Method for detecting faults in a photovoltaic installation

Abstract

The invention relates to a method for detecting defects in a photovoltaic installation comprising several strings of photovoltaic panels. The method comprises the following steps: - collection of: - structural information relating to the photovoltaic installation, - meteorological data; and - operating data for each string of photovoltaic panels; - from the structural information and meteorological data, determination of a fitted predictive production model for the photovoltaic installation; - identification of the presence of defects on each of the strings of the photovoltaic installation by submitting the fitted predictive production model, the meteorological data, and the operating data to at least one pre-trained machine learning algorithm so as to identify each string of panels containing defects. Figure for the abstract: Fig. 3

Description

Title of the invention: Method for detecting faults in a photovoltaic installation Technical field

[0001] The present invention relates to a method for detecting faults in photovoltaic installations, and devices for implementing such a detection method. Prior art

[0002] Fault detection and diagnosis are essential to ensure the normal operation of a photovoltaic installation throughout its lifetime, typically around 25 years, while minimizing significant production losses.

[0003] Furthermore, these approaches are essential to avoid destructive consequences and potential dangers for personnel coming into contact with the installations. In particular, the detection and diagnosis of faults are necessary in order to avoid human risks and / or energy losses that could cause fires.

[0004] For these various reasons, of safety and productivity, detection and diagnostic methods have been developed and aim to determine, with the greatest possible certainty, that faults have occurred on a photovoltaic installation and that said installation is therefore no longer operating within its optimal operating range. The conventional methods traditionally implemented are based on a priori knowledge, estimations and field measurements, and exploit visual methods, image processing methods, methods based on detections at the level of the electrical operation of the installation, techniques based on the implementation of protection devices, or techniques for detecting anti-reflective coating defects. There are also several machine learning approaches for detecting certain faults in photovoltaic installations.However, existing machine learning approaches are insufficient, not being able to detect or differentiate certain defects, such as snail tracks. In particular, these approaches are sensitive and dependent on the photovoltaic installation concerned and the sunlight conditions in which the installation diagnosis is carried out. For example, these approaches do not necessarily distinguish between a drop in production due to a change in sunlight conditions and a drop in production due to a fault in the installation.

[0005] However, these conventional methods of fault detection and diagnosis are limited. In particular, these conventional methods reach their own limits when fault conditions occur in several elements of the installation (inverters, string, batteries, connectors, etc.), or when multiple faults with different impact levels are present, or when faults with expressions in different domains (frequency or time) or with signatures that do not come from a single type of data such as electrical and / or thermal signatures.

[0006] Conventional methods also have difficulty detecting the occurrence of several simultaneous faults or primary faults generating more serious secondary faults. In particular, there are faults or combinations of faults which produce an electrical response almost identical to the electrical response of a non-defective installation and which these conventional methods do not allow to be identified.

[0007] Indeed, the methods conventionally implemented analyze a global response of the photovoltaic installation and do not allow the different signatures to be distinguished. They require checking each panel by visual analysis or image analysis, which is long and costly.

[0008] Difficult-to-detect defects include corrosion, snail trail, microcracks, delamination, yellowing, bypass diode failures, backplane failures, and hot spots. This list is not exhaustive.

[0009] Finally, some of these conventional methods take a long time to implement and / or require a shutdown of all or part of the photovoltaic installation during the diagnostic period, thus involving significant regular production losses. Statement of the invention

[0010] There is therefore a need for an improved method for detecting faults in a photovoltaic installation.

[0011] In particular, there is a need for a method allowing the detection of any type of defects, and in particular defects which cannot be detected by conventional methods, such as “Snail Trail”.

[0012] There is also a need for a method for detecting faults in any type of photovoltaic installation, regardless of the age of the installation or the technology used in the installation.

[0013] Generally, there is a need for an improved fault detection method that can be easily implemented for any photovoltaic installation and that does not involve a loss of production.

[0014] The invention aims to meet all or part of these needs. Summary of the invention

[0015] According to a first aspect of the invention, the invention relates to a method for detecting faults in a photovoltaic installation comprising several strings of photovoltaic panels, the method comprising the following steps: - collection: - structural information relating to the structure of the photovoltaic installation; - meteorological data; - operating data, for example voltage and current, for each string of photovoltaic panels in the photovoltaic installation; - based on structural information and meteorological data, determination of an adjusted predictive production model of the photovoltaic installation, modeling an optimal theoretical electrical production of the photovoltaic installation with regard to the meteorological data, the adjusted predictive production model being determined assuming that the photovoltaic installation does not have any faults; - identification of the presence of faults on each of the strings of the photovoltaic installation by submitting the adjusted predictive production model, the meteorological data and the operating data to at least one pre-trained machine learning algorithm so as to identify each string of panels comprising faults, called “defective string”, and preferably the types of faults.

[0016] In a particular embodiment, the adjusted production predictive model is determined by: - a preliminary generation of a reference predictive model determined based on the structural information of the photovoltaic installation then - an adjustment of the reference predictive model based on meteorological data, the reference predictive model modeling a theoretical electrical production profile for the photovoltaic installation considering the absence of faults on all photovoltaic panel strings and independently of meteorological data.

