculating the posterior probability of the corresponding kth model by formula (4), and finally iteratively calculating the mu and sigma in the model according to formulas (5), (6) and (7) when the difference between the objective functions of two adjacent iterations is less than 10-5And when so, stopping iteration.
Figure BDA0003287999350000112
Figure BDA0003287999350000113
Figure BDA0003287999350000114
Figure BDA0003287999350000115
Further, training an LSTM long-short term memory artificial neural network according to sample data, and predicting each parameter xpWith the actual value xtTwice the maximum sigma of the difference is taken as the length of the reference interval. Determining a reference interval (x) corresponding to the full boundary condition under each working conditionp-sigma,xpAnd + sigma), detecting whether the operation data exceeds a reference interval in real time, and finishing the abnormal detection work.
Among them, it should be further noted that LSTM is derived from a Recurrent Neural Network (RNN), is a kind of feedback neural network, and is improved against the problem of gradient occurring in long-term memory in RNN. Introducing threshold mechanisms such as a memory gate and a forgetting gate outside a standard cycle layer of the RNN, and connectingThe storage and forgetting of the weighting matrix control information can ensure that useful information in history data can be stored when long-time history data is processed. The basic principle is that the original RNN hidden layer only has one state h, which is sensitive to the short-term state, and on the basis, the LSTM network adds a long-term state quantity c to store the long-term state and transmits the state together with the state h. The LSTM network standard module comprises three parts of state output: long term state ctShort-term state htAnd the current state output ytThe three parts contain memory and forgetting information of historical information in long and short periods, and the memory and forgetting information and the current state output are transmitted into the next neuron of the LSTM network together, so that the memory and the propagation of time sequence information are ensured. The specific method of information regulation is to add three control gates of an input gate, a forgetting gate and an output gate in a state propagation path, which correspond to three stages in the LSTM network.
(1) Forgetting stage of long-term status: by calculating forgotten gating zfOutput c to last time sequence statet-1Processing is carried out to control which information needs to be forgotten in the long-term state in the time sequence state, and the memorized long-term information and the memory information generated at the current moment form a new unit state ctIncoming next-time LSTM network elements
(2) Selection memory stage of input: by inputting x for the current timetH transmitted from the previous timet-1Obtaining current input information matrix z by gating signal ziPerforming selective memory and long-term status after selective forgettingt -1Jointly updating the long-term memory information to form ct；
(3) An output stage: the output phase comprises yt、htAnd ctWherein h istAnd ctPassed as short-term and long-term memory states into the next temporal state, ytAccording to the current short-time state htAnd the change is obtained as the output of the current state.
Furthermore, the symptoms corresponding to the failure mode are input into an energy efficiency diagnosis knowledge base, and subsequent reasoning diagnosis work based on the Bayesian model is carried out. The method can realize the functional requirements of ontology query and ontology-based data access through the SPARQL language, and can retrieve target information conforming to users from the ontology by using the correlation mode of the SPARQL language definition information. A typical query statement for SPARQL is as follows:
SELECTx WHERE{？x rdfs:subClassOf:F01.}
the meaning of this equation is to query all subclasses of fault F01. SPARQL also supports multiple matches and literal matching. In group graph pattern queries, each triplet pattern must be matched successfully. Compared with a database SQL language, the SPARQL does not need to master complex query logic, and the knowledge related to the straight-sided query problem is suitable for a non-computer-professional user to design a query mechanism, so that the applicability of the method is improved.
And finally, transmitting the final diagnosis result and the corresponding health maintenance strategy in the ontology knowledge base to a mobile client, so that operation and maintenance personnel can directly check the operation state of the designated unit and make a related maintenance plan. In addition, the operation and maintenance personnel check out or fill in the fault mode, the fault occurrence reason, the fault symptom and related maintenance measures through the mobile terminal according to the indication, if the fault mode exists, the prior probability value in the related inference relation in the Bayesian model is adjusted after checking out, if the fault mode occurs for the first time, the operation and maintenance personnel need to fill in related information, and the system automatically enters the relevant information into the body knowledge base.
In order to implement the foregoing embodiments, the present application further provides an energy efficiency diagnosis system of a distributed photovoltaic power generation device, fig. 4 is a schematic structural diagram of the energy efficiency diagnosis system of the distributed photovoltaic power generation device according to the embodiments of the present application, and as shown in fig. 4, the diagnosis system includes a generation module 100, a determination module 200, an anomaly detection module 300, an inference diagnosis module 400, and a sending module 500.
The first obtaining module 100 is configured to analyze the distributed photovoltaic power generation device to be diagnosed through the failure mode, the impact analysis FMEA and the fault tree analysis FTA, obtain energy efficiency fault knowledge of the distributed photovoltaic power generation device, and generate a body energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation device.
