Method for determining at least one system parameter of a photovoltaic system

Abstract

The present invention relates to a computer-implemented method for determining at least one system parameter that represents an expected production rate and / or failure probability over at least a first period of at least one photovoltaic system that comprises at least one photovoltaic module or a plurality of photovoltaic modules. The invention also relates to a data processing device, a computer program product and a computer-readable (storage) medium for carrying out the method according to the invention.

Description

[0001] METHOD FOR DETERMINING AT LEAST ONE SYSTEM PARAMETER OF A PHOTOVOLTAIC SYSTEM

[0002] DESCRIPTION

[0003] The present invention relates to a computer-implemented method for determining at least one system parameter representing an expected production output, system availability and / or failure probability over at least a first period of at least one photovoltaic system comprising at least one photovoltaic module or a plurality of photovoltaic modules, as well as a device for data processing, a computer program and a computer-readable (storage) medium for carrying out this method.

[0004] There is a general effort to increase the share of electricity generated from renewable energy sources. Due to the inconsistent availability of renewable energy sources, grid operators must be able to plan in advance which portions of the required energy can be provided by renewable energy sources and which must be supplemented by alternative energy sources in order to cover the demand at a given time and avoid grid overloads.

[0005] In the case of photovoltaic systems, such forecasts are essentially based on an analysis of the expected solar radiation based on corresponding weather forecasts at the installation location of the respective photovoltaic system in combination with a target efficiency, i.e. the relationship between the nominal power output by the photovoltaic system in relation to the light output hitting the photovoltaic system. The orientation of the photovoltaic system in relation to the position of the sun in terms of azimuth and declination angles can also be taken into account. The assumed degree of aging is conventionally stored and calculated using a theoretical expected value that is based either on an empirically derived value or on the manufacturer's guarantee. In reality, however, aging depends on various other factors and can be represented as a quotient, e.g.the quotient of actual output power and target output power or the quotient of actual output power and nominal output power or otherwise derived quotients.

[0006] Contrary to theoretical assumptions, experience shows that photovoltaic systems are increasingly exhibiting either technical defects or are affected by unforeseen aging parameters, which can, on the one hand, compromise the security of supply with solar power and, on the other, negatively impact the service life of the PV systems. Often, significant aging or degeneration processes occur, usually unanticipated, leading to a significant change in efficiency, resulting in a discrepancy between the actual output power and the target output power, and even leading to the failure of the photovoltaic system. An example of aging or degeneration processes is the so-called PID effect (PID = potential induced degradation), which can lead to a significant reduction in the performance of photovoltaic modules.In many cases, this effect leads to reduced performance that is only partially reversible or even completely irreversible. The technical defects mentioned above are partly related to new technological developments, but partly also to the use of raw materials or semi-finished products that are not manufactured according to specifications (e.g., photovoltaic cells, ethylene vinyl acetate (EVA) films, glass, backsheets). These defects lead to a reduction in the resource efficiency of photovoltaic technology, as system components often no longer deliver a sufficient amount of energy before the end of their calculated service life – and often even before the end of the warranty period or even before the end of the ecological payback period, especially at the point at which the photovoltaic module recovers more resources, especially energy, than was required for its production – thus reducing system availability.

[0007] This reduces the reliability of long-term planning of the energy supply or the planning of the amount of energy fed into the grid by the photovoltaic systems.

[0008] It is therefore an object of the present invention to provide a method by means of which the long-term forecasting of the system availability for photovoltaic systems, the output power to be expected from the photovoltaic systems or the failure probability of the photovoltaic systems is improved.

[0009] This object is achieved according to the invention by a computer-implemented method for determining at least one system parameter representing an expected production output, system availability and / or failure probability over at least a first period of at least one photovoltaic system comprising at least one photovoltaic module or a plurality of photovoltaic modules, wherein the method comprises at least one simulation of the development of the efficiency of the photovoltaic modules and / or the photovoltaic system, and wherein the simulation comprises at least one of the following steps:

[0010] • Determination of at least one parameter related to the photovoltaic system, at least one inverter, at least one element of the photovoltaic system and / or the photovoltaic modules,

[0011] • Determination of at least one expected temporal course of the parameter over the first period or a specific or indefinite period,

[0012] • Determination of at least one expected progression of the degree of aging of the photovoltaic system and / or the photovoltaic modules based on the expected temporal progression of the parameter,

[0013] • Determination of at least one expected time course of an output power of the photovoltaic system and / or the photovoltaic modules based on the expected time course of the degree of aging,

[0014] • Determination of at least a probability with which a power output can be expected

[0015] • Determination of an expected efficiency over the first period based on the expected time course of the output power, and

[0016] • Calculation of the system parameters based on the expected efficiency.

[0017] It is particularly preferred that the parameter is based on at least one developer parameter, at least one developer, at least one hardware component parameter, in particular of the photovoltaic modules and / or the inverters, at least one environmental parameter, at least one planning parameter, in particular comprising information on the design and / or engineering, at least one installer parameter, in particular comprising and / or formed from at least one parameter of a project participant, preferably at least one installer, at least one developer, at least one manufacturer, at least one installer, at least one planner, at least one purchaser, at least one construction company and / or at least one general contractor, at least one assembly parameter, at least one operator parameter, at least one maintenance parameter,at least one operating parameter and / or at least one process parameter and / or a combination of at least two, preferably a plurality of the aforementioned parameters.

[0018] In the aforementioned embodiment, it is particularly preferred that the developer characteristic is determined based on the developer's experience in the development of photovoltaic systems, based on the operating time of planned systems, based on a certification of the developer, based on monitoring and / or reporting options of the developer, and / or the developer's business processes in terms of quality management.

[0019] The invention also proposes that the planning parameter is based on a certification of the planner and / or purchaser, based on a size of the photovoltaic system, based on an application area of ​​the photovoltaic system, such as industrial use, commercial use and / or private use, based on an external examination of the planner, based on proven qualifications of the planner, based on a self-disclosure of the planner, based on a comparison of the planning of the photovoltaic system and / or photovoltaic modules and / or inverters and / or other components or subcomponents, in particular photovoltaic cells and / or connectors, with standards and / or norms and / or "best practice guidelines", preferably based on the best available technology (BAT or BAT = Best Available Technology), based on a number, size, installation climate zone,site-specific environmental conditions and / or quality, in particular with regard to the solidity and / or performance indicators of the photovoltaic systems planned by the planner in the past, based on guarantees and / or warranties granted by the planner, based on the frequency of service and / or warranty claims against the planner, based on the planner's insurance parameters, such as insurer, type of insurance, amount of coverage, scope of coverage, deductible, no-claims discounts, insurance exclusions and / or other insurance conditions, based on the planner's monitoring and / or reporting capabilities, based on a comparison with at least one environmental parameter and / or a comparison of a climate zone with the environmental and / or climate zone-specific suitability and / or evaluation of the planned components and / or based on a comparison of the geological data.