[0017] Optionally, the method comprises, after the step of identifying the presence of defects, a step of presenting a report indicating the defective chain(s), preferably the types of defects, and preferably one or more maintenance actions, the maintenance actions being determined from the identification of the presence of defects on the defective chain(s), and possibly actually from the fitted production predictive model and operating data relating to each defective line, preferably by means of statistical analysis, preferably by means of a residual-based method.

[0018] One of the main advantages of the invention is that it allows universal detection. In other words, the method allows fault detection on any type of photovoltaic installation regardless of the topology of the photovoltaic installation, its technology, its age, or even the weather conditions to which the installation is subjected. For this, the invention is based on the joint analysis, preferably by means of a supervised machine learning algorithm and an unsupervised machine learning algorithm, of structural information relating to the structure of the installation, operating data of the installation and meteorological data.

[0019] The combined use of an unsupervised machine learning algorithm and a supervised machine learning algorithm, in particular in cascade, accelerates the identification of faults seriously harming the production of the photovoltaic installation by limiting the requirements for computing resources, in particular thanks to the unsupervised machine learning algorithm, while allowing the fine detection of faults and a preventive diagnosis of faults, in particular thanks to the supervised learning algorithm.

[0020] In particular, the invention makes it possible not only to detect that a photovoltaic installation has one or more faults but also to locate the string(s) of photovoltaic panels having these faults and to identify what type of faults they are.

[0021] Preferably, the meteorological data may comprise at least one irradiation and one ambient temperature. In one embodiment, the meteorological data comprises at least one irradiation, one ambient temperature and one wind speed. This list is not limiting.

[0022] The structural information may include: a brand and / or an age of the photovoltaic installation and / or a material of the photovoltaic panels and / or a technology of the photovoltaic panels and / or an inclination of the photovoltaic panels and / or a topology of the photovoltaic installation and / or information relating to a replacement of photovoltaic panels and / or a number of panels per string of panels and / or a number of strings in the photovoltaic installation. This list is not exhaustive.

[0023] Preferably, the meteorological data and / or the operating data, preferably the meteorological data and the operating data are acquired at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 10 Hz, preferably greater than or equal to 100 Hz, preferably greater than or equal at 500 Hz, preferably greater than or equal to 1 kHz.

[0024] The collection of operating data is preferably carried out by means of at least one analog sensor.

[0025] The collection of structural information may be carried out by means of a graphical interface and / or by means of a database. The collection of structural information is for example carried out by an operator. All or part of the structural information may be extracted from a database.

[0026] The invention allows in particular the detection of defects at the level of photovoltaic panel chains. Such detection at the level of a photovoltaic panel chain allows defects to be located more precisely and more quickly.

[0027] In a particular embodiment, the operating data are divided into time windows, the identification of the presence of defects being carried out for each of the time windows, at least one machine learning algorithm being pre-trained for each of said time windows.

[0028] The number of time windows in a day may be greater than or equal to 2 and / or less than or equal to 50.

[0029] The operating data are for example divided into four time windows. In particular, they can be divided into a window corresponding to the morning, a window corresponding to midday, a window corresponding to the afternoon, and a window corresponding to the night. This division is given as an example.

[0030] Preferably, the time windows are such that in normal operation, and under constant weather conditions, the production of the installation over the entire duration of the time window is substantially constant.

[0031] Preferably, the time windows make it possible to isolate particular operating cycles of the photovoltaic installation. In particular, a photovoltaic installation does not produce in the same way at the end of the day as in the middle of the day. The analysis of faults by time window thus makes it possible to detect faults detectable only when the photovoltaic installation produces little or, on the contrary, when it operates at maximum capacity.

[0032] Operating and meteorological data can be normalized.

[0033] The operating and meteorological data may be preprocessed so as to extracting attributes therefrom, the identification of the presence of defects being carried out by submitting these attributes to at least one pre-trained machine learning algorithm so as to identify each defective chain, and preferably the types of defects, from a training database.

[0034] The attributes may be determined by decomposing the operating and meteorological data, for example via wavelet decomposition, and / or may be determined using an attribute selection algorithm, for example by analysis of attribute correlations and / or by analysis of variance.

[0035] Advantageously, the invention includes an analysis of the detected, located and identified faults, so as to prioritize maintenance and repair tasks to be carried out.

[0036] The invention thus provides a decision-making support method enabling optimization of repairs and therefore optimization of the operation of the photovoltaic installation.

[0037] The prioritization of maintenance and repair tasks to be performed may depend on the type(s) of defect identified, for example depending on the presence of snail tracks, broken glass, delamination, corrosion, yellowing, microcracks in solar cells, cracks in solar cells, potential-induced degradation, a defective junction box, cracks in a backsheet.

[0038] The prioritization of the maintenance and repair tasks to be carried out may in particular depend on the production or safety benefit that the tasks to be carried out would bring, the cost generated by the tasks to be carried out, the time required for the repair, whether or not it is necessary to shut down the photovoltaic installation to carry out a task, considered alone or in combination. In particular, the prioritization may depend on a ratio between a production or safety benefit and a cost, and / or a ratio between a production or safety benefit and a time required for the repair.