The determining module 200 is configured to obtain historical operation data of each photovoltaic power generation device, and determine a reference interval of detection data of each photovoltaic power generation device under different working conditions according to the historical operation data.
The anomaly detection module 300 is configured to collect real-time operation data of each photovoltaic power generation device, and perform anomaly detection on the real-time operation data based on the corresponding reference interval.
And the reasoning diagnosis module 400 is configured to determine a sign of the abnormal data if the real-time operation data of any one of the photovoltaic power generation devices is the abnormal data, perform reasoning diagnosis on any one of the distributed photovoltaic power generation devices according to a corresponding probability relationship between the sign and a fault mode preset in the body energy efficiency diagnosis knowledge base, and determine the abnormal mode and the fault mode of any one of the distributed photovoltaic power generation devices.
A sending module 500, configured to send the diagnosis result and the maintenance policy corresponding to the diagnosis result and stored in the ontology energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance worker.
Optionally, in an embodiment of the present application, the generating module 100 is specifically configured to: converting the energy efficiency fault knowledge into a structured language recognizable by a computer; modeling is carried out through a preset open source tool based on a structured language so as to generate an energy efficiency diagnosis knowledge base of the distributed photovoltaic generator set based on the body; historical fault data of each distributed photovoltaic power generation device is obtained, prior probability assignment is carried out on fault modes corresponding to different symptoms based on Bayes theorem, and a corresponding probability relation between preset symptoms and the fault modes is generated.
Optionally, in an embodiment of the present application, the determining module 200 is specifically configured to: performing stable screening on the historical operating data, and screening out unstable data in the historical operating data; the irradiance and the temperature are used as boundary conditions, and the working condition of each photovoltaic power generation device is divided through a K-means clustering algorithm; abnormal data under each working condition are removed through a multivariate Gaussian mixture model, and normal data training samples are generated; training the long-short term memory artificial neural network LSTM through a normal data training sample, and determining a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition according to a predicted value output by the long-short term memory artificial neural network LSTM.
Optionally, in an embodiment of the present application, the determining module 200 is further configured to: aiming at each parameter, acquiring a plurality of parameter predicted values output by a long-term and short-term memory artificial neural network (LSTM) and corresponding actual values; calculating the difference value between each parameter predicted value and the actual value, and taking the parameter predicted value with the largest difference value with the actual value as a target parameter predicted value; and constructing a reference interval by taking the target parameter predicted value as a center and taking two times of the difference value between the target parameter predicted value and the actual value as an interval length.
Optionally, in an embodiment of the present application, the sending module 500 is further configured to add a fault record on the mobile client of the operation and maintenance staff according to the fault mode; and updating the posterior probability of the corresponding probability relation between the preset symptom and the fault mode according to the acquired fault record.
Optionally, in an embodiment of the present application, the sending module 500 is further configured to receive an occurred failure mode, a reason for the occurrence of the failure, a failure symptom, and a corresponding maintenance measure sent by the mobile client; judging whether the fault mode exists, and if the fault mode exists, updating the posterior probability; and if the occurred fault mode does not exist, adding the occurred fault mode to the ontology energy efficiency diagnosis knowledge base.
It should be noted that the foregoing explanation of the embodiment of the energy efficiency diagnosis method for the distributed photovoltaic power generation device is also applicable to the system of the embodiment, and details are not repeated here
To sum up, the energy efficiency diagnosis system of the distributed photovoltaic power generation equipment in the embodiment of the application can quickly locate the information of the generator set with faults and the corresponding specific faults, and provide corresponding health maintenance strategies. And moreover, through a mobile client interface, operation and maintenance personnel can access related unit information through mobile terminal equipment according to the number of the generator set, inquire a unit fault mode and maintenance measures, master the operation conditions of all the generator sets in the distributed power station area, make a maintenance plan in advance, reduce the operation and maintenance cost and improve the operation and maintenance efficiency.
In order to achieve the above embodiments, the present application also proposes a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the energy efficiency diagnosis method of the distributed photovoltaic power generation apparatus as described in any one of the above embodiments.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.

