Photovoltaic (PV) array fault diagnosis is important because it helps reduce energy and revenue losses to PV system operators. It also reduces fire hazards and electric shocks caused by PV array faults. As a result, many machine-learning-based fault diagnosis techniques have been proposed in recent times. Although the fault diagnosis accuracies associated with these techniques have been impressive, most machine learning algorithms rely on manual feature extraction, which is time consuming, expensive, and diagnostic expertise exacting. To address the problem of manual feature extraction, this paper proposes a new PV array fault diagnosis technique capable of automatically extracting features from raw data for PV array fault classification. The proposed technique utilizes long short-term memory networks, which is a deep learning algorithm, for feature extraction. The extracted features feed into a softmax regression classifier for fault diagnosis. The proposed technique exhibits high fault diagnosis accuracies on both noisy and noiseless data. In addition, the results of the proposed technique compare favorably with those of other techniques. It can, therefore, be inferred from the results that the proposed fault diagnosis technique offers an effective approach to automatically extract useful features from raw data and thus remove the need for the manual feature extraction.

/pdf/long-short-term-memory-networks-based-automatic-feature-2zbdlgfglk.pdf

Long Short-Term Memory Networks Based Automatic Feature Extraction for Photovoltaic Array Fault Diagnosis

The proper monitoring and operation and maintenance (O&M) of solar photovoltaic (PV) systems are an integral part of the service tasks required to ensure long-term reliability and prolonged lifetime of the installation. In the recent years, with the end of the feed-in-tariff legislation, the PV stakeholders and operators shifted their focus from the design and construction of utility-scale grid-connected PV plants to the development of monitoring and more efficient O&M practices. Service technical personnel have been asked to investigate and resolve unexpected failures of PV systems, whose repair may require specific strategies not initially planned during the design stage. Reliability and, consequently, availability of PV systems are dependent upon a full understanding of failure dynamics, which requires suitable practices to either mitigate or eliminate repeated components failures. Currently, guidelines or technical standards that include PV systems-specific data for reliability assessments are not available, which has caused nonstandardized fault analyses. These differences in practices have somehow compromised the understanding of PV systems reliability at industry level. Based on information gathered from more than 80 PV plants located in Italy, this article outlines remote monitoring techniques that may be used for a general standardization and as a common basis for reliability assessments and an effective O&M. By discussing the most widespread issues, major failures and unexpected events that can occur in PV systems, the authors identify novel remote monitoring techniques to improve both failure reporting and corrective action systems.

Review of O&M Practices in PV Plants: Failures, Solutions, Remote Control, and Monitoring Tools

The invention relates to an island detection method of a photovoltaic grid-connected system. In views of the problems that independently-adopted passive detection mode is effective only when effective when mismatching degrees of source and load are large and a large non-detection zone (NDZ) can be accordingly provided; and when an active detection mode is independently-adopted, the system is disturbed, and therefore the disturbance results in instability of the system, and accordingly over-voltage / under-voltage is provided. An over-frequency / under-frequency detection method is combined with an improved impedance detection method to conduct island detection to the system, and therefore the NDZ is reduced to a minimum, quality of electric energy is improved and total harmonic distortion (THD) is reduced.

Island detection method of photovoltaic grid-connected system

A cascaded H-bridge (CHB) converter has been widely used and researched in industry since it is suitable for the operation under both normal and fault conditions. This paper proposes a novel pulsewidth modulation strategy based on state transition for CHB inverter with unbalanced dc sources to achieve high-quality line-to-line output voltages and maximize the linear modulation range. In this modulation strategy, the duration time of each switching state will be modified directly through the correction value. Ranges of correction value are derived by analyzing the modulation index limitation. Then, a proper correction value is added into duration times to transform the switching states and extend modulation index to the maximum value. Meanwhile, balanced ac currents can be obtained under unbalanced dc sources condition, even under larger unbalanced coefficients. Furthermore, a three-phase power control algorithm is introduced to achieve the balanced distribution of three-phase power. Compared with the traditional zero-sequence voltage injection method, the proposed strategy is more convenient and effective theoretically, and it can be applied to the higher level CHB converter. The advantage and effectiveness of the proposed strategy are verified by simulation and experimental results.