[0039] Preferably, at each implementation of the detection method, the collected data and the identification of the defects are stored so as to allow the updating of the automatic learning algorithms. Optionally, the identification of the defects is validated by a domain expert, for example by verifying that the actual defect corresponds to the identified defect.

[0040] The invention also relates, according to another of its aspects, to a computer program comprising instructions, which when the program is executed by a computer, lead the latter to implement a method of detecting defects according to the invention.

[0041] Another aspect of the invention relates to a measuring device for collecting meteorological and operating data in a photovoltaic installation comprising several strings of photovoltaic panels connected in parallel to each other to a junction box, the device comprising: - a weather station for the acquisition of meteorological data, and - a module for measuring operating data, in particular voltage and current, the measurement module being configured to acquire the data at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 10 Hz, preferably su greater than or equal to 100 Hz, preferably greater than or equal to 500 Hz, more preferably greater than or equal to 1 kHz, and to be connected in series between at least part of the strings of panels of the photovoltaic installation, preferably all the strings of panels, and the junction box, the measuring module being configured to acquire operating data for each string of panels connected to the measuring device.

[0042] The weather station can acquire weather data within a radius of about 100 m around the measuring device, preferably within a radius of about 10 m.

[0043] The weather station may be configured to acquire weather data at a frequency identical to the acquisition frequency of the measurement module.

[0044] The measurement module may advantageously comprise at least one analog sensor. The analog sensors allow, among other things, the acquisition of data at a high sampling frequency, in particular greater than 1 kHz and less than 1 MHz.

[0045] The measuring module and the weather station can be connected by wired communication means and / or by wireless communication means.

[0046] The measurement module and / or the weather station may be configured to allow the storage and / or processing of data. In a particular embodiment, only the measurement module or the weather station is configured to allow the storage and / or processing of operating and weather data.

[0047] In one embodiment, the device is configured to be connected to at least one string, preferably at least 8, preferably at least 10, preferably at least 15, even better at least 20, even better at least 27 and / or less than 500, or less than 100 strings of photovoltaic panels.

[0048] In one embodiment, the device is configured to be connected to at least 25% of the strings of photovoltaic panels of a photovoltaic installation, preferably at least 50%, even better at least 75%. Preferably, the device is configured to be connected to all the strings of photovoltaic panels of a photovoltaic installation.

[0049] The connection of the device to the strings of panels can advantageously be carried out at the junction box of a photovoltaic installation, the junction box being used to allow the parallel connection of the strings of panels of a photovoltaic installation.

[0050] According to yet another aspect of the invention, the invention relates to a device for detecting faults in a photovoltaic installation comprising several strings of photovoltaic panels connected in parallel to each other to a box junction, the device comprising: - a measuring device for collecting meteorological and operating data according to the invention, and - a computer program comprising instructions which, when the program is executed by a computer, cause the latter to implement a fault detection method according to the invention.

[0051] The fault detection device may in particular comprise means for acquiring structural information from the photovoltaic installation, preferably the acquisition means comprising a graphical interface.

[0052] The invention also relates to a photovoltaic installation comprising several strings of photovoltaic panels in parallel connected to a junction box, in which a fault detection device according to the invention is connected in series between all or part of the strings of panels and the junction box. Definitions

[0053] The structural information defines the intrinsic characteristics of the photovoltaic installation, and in particular the characteristics relating to the technology of the photovoltaic panels and their arrangement. Among the structural information, we can cite among others the technology of the photovoltaic panels, the number of photovoltaic panels in series for each string of photovoltaic panels, the age of the photovoltaic installation, the age of the photovoltaic panels, information relating to replacements carried out in the photovoltaic installation, the inclination and / or orientation of the photovoltaic panels, the average, minimum and / or maximum height of the photovoltaic installation or of the photovoltaic panels relative to the ground.

[0054] A photovoltaic panel technology depends on the type of cells used to form the panel. Depending on the cells used, a photovoltaic panel does not produce energy in the same way and does not have the same profitability.

[0055] The technology of a photovoltaic panel can be based on panels made of monocrystalline silicon cells, polycrystalline, i.e. based on mixed silicon crystals, amorphous, i.e. based on cells formed by vaporization of a layer of silicon on an amorphous surface, such as glass, steel or polypropylene, or hybrid, i.e. based on heterojunction or tandem cells.

[0056] An adjusted production predictive model provides a prediction of the production of a string of panels or a photovoltaic installation in a theoretically ideal situation for given weather conditions, in other words in a situation for which the string of panels or respectively the photovoltaic installation does not have any faults and operates at its maximum capacity in considering the given weather conditions. An adjusted production predictive model is said to be “adjusted” because it is adapted to the given weather conditions.

[0057] A reference predictive model provides a prediction of the production of a string of panels or a photovoltaic installation in an ideal theoretical situation, independently of the weather conditions, in other words in a situation for which the string of panels or respectively the photovoltaic installation does not have any faults and operates at its maximum capacity independently of the given weather conditions.

[0058] The reference and adjusted predictive models simulate optimal theoretical electrical production for a string of panels or for a photovoltaic installation.