[0020] Furthermore, it is preferred that the hardware component characteristic is based on at least conformity of the photovoltaic system, the photovoltaic modules and / or the hardware components and / or software components of the photovoltaic system and / or the photovoltaic modules with norms and / or standards, based on at least one specification of materials and / or components used for the production of the hardware components, preferably the photovoltaic modules, the inverters and / or at least one other component, software components and / or photovoltaic system, such as manufacturing data of the raw materials, in particular silicon, wafer, cell, cell connector, glass and / or encapsulation materials, manufacturing data of junction boxes and / or connectors and / or photovoltaic module manufacturing, based on resistance to environmental influences, in particular including climate-related influences, such as temperature fluctuations,maximum and minimum absolute temperature exposure, humidity and / or dryness, solar radiation, in particular including UV radiation components, precipitation, hail, snow, pollutants, in particular of any aggregate state, such as air pollutants, and / or salinity of the ambient air, based on material combinations used, preferably also, but not only, in comparison to certified material combinations, based on circuit technologies used, such as individual inverter configurations, in particular with or without optimizer, string inverter configurations, in particular with or without optimizer, central inverter configurations, in particular with or without optimizer), based on certifications of the manufacturer and / or manufacturers of the hardware components, the software components, the photovoltaic system and / or the photovoltaic modules, based on quality ratings of the hardware components, in particular the photovoltaic modules,the inverter and / or at least one other component and / or,

[0021] Software components of the photovoltaic system, based on guarantees and / or warranties granted, in particular by the manufacturer, for the hardware components and / or software components of the photovoltaic system and / or the component, in particular the photovoltaic modules, the inverters and / or at least one other component, as well as their exclusions, in particular as far as made available, based on at least one frequency of guaranteed performance and / or warranty claims against the manufacturer of hardware components, in particular the photovoltaic modules, the inverters and / or at least one other component and / or software components of the photovoltaic system and / or based on an economic success and / or economic development of the manufacturer of the hardware components, in particular the photovoltaic modules, the inverters and / or at least one other component and / or software components of the photovoltaic system,based on an examination of the manufacturing operation of hardware components, in particular the photovoltaic modules, the inverters and / or at least one other component and / or software components of the photovoltaic system, based on a comparison with at least one environmental parameter and / or a climate zone in comparison to their suitability and / or based on a comparison of the geological data and / or weather data in comparison to the suitability for the operation of a photovoltaic system and / or based on business processes, such as quality management.

[0022] The invention also proposes that the environmental parameter is based on

[0023] • at least one installation type and / or at least one installation location of the photovoltaic system, such as ground mounting, installation integrated into agricultural land, in particular agrovoltaics, roof mounting, installation in textile architecture, building-integrated photovoltaics, floating on-water installation, floating photovoltaics, extraterrestrial installation and / or floating installation on a vehicle, in particular aircraft, land vehicle and / or watercraft, and / or based on lighter-than-air technology (LTA)

[0024] • at least one alignment of the photovoltaic system and / or the photovoltaic modules, in particular tracking the position of the sun, preferably single- or multi-axis

[0025] • at least one inclination of the photovoltaic modules, in particular with the position of the sun, preferably single- or multi-axis tracking

[0026] • at least one geographical location of the photovoltaic system, in particular for determining the climate zone, based on at least one component of the probability of precipitation, probability of hail, probability of snow, probability of wind strength, probability of shading, probability of pollution, based on expected solar radiation data, in particular including UV radiation components, temperature fluctuations, min. and / or max. temperature exposure and / or min. and / or max. humidity data, air pollutants and / or salinity of the ambient air at the location of the photovoltaic system, in particular a future development of one, several or all of these probabilities;

[0027] • at least a distance of the photovoltaic system from industrial plants, agricultural operations, salt water bodies, in particular seas and / or oceans and / or salt accumulations, in particular salt lakes, salt works, salt mountains and / or salt pans) and / or from other natural or man-made sources of increased pollutant concentrations, preferably corrosive species, in particular in the air and / or in the water

[0028] • at least one degree of environmental protection of the surroundings of the photovoltaic system, in particular by surrounding buildings or technical facilities, preferably antennas, vegetation and / or natural features

[0029] • at least one parameter relating to the foundation of the photovoltaic system,

[0030] • at least one parameter relating to drainage of the photovoltaic system, at least one parameter relating to the degree of infrastructure in the region of the photovoltaic system, in particular the distance of the photovoltaic system to supply facilities for the procurement of spare parts and / or to locations for maintenance personnel and / or

[0031] • at least one parameter of the maintenance of the photovoltaic system is determined.

[0032] Furthermore, the invention proposes that the installer parameter is based on historical data, in particular on the qualification, preferably of employees, on claims, on complaint handling, preferably based on testimonials, certificates, evaluations, profit margins, efficiency, production output, specifications, documentation and / or protocols, preferably comprising reports and / or photo documentation, and / or reliability of existing systems of the installer and / or project participants, based on a number, size and / or type, preferably rooftop systems, field systems with or without sun tracking, of the photovoltaic systems installed in the past by the installer and / or project participants, based on insurance parameters of the installer and / or project participants, such as insurer, type of insurance, coverage amount, scope of coverage, deductible, no-claims discounts,Insurance exclusions and / or other insurance conditions, based on an external quality assessment of the installer and / or project participant and / or at least one subcontractor of the installer and / or project participant, based on the conformity of a setup concept with safety, electrical, and / or construction norms and / or standards, based on guarantees and / or warranties granted by the installer and / or project participant, based on the frequency of service and / or warranty claims against the installer, based on the quality assessment procedures used by the installer for hardware and / or software components installed in the photovoltaic system - in the qualification phase and / or during the implementation phase, and / or based on the business processes within the meaning of the quality management of the installer and / or project participant.

[0033] A method according to the invention can be characterized in that the installation parameter is determined based on at least one comparison of the planned photovoltaic system with the installed photovoltaic system, and / or based on business processes in the sense of the quality management of the installation company.

[0034] It is also preferred that the operator characteristic is based on historical data on damage cases, causes of damage and / or their rectification, based on profit margins, efficiency, production output and / or operator concept of existing systems of the operator, based on a number, size, availability and / or other performance indicators, in particular comprising a relationship between output and forecast power and / or output in relation to the measured solar irradiation, mean time between maintenance and / or mean time between system shutdowns of the photovoltaic systems operated by the operator, based on insurance parameters of the photovoltaic system, such as insurer, type of insurance, coverage amount, scope of coverage, deductible, no-claims discounts,Insurance exclusions and / or other insurance conditions and / or based on the business processes in terms of the operator's quality management. Furthermore, the invention proposes that the maintenance parameter is determined based on an examination of a maintenance plan and / or monitoring system, in particular a comparison of the maintenance plan and / or monitoring system with standards, norms and / or best practices, based on historical data on damage cases of the photovoltaic systems maintained by the person maintaining the photovoltaic system, based on profit margins, efficiency, production output, operating times and / or causes of damage and / or their rectification, existing systems of the person maintaining the system, based on a number, size, concept and / or the photovoltaic systems maintained by the person maintaining the system, based on processes and / or response times of the person maintaining the system, based on insurance parameters of the person maintaining the system, such as insurer, type of insurance,The sum insured, scope of coverage, deductible, no-claims discounts, insurance exclusions and / or other insurance conditions are determined based on an external quality audit of the maintenance company and / or at least one of the maintenance company's subcontractors, based on the maintenance company's response times, based on a minimum uptime guarantee given by the maintenance company, and / or based on the business processes within the meaning of the maintenance company's quality management, in particular that of a maintenance company.