Claims (10)

1. The energy efficiency diagnosis method of the distributed photovoltaic power generation equipment is characterized by comprising the following steps of:
analyzing distributed photovoltaic power generation equipment to be diagnosed through failure modes, impact analysis FMEA and fault tree analysis FTA, acquiring energy efficiency fault knowledge of the distributed photovoltaic power generation equipment, and generating a body energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment;
acquiring historical operation data of each photovoltaic power generation device, and determining a reference interval of detection data of each photovoltaic power generation device under different working conditions according to the historical operation data;
acquiring real-time operation data of each photovoltaic power generation device, and performing anomaly detection on the real-time operation data based on the corresponding reference interval;
if the real-time operation data of any one photovoltaic power generation device is abnormal data, determining signs of the abnormal data, performing inference diagnosis on any one distributed photovoltaic power generation device through a corresponding probability relation between the signs and a fault mode preset in the body energy efficiency diagnosis knowledge base, and determining the abnormal mode and the fault mode of any one distributed photovoltaic power generation device;
and sending the diagnosis result and the maintenance strategy corresponding to the diagnosis result stored in the body energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance worker.
2. The diagnostic method of claim 1, wherein the generating the ontology energy efficiency diagnostic knowledge base of the distributed photovoltaic power generation facility comprises:
converting the energy efficiency fault knowledge into a structured language recognizable by a computer;
modeling through a preset open source tool based on the structured language to generate an energy efficiency diagnosis knowledge base of the distributed photovoltaic generator set based on the ontology;
and acquiring historical fault data of each distributed photovoltaic power generation device, and performing prior probability assignment on fault modes corresponding to different symptoms based on Bayesian theorem to generate a corresponding probability relation between the preset symptoms and the fault modes.
3. The diagnostic method according to claim 1 or 2, wherein the determining of the reference interval of the detection data of each photovoltaic power generation device under different working conditions according to the historical operation data comprises:
performing steady-state screening on the historical operating data, and screening out unsteady-state data in the historical operating data;
the irradiance and the temperature are used as boundary conditions, and the working condition of each photovoltaic power generation device is divided through a K-mean clustering algorithm;
removing fault data under each working condition through a multivariate Gaussian mixture model to generate a normal data training sample;
and training a long-short term memory artificial neural network (LSTM) through the normal data training sample, and determining a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition according to a predicted value output by the long-short term memory artificial neural network (LSTM).
4. The diagnosis method according to claim 3, wherein the determining a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition according to the predicted value of the parameter output by the long-short term memory artificial neural network LSTM comprises:
aiming at each parameter, acquiring a plurality of parameter predicted values output by the long-short term memory artificial neural network LSTM and corresponding actual values;
calculating the difference value between each predicted parameter value and the actual value, and taking the predicted parameter value with the largest difference value with the actual value as a predicted target parameter value;
and constructing the reference interval by taking the target parameter predicted value as a center and taking two times of the difference value between the target parameter predicted value and the actual value as an interval length.
5. The diagnosis method according to claim 1, after the sending the diagnosis result and the maintenance policy corresponding to the diagnosis result stored in the ontology energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance worker, further comprising:
adding fault records on the mobile client of the operation and maintenance personnel according to the generated fault modes;
and updating the posterior probability of the corresponding probability relation between the preset symptom and the fault mode according to the acquired fault record.