[0059] A reference predictive model or an adjusted production predictive model may in particular take the form of a power curve. Brief description of the drawings

[0060] The invention may be better understood by reading the detailed description which follows, non-limiting examples of its implementation, and by examining the attached drawing, in which: - [Fig.l] [Fig.l] represents the steps of a method of detecting defects according to the invention, - [Fig.2] [Fig.2] is a block diagram representing the operation of a computer program for implementing a fault detection method according to the invention, - [Fig.3] [Fig.3] represents a photovoltaic installation comprising a measuring device according to the invention, - [Fig.4] [Fig.4] shows a diagram of a fault detection device according to the invention, - [Fig.5] [Fig.5] illustrates the training of a machine learning algorithm and the enrichment of a training database, - [Fig.6] [Fig.6] is an example of fault detection according to the invention, - [Fig.7] [Fig.7] illustrates the acquisition of operating data from a panel, - [Fig.8] [Fig.8] illustrates the acquisition of meteorological data, - [Fig.9] [Fig.9] illustrates adjusted predictive models representative of the production of two determined healthy panels compared with the actual production values ​​of the healthy panels measured by means of a sensor.

[0061] In the remainder of the description, elements that are identical or have identical functions bear the same reference sign. Their description is not repeated with respect to each of the figures; only the main differences between the embodiments being mentioned. Detailed description Detection method

[0062] [Fig.l] illustrates the steps of a method 10 for detecting faults in a photovoltaic installation according to the invention.

[0063] A method 10 for detecting defects according to the invention comprises - the collection 100 of structural information 300 relating to the structure of the photovoltaic installation, meteorological data 220 and operating data 200; - the determination of an adjusted predictive model 102; - identification of the presence of defects 104, and possibly of types of defects, and - optionally the generation of a 106 report. Collection of data and information

[0064] The collection comprises the collection of structural information relating to the structure of the photovoltaic installation 300, the collection of meteorological data 220, and the collection of operating data of the photovoltaic installation 200.

[0065] The structural information 300 may comprise: a brand and / or an age of the photovoltaic installation and / or a material of the photovoltaic panels and / or a technology of the photovoltaic panels and / or an inclination of the photovoltaic panels and / or a topology of the photovoltaic installation and / or information relating to a replacement of photovoltaic panels and / or a number of panels per string of panels and / or a number of strings in the photovoltaic installation.

[0066] Preferably, the structural information 300 is entered by an operator, for example via an interface 30b, and / or extracted from a database 30a.

[0067] In particular, the structural information can be extracted from field reports produced for the photovoltaic installation and / or from technical data sheets of the photovoltaic installation.

[0068] Preferably, the meteorological data 220 comprises at least one irradiation and one ambient temperature, in the region of the photovoltaic installation. The meteorological data may comprise at least one irradiation, one ambient temperature and one wind speed.

[0069] [Fig.8] illustrates an example of meteorological data 220 composed of the acquisition of the temperature 220i, the acquisition of the irradiation 2202 and the acquisition of the wind speed 2203, within a radius of 100 m around a photovoltaic installation.

[0070] Preferably, the meteorological data is acquired by means of a weather station 22.

[0071] The meteorological data 220 are preferably acquired at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 1 kHz.

[0072] The operating data 200 preferably includes the voltage and / or current flowing in one or more strings of panels of the photovoltaic installation for which the presence of faults is to be detected.

[0073] Generally, the operating data 200 are chosen so as to allow the determination of the electrical production of the photovoltaic installation over time.

[0074] [Fig.7] illustrates an example of operating data 200 composed of acquisition of current intensities 200i, 2002, 2003, 2004 and a voltage 2005 representative of the operation of a string of photovoltaic panels.

[0075] The operating data 200 are preferably acquired at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 1 kHz.

[0076] In one embodiment, the operating data 200 and the meteorological data 220 are acquired at the same frequency.

[0077] The operating data 200 are preferably acquired by means of at least one analog sensor. The analog sensor may advantageously be configured for acquisition of the operating data at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 1 kHz, and / or less than or equal to 1 MHz.

[0078] The collection of operating and / or meteorological data can be carried out over a determined period of time, preferably continuously, greater than 24 hours.

[0079] The collection of operating and / or meteorological data can be carried out over a period of time, preferably continuously, of less than one week, or even less than 72 hours, or even less than 48 hours.

[0080] The collection of operating and / or meteorological data can be carried out over a period of time, possibly discontinuous, comprising at least one sunrise and / or one sunset.

[0081] The collection of operating and / or meteorological data can be carried out over a period of time, possibly discontinuous, during which it is not raining.

[0082] Preferably, upon completion of the collection, a set of normalization equations is applied to the operating and / or meteorological data, preferably to the operating and meteorological data.

[0083] Normalization is a step classically carried out in the field of data processing allowing data to be obtained in a particular format without loss of information.

[0084] For example, for a set of measurements in the set of real numbers, a normalization can consist of applying a set of equations so that the measurements ultimately belong to the interval [0; 1].