A PWM Strategy Based on State Transition for Cascaded H-Bridge Inverter Under Unbalanced DC Sources

In one aspect, a method for controlling the operation of a power generation system configured to supply power to an electrical grid may generally include detecting an occurrence of an islanding event associated with the power generation system and adjusting a regulator gain applied within a regulator of the power generation system to an islanding gain value in response to the detection of the islanding event, wherein the islanding gain value exceeds a maximum gain value defined for the regulator in the event of an occurrence of any ride-through transient event In addition, the method may include controlling a power converter of the power generation system based on a control signal generated by the regulator as the power generation system is being isolated from the electrical grid and shutting down the power generation system upon isolation of the power generation system from the electrical grid

System and method for controlling a power generation system based on a detected islanding event

The invention discloses an island detection method of a high-power photovoltaic grid-connected converter. The method comprises the following steps of: determining whether a system operates at a grid-connected power generation state, and starting up an island detection algorithm instantly if the system is at the grid-connected power generation state; detecting whether the frequency of the voltage of a port power grid is within the normal range in real time, if the frequency is not within the normal range, considering that the system is at the island state, and carrying out automatic safety stop on the system; adding reactive current Iqdisturb with the duration of td periodically within the normal range; and superposing the reactive current Iqdisturb and reactive control Icqset of the converter, thus an axis q component reference Icqref of grid-connected current of the converter is formed finally. The contradiction between two aspects of reliability of detection and the quality after grid-connected electric energy reduction for the traditional island detection algorithm is effectively solved by the method provided by the invention, and the implementation method is simple and effective.

Island detection method of high-power photovoltaic grid-connected converter

The invention discloses a three-phase inverter parallel-connection control method without an output isolation transformer. The method comprises the following steps: an output end of each three-phase inverter is connected with an LCL filter, and a filtering capacitor of each LCL filter is connected with a dumping resistor in series, so that the stability is improved, the harmonic wave of a harmonic point is limited, and stability range of control parameters is increased. By means of pre-synchronization before parallel connection and carrier synchronization in the process of parallel connection operation, the phases among the three-phase inverters and parallel-connected communication buses are synchronized, carriers among the three-phase inverters connected in parallel are synchronized, and stability of parallel connection is guaranteed. By means of the algorithms combining pole assignment capacitor voltage control and virtual resistance matching control, the response speed of the inverters is improved, the control performance is improved, and the influence on parallel-connection current sharing by circuit resistance is lowered. By means of seamless mode switching and alternative current over-current and short-circuit failure protection base on three-ring control, under the condition of absence of the output isolation transformer, the three-phase inverter parallel-connection control method is used for independently resolving short-circuit failures of a parallel-connection micro power grid.

Three-phase inverter parallel-connection control method without output isolation transformer

Method for controlling three-phase current transformer in photovoltaic grid-connected power generating system

The invention discloses a square-matrix insulation resistance detection system for a photovoltaic power generation system. The system includes a positive-pole-end sampling circuit, and a negative-pole-end sampling circuit. A microprocessor calculates sampling signals, so that a positive-pole to-earth insulation resistance and a negative-pole to-earth insulation resistance of a photovoltaic square matrix are obtained and compared with set values input to the microprocessor by an insulation resistance input circuit for judgment and alarm output. The square-matrix insulation resistance detection system is capable of measuring and calculating a direct current insulation resistance between an input end of a power generation system connected with the photovoltaic square matrix and the ground, so that when the insulation resistance does not comply with a limit-value demand, the system indicates a fault and reports the fault to an external controller. Therefore, the square-matrix insulation resistance detection system is capable of effectively ensuring insulation performance detection of a solar-panel square matrix of a photovoltaic power generation system and is a simple, reliable and rapid detection system which provides powerful guarantee for safe operation of the photovoltaic power generation system.

Square-matrix insulating resistance detection system for photovoltaic power generation system

When designing a series resonant converter, a resonant inductor is usually connected on series in the circuit which leads to an increase in the weight of the device. If the leakage inductance of the transformer is used as a resonant inductor, the weight will be reduced, but the high-frequency leakage flux will bring higher losses. In order to solve this problem, this paper deduces the analytical expression of leakage inductance with respect to winding spacing by magnetic field partitioning method, and constructs a mathematical model about the weight of series resonant inductor. The minimum weight is set as the optimization target, the transformer efficiency as constraints. The Fmincon function in matlab is used to obtain the optimal value of the winding spacing. A three-winding transformer prototype with rated working frequency of 15 kHz and rated power of 25 kW was designed and wounded through this method. The experimental results of the transformer were consistent with the theoretical calculations. The optimization method in this paper can provide reference for the design of high frequency transformers for series resonant converters.

Optimum Design of High Frequency Transformer Based on Winding Spacing

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Guangsen Wang

Author Tools

Chat about Author