[0035] It is also preferred that the operating parameter is based on a degree of completeness of the documentation necessary for the design, planning, purchasing of components, installation, operation and / or maintenance of the photovoltaic system, based on an examination and / or qualified self-declaration of data sheets of the materials, components, hardware components used in the photovoltaic system, in particular the photovoltaic modules, inverters and / or at least one other hardware component and / or software components, based on an examination and / or qualified self-declaration of design plans, wiring plans and / or diagrams, based on an examination and / or qualified self-declaration of handling and / or installation protocols, based on the examination and / or qualified self-declaration of acceptance protocols,based on a test and / or qualified self-declaration of the conformity of the implemented photovoltaic system with the planning of the photovoltaic system, based on a test and / or qualified self-declaration of the supporting structure, the hardware components, in particular comprising the photovoltaic modules, the inverters and / or at least one further hardware component, the software components, and / or the design, based on at least one test and / or qualified self-declaration for hotspots and / or based on at least one electroluminescence measurement, preferably for cracks, breaks, defects, inactive cells and / or cell parts and / or other suitable criteria, in particular using an infrared measuring method, an electroluminescence recording device and / or at least one other suitable measuring or documentation device,based on at least one test and / or qualified self-declaration of the photovoltaic system, at least one photovoltaic module, at least one hardware component, based on a comparison of the serial number of the hardware components with the installed position and / or the installation location of the installed photovoltaic system, in particular including a position planned for installation and / or geodesics, based on a test of the traceability of the supply chain for the hardware components, in particular the photovoltaic modules, the inverters and / or at least one other hardware component, and / or the software component and / or based on the quality of the connections and / or links of the hardware and software components and / or the quality of the substructure, in particular including at least one tracking system, preferably for stability, susceptibility to corrosion,Natural frequency and / or other essential quality characteristics and / or the quality of the cabling, in particular based on a quality of the cable specification and / or a quality of a laying.

[0036] Furthermore, in an additional embodiment, the invention proposes that the determination of at least one variable and / or a temporal progression of the variable is carried out using a neural network, using artificial intelligence, using machine learning and / or using empirically and / or empirically and / or assumed values ​​of derived links, wherein the variable is one or more variables selected from the group comprising the characteristic variable, the degree of aging, the power output, the efficiency, the developer characteristic variable, the hardware component characteristic variable, the manufacturer's characteristic variable of hardware or software components, the planning characteristic variable, the environmental characteristic variable, the operator characteristic variable, the installer characteristic variable, the maintenance characteristic variable, the assembly characteristic variable, the operating characteristic variable and / or other relevant characteristics in relation to the production of hardware and / or software components,on the construction and / or operation of a photovoltaic system. In the aforementioned embodiment, it is preferred that training data be used to determine the variables by means of the neural network, artificial intelligence, or machine learning. This training data preferably includes data on damage cases, insurance data, the development of the system parameters, the variables, the characteristics and / or the performance data and / or the status data of the photovoltaic system, specifically in comparison with the parameters previously determined for other systems and the actual parameters, preferably for existing photovoltaic systems.

[0037] It is also proposed that the use of a neural network, the use of artificial intelligence, the use of machine learning and / or the use of empirically and / or empirically assumed values ​​of links derived from the use of at least one dimensionality reduction, at least one principal component analysis, PCA (principal component analysis), at least one clustering, at least one classification, at least one multivariate regression analysis, at least one, preferably statistical, time series analysis, and / or at least one evaluation criterion with regard to conformity with relevant norms, standards and / or best practices, in particular for the processing and / or editing of training data.

[0038] A method according to the invention can be characterized in that the hardware components comprise at least one photovoltaic module, at least one photoactive cell for converting photons into electrons or vice versa, at least one inverter, at least one converter, at least one connecting line, at least one cabling, at least one distribution box and / or at least one device for merging and / or separating and / or distributing cables and / or cable harnesses, at least one plug connection, at least one holding element, at least one supporting element, at least one supporting structure, at least one tracking system (“tracker system”), at least one drive unit, in particular comprised by the tracker system, at least one control unit, at least one sensor, at least one light sensor, at least one precipitation sensor, at least one sun position sensor and / or at least one optimizer,preferably for at least one of the photovoltaic modules and / or operatively connected to the photovoltaic module. Furthermore, it is preferred that the software component comprises at least one control software, at least one tracker software, at least one inverter software, and / or an operating and / or maintenance software.

[0039] Furthermore, the invention proposes that the first period of time is determined by at least a first time and / or start time and a second time and / or end time.

[0040] It is also preferred that the first point in time is the start of a preliminary planning phase, the second point in time is the end of the preliminary planning phase, the end of a planning phase, the end of a development phase, the end of a planning phase for the installation of the photovoltaic system, the start of the assembly phase and / or the Notice to Proceed (NTP) point in time, the end of an assembly phase, the start of a commissioning phase, the end of a commissioning phase, the end of the preliminary and / or final acceptance phase, in particular the preliminary and / or final commissioning, the certification, a point in time during regular operation, a point in time during a change of ownership and / or after completion of a maintenance phase orRepair phase of the photovoltaic system, the first point in time is the commissioning of the photovoltaic system and the second point in time is the end of a predetermined start-up phase of the photovoltaic system, the first point in time is the commissioning of the photovoltaic system and the second point in time is the end of a predetermined simulation interval, the first point in time is the end of a previous simulation interval and the second point in time is the end of the predetermined simulation interval, the second point in time is the point in time of a change in the operator of the photovoltaic system, the second point in time is the point in time of a change in insurance for the photovoltaic system, and / or the first point in time is the point in time of a new financing of the photovoltaic system and / or the point in time is the point in time of a partial or complete revision of the photovoltaic system, in particular a repowering.

[0041] The invention also proposes that the system parameter is determined based on a system parameter determined for a preceding second period, the system parameter for the second period is compared with the system parameter determined for the first period, the system parameter for the first period is determined at least partly based on a modification of the system parameter and / or the

[0042] Operating parameter, maintenance parameter and / or at least one

[0043] Maintenance parameter is determined for the second period, wherein preferably the first period and the second period have approximately identical lengths, in particular one year, the second period is shorter than the first period or the second period is longer than the first period and / or the interval between the determinations is arbitrarily determined and / or the evaluation of the parameters of a photovoltaic system is determined and specified for the first time after any operating period.

[0044] Furthermore, the invention provides a device for data processing comprising at least one processor which is configured to carry out a method according to the invention.

[0045] Furthermore, the invention provides a computer program and / or evaluation program comprising instructions which, when the program is executed by at least one computer, cause the computer to carry out a method according to the invention.

[0046] Finally, the invention provides a computer-readable (storage) medium comprising instructions which, when executed by at least one computer, cause the computer to carry out a method according to the invention.

[0047] The invention is based on the surprising finding that by including additional parameters relating to the photovoltaic system, the predictability of system availability, failure probability and / or production output of a photovoltaic system can be significantly improved compared to an exclusive consideration of a target efficiency of the photovoltaic system in combination with corresponding weather forecasts.