[0085] The normalization equations may depend on minimum and maximum threshold values ​​of the data, these threshold values ​​may be a function of the operating ranges of the sensors. For example, for a power density sensor operating over a range of values ​​[0; 1500 W / m2], the minimum and maximum threshold values ​​considered for the normalization equations may be defined as being 0 W / m2 and 1350 W / m2 respectively. Such a normalization method makes it possible to obtain a good distribution of the normalized data in the interval [0; 1].

[0086] This not only improves the visualization of data relating to the photovoltaic installation, but also improves the accuracy of fault detection algorithms. Determination of the adjusted predictive model

[0087] The determination 102 of an adjusted predictive model comprises the determination of an adjusted predictive production model 320 of the photovoltaic installation, from the structural information 300 and the meteorological data 220.

[0088] The objective of this step is to enable the analysis of any type of photovoltaic installation, regardless of the number of strings of panels it contains, the technology used, its orientation, its age, etc., and regardless of the environment in which it is located, by means of modeling capable of predicting the production of the photovoltaic installation whose faults are to be detected.

[0089] For this, one method consists of determining a production of a theoretical installation of the same type as the photovoltaic installation whose faults are to be detected, that is to say whose structural information is similar, preferably identical, the theoretical installation corresponding to an installation in which all the strings of photovoltaic panels are healthy and therefore operate at their maximum capacity.

[0090] Production can be expressed by a power.

[0091] The fitted predictive model can take the form of a power curve, as illustrated in Fig. 9b of [Fig.9].

[0092] The determination of the adjusted predictive model 320 can be carried out via the exploitation of knowledge-based models, for example by means of generative models as described in Theis, et al. "A note on the evaluation of generative models." arXiv preprint arXiv: 1511.01844 (2015), by means of domain adaptation learning models considering the invariant domain as described in Bak- tashmotlagh, M et al. “Unsupervised Domain Adaptation by Domain Invariant Projection” 2013 IEEE International Conference on Computer Vision, pages 769-776, 2013, or preferably using equations simulating the behavior of a photovoltaic installation as described in Herteleer, B., et al. “Normalized efficiency of photovoltaic Systems: Going beyond the performance ratio.” Solar Energy, vol. 157, pages 408-418, 2017.

[0093] The approach using predetermined equations simulating the operation of a photovoltaic installation allows in particular a high precision of prediction of the behavior of a photovoltaic installation and is also easily adaptable to the real data measured on any type of photovoltaic installation.

[0094] In one embodiment of the invention, a reference predictive model is determined from the structural information of the photovoltaic installation, representing the ideal operation of a photovoltaic installation comprising healthy panels and having the same intrinsic characteristics as those collected in the structural information relating to the photovoltaic installation, then from the reference predictive model, an adjusted predictive model is determined from the meteorological data, determining an ideal production profile of the photovoltaic installation under meteorological conditions as collected.

[0095] Preferably, the adjusted predictive model makes it possible to model the ideal operation of each of the chains of the photovoltaic installation. Ideal operation means fault-free operation of the installation.

[0096] In the absence of a defect, the adjusted predictive model therefore corresponds substantially to actual production.

[0097] [Fig. 9] illustrates a comparison between fitted predictive models (Fig. 9b) and operating data of fault-free photovoltaic panel strings (Fig. 9a), the fitted predictive models modeling the current intensity of the photovoltaic panel strings over time, and the operating data being data acquired during the actual operation of the photovoltaic panel strings. Fig. 9c is an overlay of the fitted predictive models and the operating data. In the example of [Fig. 9], for the panel strings with no faults, the current intensity curves representing the predictive models and the actual operating data are substantially identical.

[0098] The determination of the adjusted predictive model is implemented by computer means 32, the computer means preferably exploiting methods based on knowledge-based models.

[0099] The computer means comprise at least one computer program comprising lines of code which, when implemented by a computer, enable the determination of the adjusted predictive model. Defect identification

[0100] Identifying faults includes determining the presence of faults on the string(s) of the photovoltaic installation by subjecting the adjusted predictive production model, the meteorological data and the operating data to at least one pre-trained machine learning algorithm so as to identify one or more strings of panels comprising faults, called “defective string(s)”.

[0101] By submission of the adjusted production predictive model, meteorological data and operating data, it is meant submission of raw, pre-processed data and / or attributes extracted and / or selected from the raw or pre-processed data.

[0102] Preferably, the identification of the defects further comprises the determination of the types of defects detected.

[0103] The identification of the presence of defects, and possibly the identification of the types of defects, can be carried out by means of several machine learning algorithms. Preferably, at least a first machine learning algorithm identifies the presence of defects and then at least a second machine learning algorithm identifies the type of defect.

[0104] If the defect cannot be identified, a manual identification step, preferably performed by a qualified technician, may be performed. Preferably, after such a manual identification step, the machine learning algorithm is enriched so as to allow for the integration of future detection of such a defect.

[0105] Preferably, the at least one second machine learning algorithm is a supervised machine learning algorithm.

[0106] One type of defects may be snail tracks, broken glass, delamination, corrosion, yellowing, solar cell microcracks, solar cell cracks, potential induced degradation, defective junction box, backsheet cracks.

[0107] Generally, a machine learning algorithm is trained from a training database DB so as to learn to be able to identify, from input data, output data.