6. The diagnostic method of claim 5, wherein adding a fault record on the mobile client of the operation and maintenance personnel comprises:
receiving the failure mode, the failure reason, the failure symptom and the corresponding maintenance measure sent by the mobile client;
the updating of the posterior probability of the corresponding probability relationship between the preset symptom and the fault mode according to the acquired fault record includes:
judging whether the fault mode exists, and if the fault mode exists, updating the posterior probability;
and if the fault mode does not exist, adding the fault mode to the ontology energy efficiency diagnosis knowledge base.
7. An energy efficiency diagnostic system for a distributed photovoltaic power generation facility, comprising:
the generating module is used for analyzing the distributed photovoltaic power generation equipment to be diagnosed through failure mode and influence analysis FMEA and fault tree analysis FTA, obtaining energy efficiency fault knowledge of the distributed photovoltaic power generation equipment and generating a body energy efficiency diagnosis knowledge base of the distributed photovoltaic power generation equipment;
the determining module is used for acquiring historical operating data of each photovoltaic power generation device and determining a reference interval of detection data of each photovoltaic power generation device under different working conditions according to the historical operating data;
the abnormity detection module is used for acquiring real-time operation data of each photovoltaic power generation device and carrying out abnormity detection on the real-time operation data based on the corresponding reference interval;
the reasoning diagnosis module is used for determining the sign of the abnormal data if the real-time operation data of any photovoltaic power generation equipment is abnormal data, carrying out reasoning diagnosis on any distributed photovoltaic power generation equipment through the corresponding probability relation between the sign and the fault mode preset in the body energy efficiency diagnosis knowledge base, and determining the abnormal mode and the fault mode of any distributed photovoltaic power generation equipment;
and the sending module is used for sending the diagnosis result and the maintenance strategy corresponding to the diagnosis result stored in the body energy efficiency diagnosis knowledge base to a mobile client of an operation and maintenance worker.
8. The diagnostic system of claim 7, wherein the generation module is specifically configured to:
converting the energy efficiency fault knowledge into a structured language recognizable by a computer;
modeling through a preset open source tool based on the structured language to generate an energy efficiency diagnosis knowledge base of the distributed photovoltaic generator set based on the ontology;
and acquiring historical fault data of each distributed photovoltaic power generation device, and performing prior probability assignment on fault modes corresponding to different symptoms based on Bayesian theorem to generate a corresponding probability relation between the preset symptoms and the fault modes.
9. The diagnostic system of claims 7 and 8, wherein the determination module is specifically configured to:
performing steady-state screening on the historical operating data, and screening out unsteady-state data in the historical operating data;
the irradiance and the temperature are used as boundary conditions, and the working condition of each photovoltaic power generation device is divided through a K-mean clustering algorithm;
removing fault data under each working condition through a multivariate Gaussian mixture model to generate a normal data training sample;
and training a long-short term memory artificial neural network (LSTM) through the normal data training sample, and determining a reference interval corresponding to the boundary condition of each photovoltaic power generation device under each working condition according to a predicted value output by the long-short term memory artificial neural network (LSTM).
10. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the energy efficiency diagnosis method for a distributed photovoltaic power generation apparatus according to any one of claims 1 to 6.
CN202111153747.7A 2021-09-29 2021-09-29 Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment Pending CN113888353A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111153747.7A CN113888353A (en) 2021-09-29 2021-09-29 Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111153747.7A CN113888353A (en) 2021-09-29 2021-09-29 Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment

Publications (1)

Publication Number Publication Date
CN113888353A true CN113888353A (en) 2022-01-04

Family

ID=79008362

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111153747.7A Pending CN113888353A (en) 2021-09-29 2021-09-29 Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment

Country Status (1)

Country Link
CN (1) CN113888353A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115691026A (en) * 2022-12-29 2023-02-03 湖北省林业科学研究院 Intelligent early warning monitoring management method for forest fire prevention
CN115695150A (en) * 2022-11-01 2023-02-03 广州城轨科技有限公司 Method and device for detecting networking equipment based on distributed heterogeneous fusion
CN116628608A (en) * 2023-04-23 2023-08-22 华能国际电力江苏能源开发有限公司 Photovoltaic power generation fault diagnosis method and system

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115695150A (en) * 2022-11-01 2023-02-03 广州城轨科技有限公司 Method and device for detecting networking equipment based on distributed heterogeneous fusion
CN115695150B (en) * 2022-11-01 2023-08-08 广州城轨科技有限公司 Method and device for detecting networking equipment based on distributed heterogeneous fusion
CN115691026A (en) * 2022-12-29 2023-02-03 湖北省林业科学研究院 Intelligent early warning monitoring management method for forest fire prevention
CN115691026B (en) * 2022-12-29 2023-05-05 湖北省林业科学研究院 Intelligent early warning monitoring management method for forest fire prevention
CN116628608A (en) * 2023-04-23 2023-08-22 华能国际电力江苏能源开发有限公司 Photovoltaic power generation fault diagnosis method and system

Similar Documents

Publication Publication Date Title
Javed et al. State of the art and taxonomy of prognostics approaches, trends of prognostics applications and open issues towards maturity at different technology readiness levels
KR102354112B1 (en) Apparatus and method for estimating status of battery based on artificial intelligence
CN113888353A (en) Energy efficiency diagnosis method, system and medium for distributed photovoltaic power generation equipment
Wang et al. Remaining useful life estimation using functional data analysis
JP2020119605A (en) Abnormality detection system, abnormality detection method, abnormality detection program, and method for generating learned model
CN109255440B (en) Method for predictive maintenance of power production equipment based on Recurrent Neural Networks (RNN)
CN114021932A (en) Energy efficiency evaluation and diagnosis method, system and medium for wind turbine generator
Chen et al. A health index construction framework for prognostics based on feature fusion and constrained optimization
CN115822887A (en) Performance evaluation and energy efficiency diagnosis method and system of wind turbine generator
Kharlamova et al. A digital twin of battery energy storage systems providing frequency regulation
Wang et al. A remaining useful life prediction model based on hybrid long-short sequences for engines
Abdelli et al. Machine learning based data driven diagnostic and prognostic approach for laser reliability enhancement
Melendez et al. Self-supervised Multi-stage Estimation of Remaining Useful Life for Electric Drive Units
CN117169761A (en) Battery state evaluation method, apparatus, device, storage medium, and program product
Bond et al. A hybrid learning approach to prognostics and health management applied to military ground vehicles using time-series and maintenance event data
US11339763B2 (en) Method for windmill farm monitoring
CN115600695A (en) Fault diagnosis method of metering equipment
RU2699685C1 (en) Method of analyzing and monitoring the state of a technical installation comprising a plurality of dynamic systems
CN114548701A (en) Process early warning method and system for analyzing and estimating coupling structure of full-scale measuring point
Esteves et al. Prognostics health management: perspectives in engineering systems reliability prognostics
Khan et al. Outliers detection and repairing technique for measurement data in the distribution system
CN117272844B (en) Method and system for predicting service life of distribution board
CN117647725B (en) Aging test method and system for PCBA
RU2818858C1 (en) Method for diagnosing a complex of on-board equipment of aircraft based on unsupervised machine learning with automatic determination of model training parameters
Wang et al. Complex equipment diagnostic reasoning based on neural network algorithm

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