[0048] Furthermore, the method according to the invention also makes it possible to provide an improved forecast for the profitability of the photovoltaic system based on the system parameters determined by the method. In particular, it is possible to determine which system components are likely to require replacement or repair before the end of their calculated service life, thereby optimizing additional necessary investments.

[0049] Furthermore, the method makes it possible to determine the probability of a lack of availability of the photovoltaic system based on the specific system parameters, for example due to technical failure due to aging processes or due to at least partial destruction by environmental influences such as hail damage, wind damage, or the like. By incorporating various parameters such as a developer parameter, a planning parameter, an environmental parameter, a location parameter, particularly indicative of climate zones, special weather zones, a hardware component parameter, an environmental parameter, an installer parameter, an assembly parameter, an operator parameter, a maintenance parameter, and / or an operating parameter, the method according to the invention makes it possible to provide more reliable information about the expected output power in relation to solar radiation and the system availability.

[0050] In order to provide a reliable basis for determining the system parameters, it is particularly intended that the data associated with the determination of the respective parameters, in particular from existing and new photovoltaic systems, are first evaluated, whereby neural networks, machine learning methods, artificial intelligence or algorithms initially determined empirically or derived from experience are preferably used for this evaluation.

[0051] In particular, the use of these technologies makes it possible to establish and recognise relationships between the parameters describing the individual phases of construction and operation of the photovoltaic system, starting with planning (represented by the planning parameter) and development (represented by the developer parameter), through the selection of hardware and / or software elements and / or components (represented by the hardware component parameter), installation (represented by the installer parameter and / or assembly parameter) and operation (represented by the environmental parameter, the operator parameter, the maintenance parameter and / or the operating parameter), and ultimately to make reliable statements about future system availability and output.

[0052] By incorporating the various parameters, the method according to the invention enables the simulation to incorporate not only product or hardware parameters, but also data at the process or execution level, in particular the planning development, implementation, and maintenance of the photovoltaic system, as well as the process reliability of the respective process participants in terms of quality management. This makes it possible to better plan and control resources for the power grid, avoid grid overloads, and, on the other hand, ensure sufficient supply security.

[0053] In particular, it can be provided that the various parameters are combined with different weightings to form an overall parameter, on the basis of which the system parameter is then determined.

[0054] In particular, the simulation includes the evaluation of different mechanisms that can lead to performance degeneration or an aging process.

[0055] Using (training) data from existing systems, machine learning of performance data and error analyses can be used to draw conclusions about how the various parameters for determining the different key figures, such as the planning key figure, the developer key figure, the hardware component size, the maintenance key figure and the operator key figure, influence the availability or the output power of the photovoltaic system.

[0056] In particular, production and construction data are incorporated into the key performance indicators. This is relevant information that is crucial for the construction of a photovoltaic system. This includes, for example, information relating to the photovoltaic modules used in the photovoltaic system, the specifications of the raw materials used, and production data from silicon, wafer, cell, glass, and / or module production, as well as the production of other module components. The data or parameters incorporated into the operator key performance indicators include, among other things, operating data to compare the system status with the expected value in terms of safety and power output. Finally, the influence of maintenance and repair data makes it possible to derive and map all measures that serve to maintain or restore the safety and availability of the photovoltaic system.

[0057] When using a machine learning system to evaluate the (training) data, it can be provided, in particular, that the relevant parameters present in the data are first determined by dimensionality reduction in order to determine which data or parameters have the greatest influence on the system parameters determining the output power and availability of the photovoltaic system. Furthermore, it can be provided that, using clustering methods, data from existing photovoltaic systems with comparable trends in the condition of the photovoltaic systems, in particular the output power and system availability, are summarized. This enables the classification method to learn patterns regarding the technical condition of the photovoltaic systems based on certain initial data, in particular via a deeper neural network.Multivariate regression analyses and statistical time series analyses are preferred for determining system parameters. Compared to other Lemv methods, this approach incorporates data from several different sources and of different types, allowing various factors and parameters to be incorporated into the determination of system parameters to determine the reliability and detection capability of the photovoltaic system.

[0058] The use of artificial intelligence makes it possible, in addition to determining the system parameters more precisely, to also verify the system by linking the different parameters to the key figures and / or to carry out a deviation analysis (so-called FMEA - failure mode and effects analysis, e.g. design FMEA, system FMEA, hardware FMEA, software FMEA, and / or process FMEA).

[0059] Finally, the use of artificial intelligence (“AI”) also makes it possible to make targeted suggestions as to how certain parameters, or the parameters influencing them, can be changed in order to achieve even higher power output or even higher system availability, or to initiate corrective measures for the future production of photovoltaic systems or their components.

[0060] In particular, to check and improve the output power and system availability, it can be provided that in the method according to the invention the first period is selected such that it only comprises partial phases of the construction and operation of the photovoltaic system through the suitable selection of a first point in time at which the first period begins and a second point in time at which the first period ends. For example, the system parameter for a phase 0, in particular a pre-planning phase of the system parameters, can be determined exclusively based on the hardware parameter. A pre-qualification system parameter is then determined. In a subsequent phase 1, in which a plan for the photovoltaic system preferably already exists, it can then be provided that the system parameter is determined based on both the hardware parameter and the planning parameter in order to determine a system design system parameter.In phase 2, in which a system development in particular has been completed, it can be provided that the system parameter is determined based on a combination of the parameters comprising hardware parameter, planning parameter and developer parameter, whereby a planning system parameter can be determined. It is also possible in phase 3, in which in particular the procurement of components and the installation of the photovoltaic system have been prepared, to use the installer parameter in addition to the parameters - if necessary including useful subcategories - whereby a design system parameter can be determined. If in phase 4, in which in particular the installation of the system has been completed, the installation parameter is also used, a commissioning system parameter can be determined.In phase 5, in which the system has preferably been commissioned, an operating system parameter can then be determined by additionally considering operator and / or operational parameters. In phase 6, in which the first maintenance intervals have preferably been completed – but as a prerequisite, at least the maintenance concept is available – the maintenance parameter can then be additionally used to determine a maintenance system parameter. In one embodiment, the planning parameter and / or the purchasing parameter are linked to the hardware and / or software component parameter, which in turn are linked to the installation parameter, creating an integrated operator parameter.

[0061] In particular, the system parameters created in the respective phases can be used to make improvements to the photovoltaic system or to adapt the subsequent phases in order to achieve the best possible system availability or maximized or, in terms of continuity, improved production output.

[0062] Preferably, the system parameters are re-determined at regular intervals, for example, annually, in order to detect any changes in the aging processes and thus make adjustments to the photovoltaic system or at least adjustments to the forecast values ​​in order to optimize or predict the system availability or production output of the photovoltaic system. A uniform data taxonomy can also be provided, which allows the interface between SCADA and other data acquisition methods (e.g., but not limited to, the acquisition of performance data from inverters or optimizers) of the photovoltaic system operators to be implemented in such a way that the data can be captured by the evaluation system.

[0063] An optimizer, as defined in the present invention, is understood in particular to be a power optimizer that is embedded in the solar modules instead of the junction box or is subsequently attached to photovoltaic modules by installers. These power optimizers, in particular, optimize the output power of the modules through MPP control and, in particular, ensure a fixed string voltage. This allows greater flexibility in the design of optimal photovoltaic systems (e.g., strings of different lengths or orientations can be connected in parallel).