[0108] For training a supervised machine learning algorithm, the training database DB comprises an input data set and an output data set. For training an unsupervised machine learning algorithm, the training database DB comprises at least one input data set.

[0109] Training a machine learning algorithm involves providing at least part of the input data set and possibly at least one part of the output set of data from the training database DB, so that the algorithm learns to determine, from an input data, an output data.

[0110] In the present case, an input data comprises an adjusted production predictive model, meteorological data and operating data, or attributes extracted from said operating and meteorological data, raw or pre-processed, and an output data corresponds to the detection of the presence of defects on a chain and possibly the determination of the type of defect detected on the chain.

[0111] Preferably, the input data are preprocessed. Conventionally, the operating and / or meteorological data are at least normalized.

[0112] The input data may be decomposed, for example, into time windows and / or attributes may be extracted and / or selected from the operating data and / or the meteorological data.

[0113] In particular, attributes may be extracted from the operating and / or meteorological data, for example by applying an attribute extraction algorithm, in particular via a decomposition of the multi-resolution operating data followed by an extraction of statistical attributes.

[0114] In particular, a multi-resolution decomposition allows a temporal and frequency analysis of the operating data.

[0115] The attributes can be chosen in particular from: - an asymmetry of the data compared to the mean of said data; - a flattening measuring the peak of a probability distribution of the data; - a variance of the data; - a peak-to-peak distance, corresponding to the distance between the highest amplitude peak and the lowest amplitude valley; and - energy calculated from the data.

[0116] This list is for information purposes only and is not exhaustive.

[0117] Preferably, when the data is decomposed into time windows, at least one machine learning algorithm is constructed for each of the time windows.

[0118] Such a decomposition makes it possible in particular to analyze the behavior of photovoltaic panels under different irradiation conditions of the day, in particular: morning, midday, afternoon and night.

[0119] An extraction of attributes can be performed in each of these windows.

[0120] Machine learning can be performed using an algorithm using k-Nearest-Neighbors (kNN) algorithms, support vector machines, or other algorithms. in English, also called “SVM”) and / or decision trees (“Decision Trees” in English, or “DT”).

[0121] A combination of these algorithms with relative majority voting can be performed.

[0122] “k-Nearest-Neighbors”

[0123] The kNN is one of the most widely used models for classification due to its simplicity. For a given input sample x' = |x / ', ..., %„ / ] with attributes to classify, the kNN finds its nearest neighbors among the already classified samples based on a distance metric, usually the Euclidean distance.

[0124] A kNN assigns the input sample to the most common class among its k nearest neighbors.

[0125] As the value of k increases, the kNN model can tolerate more noise because the impact of variance caused by random error is reduced, but there is a risk of missing a small but important pattern in the data. The key to choosing an appropriate value of k is to find a balance between overfitting and underfitting.

[0126] “Vector Machines Support”

[0127] SVMs are among the most powerful and complex classification algorithms. The SVM classifier searches for one or a set of optimal separating hyperplanes, maximizing the margin between classes constituting the outputs. The SVM is based on the idea that it is possible to map an input space into a higher-dimensional attribute space via a kernel function, and then apply a linear SVM on this space, thus making it possible to distinguish classes that were not linearly separable.

[0128] “Decision Trees”

[0129] The DT model is characterized by decision nodes having several branches, and leaf nodes corresponding to the outputs of the decision nodes. DTs are created from a recursive splitting of the sample set based on a set of splitting rules related to the attributes.

[0130] There are many variants of DT in the literature, such as Iterative Dichotomies 3 (ID3), ID3 Successor (C4.5), Chi-Square Automatic Interaction Detector (CHAID), Classification and Regression Tree (CART). This list is not exhaustive.

[0131] Majority vote

[0132] In a particular embodiment, the pre-trained machine learning algorithm to identify one or more chains of panels comprising defects comprises the implementation of several classifiers, for example the three classifiers kNN, SVM and DT, combined using the principle of majority voting.

[0133] Each of the classifiers can in fact be considered as a voter, the prediction results, i.e. the outputs, of each classifier being compared to determine a final prediction.

[0134] The principle of majority voting can be a weighted majority vote, a relative majority vote or an absolute majority vote.

[0135] Once trained, the algorithm can be used to predict an output, here a detection of the presence, and possibly the type, of faults present on one or more string(s) of photovoltaic panels of a photovoltaic installation, from operating data, meteorological data, and / or attributes extracted from this data, possibly for a time window, and from the adjusted predictive model.

[0136] For each subsequent operation of the training of the machine learning algorithm, the input data E and the output data S can be recorded in the database DB, as shown schematically in [Fig.5].

[0137] The machine learning algorithm can be updated, thanks to the enrichment of the DB database.

[0138] Preferably, output data S are checked by an operator before being added to the database DB. Generating a report

[0139] The method preferably comprises the optional step of presenting a 360 report indicating the defective chain(s), and preferably the types of defects.

[0140] In a preferred embodiment of the invention, the method comprises determining one or more maintenance actions, the maintenance actions being determined from the identification of the presence of defects on the defective line(s), and optionally from the adjusted predictive production model and operating data relating to each defective line, preferably by means of a statistical analysis, preferably by means of a residual-based method.