[0064] Further features and advantages of the invention will become apparent from the following description, which provides an example for determining the aforementioned system parameters for an exemplary photovoltaic system.

[0065] Whenever appropriate, reference is made to the attached figures, whereby

[0066] Fig. 1 schematically shows the degradation of photovoltaic modules after one and several cycles of ageing processes;

[0067] Fig. 2. schematically shows the degradation of photovoltaic modules after a cycle of aging processes of a second system;

[0068] Fig. 3A shows the graphical plot of the maximum power PmppB at the maximum power point of a module B versus the maximum power PmppA of a module A of random module pairings A and B of a first photovoltaic module type;

[0069] Fig. 3B shows the graphical plot of the maximum power PmppB at the maximum power point of a module B versus the maximum power PmppA of a module A of random module pairings A and B of a second photovoltaic module type; Fig. 4 shows a graphical plot of damage amounts versus the investment amount for a plurality of photovoltaic systems;

[0070] Fig. 5 shows a graphical representation of the output power of a photovoltaic system over time;

[0071] Fig. 6 shows a graphical representation of the power output of a photovoltaic system over time and the associated degradation and partial system failures in the example;

[0072] Fig. 7 is a graphical representation of the degrading power output of a photovoltaic system, showing dirt and sand deposits on the modules over time without module cleaning; and

[0073] Fig. 8A and 8B graphical representations of the test results of different

[0074] Batches of produced photovoltaic cells with regard to the distribution of the surface resistance per cell versus the position of the cell in the rack in the diffusion furnace, in the given example a rack with 250 cells.

[0075] The following describes, using exemplary embodiments, parameters and characteristics that can be used to determine the system parameters, based on exemplary photovoltaic systems. These parameters and characteristics can be used alternatively to one another or in any combination to determine the system parameters. Various types of system parameters that can be determined according to the invention are also described. These can be determined alternatively or combined in any combination to form an overall system parameter.

[0076] Based on the specific system parameter, at least one variable of the photovoltaic system can then be adjusted and / or influenced.

[0077] Example 1: Consider a newly constructed photovoltaic system on an industrial scale (so-called "utility scale", i.e., generally > 5 MWp). The first point in time is therefore preferably the beginning of the preliminary planning phase, and the second point in time is the commissioning of the photovoltaic system. Between the first and second points in time, additional interim points may be necessary, at which parameters are recorded, insofar as these are necessary to determine or calculate the parameters for the second point in time. The parameters of the photovoltaic system comprise the following chapters:

[0078] General evaluation and pre-qualification of the participating companies based on structured criteria: o Evaluation of the developer to determine a developer characteristic o Evaluation of a manufacturer of hardware and software components, e.g. a module manufacturer and an inverter manufacturer, to determine at least one manufacturer characteristic o Evaluation of an EPC to determine at least one installer characteristic o Evaluation of the operator responsible for operation, maintenance and repair to determine at least one operator characteristic or at least one maintenance characteristic

[0079] General evaluation of the hardware components including a differentiated assessment of suitability for specific climatic and environmental conditions to determine the various hardware component parameters, e.g. o at least one module parameter o at least one inverter parameter (if necessary with optimizer)

[0080] Project-specific assessment o Development of the site (usually carried out by the developer) to determine at least one site parameter, at least one environmental parameter, and the level of infrastructure in the surrounding area o Planning of the photovoltaic system (usually carried out by the installer) to determine at least one planning parameter o Purchasing of the hardware and software components and, if applicable, subcontractors to determine at least one purchasing parameter or supply chain parameter that is part of the installer parameter o Construction / erection of the photovoltaic system to determine at least one installation parameter eo Testing and acceptance of the photovoltaic system to determine at least one conformity parameter that is part of the system parameter o Operation, maintenance and servicing concept of the photovoltaic system to determine at least one maintenance parameter and / or at least one operator parameter

[0081] Input parameters:

[0082] For each of the above-mentioned chapters, an expert performs evidence-based individual assessments of the criteria specified in an assessment questionnaire, for example, using a whole number between 0 and 10, or any arbitrary number format or range. Each criterion is assigned a weighting. Among the criteria, there may be one or more knockout criteria. A knockout criterion is defined as a failure to meet a minimum level that is so critical that certification must be excluded or—if, for example, it is not safety-relevant—at least a significant downgrade must be applied.If the assessment of a KO criterion is determined to be below a certain threshold, points are deducted in an overall assessment of the chapter through a KO downgrade such that non-fulfillment of the minimum threshold of at least one KO criterion automatically leads to the determination of a penalized normalized score or system parameter in the sense of the table below, e.g. a rating of C or D. On the other hand, full fulfillment of all criteria leads to a very high normalized score or system parameter and thus to a rating of e.g. AAA.

[0083] Input parameters also include environmental parameters such as the geodesics of the installation site, a comprehensive key figure-based input of environmental criteria including the climate zone and temperature range; further environmental data relate to the immediate surroundings, for example, proximity to a farm (ammonia emissions), proximity to a coast (salt content in the air), and special expected weather events.

[0084] For each of the aforementioned assessments, a score is determined using a structured assessment questionnaire. Each assessment point is weighted, for example, with a whole number between 0 and 10. The result is a total calculated by adding all individual assessments times the corresponding weighting factors. Points are deducted from this total if at least one of the so-called knockout criteria is not met.

[0085] The maximum possible total per test sheet is calculated by adding the maximum value (e.g., 10) times the respective weighting factor. This maximum total is normalized to 1000, resulting in a conversion factor U (U = 1000 / Mo). The result of the chapter total, i.e., the absolute score achieved, is normalized with U according to the following formula:

[0086] Pn = Po * U = Po * 1000 / Mo

[0087] P n = normalized score of the assessment; P o Absolute score of the assessment, M o the maximum possible absolute score, the number 1000 reflects the maximum possible normalized score.

[0088] A more complex procedure is used for the operator parameter, since sub-links are necessary to determine it at the current stage of development.

[0089] Output parameters:

[0090] A normalized total score for each chapter and a rating derived from this for each chapter

[0091] For photovoltaic systems as system parameters: o Overall assessment (score) and rating according to Table 1 o Expected value regarding the service life o Forecast value of the normalized output power in relation to the nominal power and the expected irradiation value o Yield probability of the calculated service life (e.g. 20 years) P50, P75, P90, P99, i.e. the delivered power that will be exceeded with a probability of 50%, 75%, 90%, or 99%; this is determined as an integral over the service life, taking into account the expected degradation derived from the assessment. o Failure probability in % / a for the following 3 years

[0092] Table 1 The rating categories AAA to BBB correspond to a plant parameter that corresponds to a so-called investment grade, i.e. excellent to good reliability including compliance with all relevant norms and standards. The rating categories BB and B correspond to a plant parameter that corresponds to a minimum fulfillment of the relevant norms and standards. The rating categories C and D correspond to a plant parameter that corresponds to a minor (C) or severe (D) deviation or non-conformity with relevant norms and standards and thus, for example, to a non-approval for grid-connected regular operation or for island operation or for operation at all. For smaller plants, a simplified rating matrix may result, which is derived according to the invention from the relatively complex matrix mentioned above.The various evaluations for determining the system parameter are linked, for example, according to the following Formula 1, Formula 2, Formula 3, Formula 4, and / or Formula 5, and ultimately result in an overall evaluation in the form of the system parameter. In the following example, the overall evaluation or the system parameter is derived from Formula 1, which in turn is derived from the following formulas: Formulas 2 to 4 are incorporated into Formula 5, and Formula 5, together with the results from the other determined key figures in the above-mentioned chapters, is incorporated into Formula 1.