[0141] For example, the determination of the maintenance actions may comprise: - the determination of a power threshold value, the calculation of a centroid of the power of the healthy panel strings or of a centroid of a theoretical power for a panel string determined for example from the adjusted predictive model, the healthy panel strings being the strings for which no fault has been identified; - for each defective panel string, comparing the power of the defective panel string with the power threshold value in order to determine an amplitude of the difference between the power of the defective string and the power threshold value; the maintenance priority of defective panel strings being determined by the magnitude of the difference (the greater the magnitude, the higher the maintenance priority).

[0142] The threshold value may be a centroid of the power of the healthy panel strings, the healthy panel strings being the panel strings for which no faults have been identified.

[0143] The threshold value may be a centroid of the theoretical powers calculated for each of the strings of panels in the installation, the theoretical powers being able to be determined from the adjusted predictive model.

[0144] In one embodiment, the threshold value is a centroid of the theoretical powers and the powers of the healthy panel strings. Devices Computer program

[0145] [Fig.2] shows a diagram of a computer program 1 comprising instructions which, when the program is executed by a computer, cause the latter to implement a fault detection method according to the invention.

[0146] The program includes a prediction module 32 which makes it possible to determine the adjusted predictive production model 320, from the collected meteorological data 220 and the collected structural information 300.

[0147] The program comprises a detection module 34 which allows the identification of the presence of defects on the string(s) of the photovoltaic installation and preferably the types of defects 340, from the adjusted predictive production model 320, meteorological data 220 and operating data 200 and / or attributes extracted from said data 220;200. The detection module comprises at least one pre-trained machine learning algorithm so as to identify one or more strings of panels comprising defects, called “defective string”, and preferably the types of defects detected.

[0148] The program 1 may comprise a module for generating a report 36 determining maintenance actions, based on the identification of the presence and possibly the types of defects 340, and possibly the adjusted predictive production model 320, and / or the operating data 200.

[0149] The generation module 36 preferably allows the presentation of the 360 ​​report to an operator, for example, by sending the 360 ​​report to a graphical interface 38, such as a computer screen, a mobile phone, a tablet or any other well-known display means. Photovoltaic installation

[0150] [Fig.3] illustrates a photovoltaic installation 4 comprising strings 40 of photovoltaic panels each formed from one or more photovoltaic panels 42. The strings 40 of photovoltaic panels being connected to a junction box 44.

[0151] The photovoltaic installation conventionally comprises an inverter 46.

[0152] In [Fig.3], a measuring device 2 for collecting meteorological and operating data is positioned on the photovoltaic installation 4.

[0153] The invention also relates to a photovoltaic installation 4 comprising such a measuring device 2 according to the invention. Measuring device

[0154] A measuring device 2 according to the invention comprises at least: - a weather station 22 for the acquisition of meteorological data 220, and - an operating data measurement module 20, in particular voltage and current, the measurement module being configured to acquire the data at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 1 kHz, and to be connected in series between at least part of the strings 40 of panels of the photovoltaic installation, preferably all the strings of panels, and the junction box 44 so as to acquire the operating data 200 for each string of panels connected to the measurement device.

[0155] In a preferred embodiment of the invention, the measurement module 20 comprises one or more analog sensor(s), in particular for measuring the voltage and intensity for each string of panels of the photovoltaic installation.

[0156] The measuring device advantageously comprises connection means 24, preferably deactivatable, of the measuring module 20 with the chains 40 of photovoltaic panels.

[0157] The connection means 24 may comprise means for connecting the measurement module 20 to the weather station 22, the connection means being able to comprise wireless and / or wired communication means.

[0158] The connection means 24 may comprise means for connecting the measurement module 20 and / or the weather station 22, with a battery 26.

[0159] Preferably, the measuring module 20 and the weather station 22 are configured so as to be electrically powered independently of each other.

[0160] The battery advantageously provides autonomy to the measuring device 2. Fault detection device

[0161] [Fig.4] is a block diagram of a fault detection device 3 according to the invention.

[0162] A fault detection device 3 comprises: - a measuring device 2 according to the invention; - a computer program 1 for implementing a method of detecting defects 10 according to the invention.

[0163] A fault detection device may comprise one or more features of a computer program 1 for implementing a detection method as described previously. Examples

[0164] [Fig.6] illustrates an example of classification for determining the presence of the “snail track” type defect on a photovoltaic panel in a string of photovoltaic panels.

[0165] The operating data 200 are presented in the form of a time curve. The curve illustrates the current intensity, in amperes, of a panel during one day. The operating data 200 are divided into 4 time windows 200a, 200b, 200c and 200d.

[0166] For each time window, three classifiers are applied to the operating data from panel chains that have been previously detected as defective.

[0167] Submitting the operating data to the three classifiers allows the types of faults to be determined.

[0168] In this case, a kNN, an SVM and a DT are implemented to determine the presence of a defect of the “snail tracks” type or the absence of a defect. In the example, if the classifier identifies the presence of a defect, its output is “snail tracks” while in the absence of a defect, its output is “healthy”.

[0169] A majority vote is then implemented, based on the outputs of the classifiers, for each of the time windows. A prediction of the presence of a defect is then obtained for each of the time windows.