[0093] Formula 1 : x Developer key figure - normalized key figure determined from the evaluation chapter “Developer” y Key figure “Test and Acceptance” - normalized key figure determined from the evaluation chapter “Test and Acceptance” z “Installation key figure” (EPC) Interim result as interim result “CZ”, normalized key figure see formula 5 below

[0094] R Minimum key figure for a rating “B” Specified key figure (e.g. 680) as a minimum key figure that meets the minimum requirements m, n A whole, rational or irrational number between 0 m, n, p, q are empirically and 10 determined or based on experience p, q A rational or irrational number between 0 and 1 assumed or derived by AI / ML

[0095] Formula 2: x Planning parameter initial assessment y Project-relevant hardware parameter modules z Project-relevant hardware parameter inverter a* A factor between 0 and 1 (rational or irrational number) ß* A factor between 0 and 1 (rational or irrational number) y* A factor between 0 and 1 (rational or irrational number)

[0096] R Minimum score for a rating “B”

[0097] * a + ß + y = 1

[0098] Formula 3: x Initial key figure for purchasing y Project-relevant hardware key figure for modules z Project-relevant hardware key figure for inverters a* A factor between 0 and 1 (rational or irrational number) ß* A factor between 0 and 1 (rational or irrational number) y* A factor between 0 and 1 (rational or irrational number)

[0099] R Minimum score for a rating “B”

[0100] * a + ß + y = 1 Formula 4: 3,4, 5} x4 Hardware characteristic module, applicable for cold and moderate climates, combined, normalized characteristic x2 Hardware characteristic module, applicable for warm and dry climates, combined, normalized characteristic x3 Hardware characteristic module, applicable for warm and humid climates, combined, normalized characteristic x4 Hardware characteristic module, applicable for hot and dry climates, combined, normalized characteristic x5 Hardware characteristic module, applicable for hot and humid climates, combined, normalized characteristic y Hardware characteristic inverter, applicable for cold and moderate climates, combined, normalized characteristic y2 Hardware characteristic inverter, applicable for warm and dry climates, combined, normalized characteristic y3 Hardware characteristic inverter, applicable for warm and humid climates, combined, normalized characteristic y4 Hardware characteristic inverter,applicable for hot and dry climates, combined, normalized index y5Hardware index inverter, applicable for hot and humid climates, combined, normalized index,

[0101] R Minimum score for a rating “B”

[0102] Formula 5: e, p, c), e < R \ / p < R \ / c < R + yc), otherwise e planning and engineering, combined, normalized calculation see

[0103] Key figure formula 2 p purchasing key figure, , combined, normalized key figure calculation see

[0104] Formula 3 c Plant engineering, combined, normalized key figure Calculation see

[0105] Formula 4 a* A factor between 0 and 1 (rational or irrational a, ß, y are empirically determined or empirically determined number) ß* A factor between 0 and 1 (rational or irrational assumed or number) Y* A factor between 0 and 1 (rational or irrational by KI / ML number) derived values ​​o Process / organizational key figure Establishing company Determination from (EPC), combined, normalized key figure Assessment chapter “Divider”

[0106] R Minimum key figure for a rating “B” Specified key figure (e.g. 680) as a minimum key figure that

[0107] Minimum requirements corresponds to a + ß + Y = 1

[0108] The identified key performance indicators allow for an assessment of individual disciplines as well as a combined assessment at the overall project level. For additional guidance, the identified key performance indicators and / or the specific asset parameters can be translated into financial nomenclature (e.g., AAA - D) using the table above:

[0109] For smaller systems, a simplified evaluation matrix may result, which is derived according to the invention from the above-mentioned, relatively complex matrix.

[0110] Example 2:

[0111] Consider an industrial-scale photovoltaic system (so-called “utility scale”, i.e. generally > 5 MWp) that has been in operation for at least 1 year. The first point in time is therefore the beginning of the simulation interval year 1, and the second point in time is a point in the future. The determination of the key figures, in particular the system parameters of the photovoltaic system, covers the chapters analogous to those in example 1 for the first period. However, to evaluate the subsequent periods, the performance data recorded via an interface and the actually determined maintenance parameters are now also recorded over time, evaluated, and compared with the expected values ​​and / or simulation values ​​determined for the first period. This comparison is carried out, among other things, for the further adjustment of the entire system through the application of artificial intelligence (“AI”) and machine learning (“ML”).The significance of the parameters determined in the first period, which form the basis of the development phase including the construction phase of the photovoltaic system, decreases with increasing operating time, so that the following evaluation shares, shown in Table 2, result from the initial evaluation and from the actual data of a specific photovoltaic system, assuming an annual review:

[0112]

[0113] Table 2 Example 3:

[0114] Consider a group of photovoltaic systems whose performance data was evaluated using AI and ML by obtaining data from the field. It turns out that the photovoltaic systems equipped with Type A modules from manufacturer X deviate significantly from the expected value in terms of their power output, as they exhibit a degradation rate 1 percentage point per year greater than the tested value, contrary to previous tests. These parameters represent a hardware component characteristic that is incorporated into the determination of the system parameter. This leads to a re-evaluation of the photovoltaic modules and thus of the system parameter according to Table 3.

[0115] Table 3

[0116] This will result in a corresponding reduction in the ratings and project parameters of all photovoltaic systems equipped with this module type; in particular, the system parameters of these systems will be adjusted. This means that all photovoltaic systems equipped with this module will be re-evaluated at the latest in the next cycle and will be re-evaluated and downgraded accordingly immediately, but no later than after the expiration of the rating certificate (e.g., after a maximum of one year). In particular, the time for the necessary replacement of type X modules will be postponed. In other words, the photovoltaic system itself, or its control system, will be modified due to the changed system parameters.The photovoltaic system is still classified as safe, but the expected service life, failure probability, and yield forecasts change accordingly, which is addressed by correspondingly shortened maintenance and decommissioning intervals for the modules. Example 4:

[0117] Consider the review assessment of a manufacturer (analogously valid for the review and evaluation process of a developer, installer, or operator) with regard to the aforementioned criteria, e.g., financial strength, experience (e.g., references), soundness of business processes (of quality management), and other relevant checkpoints for determining the manufacturer's key performance indicator (the developer's key performance indicator, the installer's key performance indicator, etc.). The following example describes how a manufacturer's key performance indicator (and so on) is corrected if new findings emerge after determining the first key performance indicator, which was initially determined based on a qualified self-assessment due to travel restrictions.