[0170] Subsequently, an absolute majority vote is applied to the predictions obtained for each time window, determining whether the string of panels for which the operating data was acquired is defective or not. In the example, at least one panel in the string of panels has a snail track type defect.

[0171] As is clearly apparent, the invention allows for more precise, rapid and precise detection than conventional methods.

[0172] Furthermore, the invention does not require production shutdown, the data being able to be acquired during operation of the photovoltaic installation.

[0173] Furthermore, it is not necessary to have the complete history of the photovoltaic installation to implement the fault detection method according to the invention, Unlike many well-known methods, which exploit the history of the panels in order to precisely model their aging. Unlike these methods which can only be applied with very precise knowledge of the photovoltaic installation, the proposed method makes it possible to diagnose faults on any type of installation, without the need for an exhaustive history.

[0174] Obviously, the invention is not limited to the embodiments described.

[0175] In particular, once a fault has been detected at the level of a string of panels, the person skilled in the art would have no difficulty in adapting the method as described to apply it no longer to the photovoltaic installation to detect a fault at the level of a string of panels but to the string of panels to detect on which panel a fault is present.

[0176] Other types of machine learning algorithms could also be implemented, in combination or alone.

Claims

Claims

1. Method for detecting faults (10) in a photovoltaic installation comprising several strings of photovoltaic panels, the method comprising the following steps: - collecting: - structural information (300) relating to the structure of the photovoltaic installation, - meteorological data (220); and - operating data (200) for each string of photovoltaic panels of the photovoltaic installation; - from the structural information and the meteorological data, determining an adjusted predictive production model of the photovoltaic installation, said adjusted predictive production model modeling an optimal theoretical electrical production of the photovoltaic installation with regard to the meteorological data, the adjusted predictive production model being determined assuming that the photovoltaic installation does not have a fault;- identifying the presence of faults on each of the strings of the photovoltaic installation by subjecting the adjusted predictive production model, the meteorological data and the operating data to at least one pre-trained machine learning algorithm so as to identify each string of panels comprising faults, called "defective string", and preferably the types of faults; - optionally, presenting a report indicating the defective string(s), preferably the types of faults, and preferably one or more maintenance actions, the maintenance actions being determined from the identification of the presence of faults on the defective string(s), and possibly from the adjusted predictive production model and the operating data relating to each defective string, preferably by means of a statistical analysis, preferably by means of a residual-based method.;

2. The detection method of claim 1, wherein the meteorological data comprises at least one irradiation, and one ambient temperature.

3. A detection method according to any preceding claim, wherein the meteorological data and / or the data operating, preferably the meteorological data and the operating data are acquired at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 1 kHz.

4. A detection method according to any one of the preceding claims, wherein the structural information comprises: a brand and / or an age of the photovoltaic installation and / or a material of the photovoltaic panels and / or a technology of the photovoltaic panels and / or an inclination of the photovoltaic panels and / or a topology of the photovoltaic installation and / or information relating to a replacement of photovoltaic panels and / or a number of panels per string of panels and / or a number of strings in the photovoltaic installation.

5. A detection method according to any preceding claim, wherein the collection of operating data is performed by means of at least one analog sensor.

6. A detection method according to any preceding claim, wherein the collection of structural information is performed by means of a graphical interface and / or by means of a database.

7. A detection method according to any preceding claim, wherein the operating data is divided into time windows, the identification of the presence of defects being carried out for each of the time windows, at least one machine learning algorithm being pre-trained for each of said time windows.

8. A detection method according to the preceding claim, wherein the operating data is divided into time windows.

9. A computer program (1) comprising instructions which, when the program is executed by a computer, cause the computer to implement a fault detection method according to any one of the preceding claims.

10. Measuring device (2) for collecting meteorological (220) and operating (200) data in a photovoltaic installation composed of one or more strings of photovoltaic panels connected to a junction box, the device comprising: - a meteorological station (22) for acquiring meteorological data, and - a measuring module (20) for operating data, in particular voltage and current, the measurement module being configured to acquire the data at a frequency greater than or equal to 1 Hz, preferably greater than or equal to 1 kHz, and to be connected in series between at least part of the strings of panels of the photovoltaic installation, preferably all the strings of panels, and the junction box so as to acquire the operating data for each string of panels connected to the measurement device.

11. Measuring device according to claim 10, wherein the measuring module comprises at least one analog sensor.

12. Device (3) for detecting faults in a photovoltaic installation comprising several strings of photovoltaic panels connected to a junction box (44), the device comprising: - a measuring device (2) for collecting meteorological and operating data according to any one of claims 10 and 11, and - a computer program (1) according to claim 9.

13. Detection device according to claim 12, comprising means for acquiring structural information from the photovoltaic installation, preferably the acquisition means being a graphical interface (30a; 30b).

14. Detection device according to claim 13, in which the installation comprises at least 8, better still at least 15, better still at least 20 chains of panels (40).

15. Photovoltaic installation (4) comprising several strings of photovoltaic panels (40) in parallel connected to a junction box (44), in which a fault detection device (3) according to claim 12 is connected in series between the strings of panels and the junction box.

Images