[0118] Based on the qualified and evidence-based self-assessment, a so-called preliminary desktop assessment was determined. In the on-site reviews following the "paper assessment," structured interviews revealed glaring discrepancies between the content of the quality manual and the understanding of how a quality organization operates. The interview results of 17 employees, including managers from the first three levels and selected individuals from production and quality management, yielded the following picture, presented in Table 4 (full compliance: 1000 points, minimum score for a "B" = pass: 681 points):

[0119] Table 4

[0120] Based on the expected result, the manufacturer’s characteristic is corrected as shown in Table 5:

[0121] Table 5 Example 5:

[0122] Consider a selection of 60-cell modules that were subjected to accelerated aging tests as part of the test protocol. The results of these aging tests are shown in Figure 1. Following exposure to accelerated aging, the electroluminescence (EL) method was used to determine how many cells were still fully or partially active after the end of the exposure, in particular to determine an operating parameter. Furthermore, performance was measured using so-called flash tests. The measured values ​​were compared with the forecast model. The agreement between the measured value and the calculated value can be described as very good, as can be seen from the highlighted area of ​​the table in Figure 1. The calculation model is accordingly incorporated into the definition criteria for the module parameter, in particular for determining the operating parameter, based on which the system parameter is in turn determined.

[0123] Furthermore, in the example shown in Figure 1, it can be seen that the degradation of the modules after the last exceptionally harsh stress test remained within a very narrow range, which, considering the uniformity of the expected aging progression, is rated positively with full marks, thus a correspondingly high operating parameter. This stress test was specifically designed to expose the modules to a hot and humid climate zone. The absolute result of the stress test shown in Table 5 is incorporated into the module parameters for the various climate zones, particularly as an environmental parameter. Table 5

[0124] Example 6:

[0125] Consider two 60-cell modules of the same type that were subjected to a PID test, which also simulates aging to determine the operating parameters. As in Example 5, the EL method is used here. The results are shown in Figure 3.

[0126] The highlighted area of ​​the table in Figure 3 shows that the degradation of the two test modules differs significantly from each other in this test, resulting in significant point deductions in the rating model, respectively a quantitatively reduced operating characteristic and a quantitatively reduced system characteristic, since the power output of a string is based on the module with the lowest performance. These deviations are incorporated into the forecast models according to Table 6 and into the AI / ML for determining the system characteristic.

[0127] Table 6

[0128] Example 7:

[0129] As is known from the relevant literature, modules have different performance levels after production, and in particular, they have different hardware component characteristics. The different performance levels are usually classified during the manufacturing process, firstly at the cell production level and finally after module manufacture. Nevertheless, modules age differently, which leads to diverging module performance levels for modules of the same performance class. Figures 3A and 3B show diagrams that list the performance data Pmpp after aging for randomly selected module pairs of the same photovoltaic module type. The maximum power Pmpp at the maximum power point of module B is plotted against the maximum power Pmpp of a randomly selected module B of the same photovoltaic module type.In the case where modules A and B are of the same age, the measurement points lie on the straight line shown in Figures 3A and 3B. The set of points for the module pairs A and B resulting from this data determination demonstrates a more or less good proximity to the ideal diagonal line. In example 1 shown in Figure 3A, a very good, close distribution and proximity to the ideal line is observed, whereas in example 2 shown in Figure 3B, an almost random distribution of the pairings can be observed. A significant deviation from the ideal line leads to significant underperformance in the string. These results are incorporated into the evaluation system via KI / ML and thus into the module characteristic.

[0130] The determination of these module pairings is either carried out over time after at least a first period, a second period, etc., or alternatively by simulated accelerated aging tests.

[0131] Example 8:

[0132] A study examined approximately 3,600 insurance claims in the photovoltaics sector. The range of loss amounts over investment is shown in Figure 4.

[0133] The corresponding module types and inverter types are clustered from the data of the above study using AI / ML and assigned to the evaluation system as a hardware component characteristic, as shown in the diagrams for modules (left) and inverters (right) contained in Figure 5. Ml..M35 stands for modules from one of 35 module manufacturers, W1..W17 for inverters from one of 17 inverter manufacturers.

[0134] The damage rate determined using AI and ML from so-called internal damage causes - i.e. causes that are, for example, design-related - flows into the evaluation system and thus into the determination of the module and inverter parameters as well as into the calculation of the failure probability of a photovoltaic system equipped with the respective hardware components.

[0135] Example 9:

[0136] In the following example, data was transmitted via the SCADA interface, resulting in a significant loss of performance in the photovoltaic system that was not detectable by the evaluation system at the time. At this earlier stage of development, a continuously recurring evaluation was not planned. The insights from this application example have led to the expansion of procedures so that the temporal progression of the key parameters can be recorded repeatedly and with the help of AI / ML.

[0137] For clarification: COD is the commercial commissioning date. The review after approximately 3 years revealed that the photovoltaic system had suffered an increasingly significant loss of performance over the first 3 years, as shown in Figure 6. The data suggested that the deviation had already begun after approximately 1 year (start of deterioration). After that, the performance of the photovoltaic system deteriorated progressively. Only through extensive measures (approximately 38% reinvestment) were the causes corrected, so that after completion of the measures, the photovoltaic system was once again able to deliver the contractually agreed minimum output. Measurement data such as that shown in the example above lead to at least a temporary devaluation of the performance indicators and, as a consequence, possiblyto a devaluation of the hardware component parameters of the hardware components that are the cause of the accelerated degradation (in this case the modules), furthermore the.

[0138] The purchaser's key figures, the installer's key figures, the planning key figures, the operator's key figures, and the maintenance key figures are corrected. The project-specific key figures and the resulting rating of the photovoltaic system can be reassessed after the measures have been completed, provided the originally agreed performance data is achieved again.

[0139] Example 10:

[0140] Different climatic conditions (environmental factors) lead to different module requirements. The given example shows the temperatures for a hot, arid climate and a moderate climate, as well as the consequences for the diode temperature. The critical temperature according to standards is 85°C.

[0141] In arid, hot climates, the daytime temperature can rise well above 50°C; in some cases, there are also large spreads in the daily temperature curve (day / night).

[0142] In contrast, the maximum temperatures and temperature spreads in moderate climates are significantly lower.

[0143] As a result, individual electronic components of the hardware components reach even higher temperatures, for example, diode temperatures. It is known from the literature that temperatures of 40°C (irradiation approx. 1000 W / m 2 ) can already lead to a diode temperature in the range of 80°C.

[0144] For outside temperatures in the range of 50°C (irradiation approx. 1100 W / m 2) Diode temperatures well over 100°C can occur, which is significantly too high for the relevant standards (max. 85°C). As a result, the excessive temperature rise leads to reduced performance and significantly increased degradation, especially around the overheated components of the hardware.

[0145] For example, this led to corresponding downgrades for one module in warm and especially hot climates, as shown in Table 7:

[0146] Table 7 11:

[0147] A photovoltaic system is installed and operated in an arid climate zone (e.g., a desert). Initial performance is as predicted. The maintenance concept and the maintenance company were convincing in the initial evaluation.

[0148] It turns out that the maintenance company hadn't adequately cleaned the plant of the constantly accumulating fine sand. The performance drop is usually detected by the SCADA system. Typical performance drops follow the curve shown in Figure 7.

[0149] Through the analysis using AI and ML, particularly the maintenance metric and / or the environmental metric, it is determined in the example that the photovoltaic system has not been cleaned for more than 35 days, which is due to inadequate maintenance. The maintenance company did not perform the required work according to the maintenance schedule, and the operator failed to remind the maintenance company to perform proper maintenance / fulfill the contract. This leads to a temporary devaluation of the system's metrics, the operator metric, and the maintenance metric according to Table 8.

[0150] Table 8

[0151] A return to the original assessment area requires corrective measures and the corresponding evidence:

[0152] (1) Revision and effective implementation of the processes on the operator’s side

[0153] (2) Revision and effective implementation of the processes on the part of the maintenance company

[0154] (3) Power output is achieved within the original expected corridors. Subsequently, further photovoltaic systems of the operator or the maintenance party can be subjected to an inspection, if necessary combined with a devaluation of the parameters, expected values ​​and failure probabilities of these and other photovoltaic systems.

[0155] Example 12:

[0156] In cell production for photovoltaic modules, the so-called emitter sheet resistance distribution can be measured after the diffusion process (doping). As can be seen from the literature (e.g., L. Shen, ZC Liang, CF Liu, TJ Long, DL Wang: Optimization of oxidation processes to improve crystalline silicon solar cells), there is a relationship between cell efficiency and sheet resistance such that cell efficiency reaches a maximum at a certain sheet resistance, while at higher or lower sheet resistances than the optimum, cell efficiency decreases. Excessive doping leads to excessively low resistance, while excessively low doping leads to excessively high resistance.

[0157] Measurements taken during industrial cell production can demonstrate how sheet resistances vary across a batch. If available, this information, in conjunction with other production data (e.g., cell sorting, module assignment of pre-sorted cells), can be incorporated into module evaluation using AI / ML. A homogeneous distribution within the target values ​​(green area in Figures 8A and 8B) leads to maximum good ratings; a distribution within the permissible limits (green and yellow areas in Figures 8A and 8B) leads to good ratings; and a distribution with a high proportion of values ​​outside the permissible limits (red area in Figures 8A and 8B) leads to a medium to poor rating.Figure 8A shows a very good result for batch 2, i.e. a quantitatively high hardware component characteristic, and a mediocre result in the distribution for batch 3, i.e. a quantitatively lower hardware component characteristic.

[0158] The results of the actual use of the cells and their installation in the modules can be incorporated into the cell and module parameters via data transfer and subsequent AI / ML, and thus also into the system parameters. The resulting module parameter is typically assigned to a specific project using allocation keys. This requires appropriate data management and linking using AI and / or ML.

[0159] The features specified in the preceding description and the claims may be essential to the invention in its various embodiments, both individually and in any combination.

Claims

CLAIMS 1. Computer-implemented method for determining at least one system parameter representing an expected production output, system availability and / or failure probability over at least a first period of at least one photovoltaic system comprising at least one photovoltaic module or a plurality of photovoltaic modules, wherein the method comprises at least one simulation of the development of the efficiency of the photovoltaic modules and / or the photovoltaic system, and wherein the simulation comprises Determination of at least one parameter related to the photovoltaic system, at least one inverter, at least one element of the photovoltaic system and / or the photovoltaic modules, Determination of at least one expected temporal course of the parameter over the first period and / or at least a specific and / or indefinite period, Determination of at least one expected progression of the degree of aging of the photovoltaic system and / or the photovoltaic modules based on the expected temporal progression of the parameter, Determination of at least one expected temporal progression of the output power of the photovoltaic system and / or the photovoltaic modules based on the expected temporal progression of the degree of aging, Determination of at least a probability with which a discharge can be expected, Determination of an expected efficiency over the first period based on the expected time course of the output power, and / or calculation of the system parameter based on the expected efficiency.

2. Method according to claim 1, characterized in that the parameter is based on at least one developer parameter, at least one developer, at least one hardware component parameter, in particular of the photovoltaic modules and / or the inverters, at least one environmental parameter, at least one planning parameter, in particular comprising information on the design and / or the engineering, at least one installer parameter, in particular comprising and / or formed from at least one characteristic of a project participant, preferably at least one installer, at least one developer, at least one manufacturer, at least one erector, at least one planner, at least one purchaser, at least one construction company and / or at least one general contractor, at least one assembly characteristic, at least one operator characteristic, at least one maintenance characteristic, at least one operating characteristic and / or at least one process characteristic and / or a combination of at least two, preferably a plurality of the aforementioned characteristics.

3. Method according to claim 2, characterized in that the developer characteristic is determined based on the developer's experience in the development of photovoltaic systems, based on the operating time of planned systems, based on a certification of the developer, based on monitoring and / or reporting options of the developer, and / or the business processes in the sense of quality management.

4. Method according to claim 2 or 3, characterized in that the planning parameter is based on a certification of the planner and / or purchaser, based on a size of the photovoltaic system, based on an area of ​​application of the photovoltaic system, such as industrial use, commercial use and / or private use, based on an external examination of the planner, based on proven qualifications of the planner, based on a self-disclosure of the planner, based on a comparison of the planning of the photovoltaic system and / or photovoltaic modules and / or inverters and / or other components or subcomponents, in particular photovoltaic cells and / or connectors, with standards and / or norms and / or “best practice guidelines”, preferably based on the best available technology (BAT orBAT = Best Available Technology), based on the number, size, installation climate zone, site-specific environmental conditions and / or quality, in particular with regard to solidity and / or performance indicators of the photovoltaic systems planned by the planner in the past, based on guarantees and / or warranties granted by the planner, based on the frequency of service and / or warranty claims against the planner, based on the planner's insurance parameters, such as insurer, type of insurance, amount covered, scope of coverage, deductible, no-claims discounts, insurance exclusions. and / or other insurance conditions, based on the planner's monitoring and / or reporting capabilities, based on a comparison with at least one environmental parameter and / or a comparison of a climatic zone with the environmental and / or climatic zone suitability and / or assessment of the planned components and / or based on a comparison of the geological data.

5. Method according to one of claims 2 to 4, characterized in that the hardware component characteristic is based on at least conformity of the photovoltaic system, the photovoltaic modules and / or the hardware components and / or software components of the photovoltaic system and / or the photovoltaic modules with norms and / or standards, based on at least one specification of materials and / or components used for the production of the hardware components, preferably the photovoltaic modules, the inverters and / or at least one other component, software components and / or photovoltaic system, such as manufacturing data of the raw materials, in particular silicon, wafer, cell, cell connector, glass and / or encapsulation materials, manufacturing data of junction boxes and / or connectors and / or photovoltaic module manufacturing, based on a resistance to environmental influences, in particular comprising climate-related influences, such as temperature fluctuations,maximum and minimum absolute temperature exposure, humidity and / or dryness, solar radiation, in particular including UV radiation components, precipitation, hail, snow, pollutants, in particular of any aggregate state, such as air pollutants, and / or salinity of the ambient air, based on material combinations used, preferably also, but not only, in comparison to certified material combinations, based on circuit technologies used, such as individual inverter configurations, in particular with or without optimizer, string inverter configurations, in particular with or without optimizer, central inverter configurations, in particular with or without optimizer), based on certifications of the manufacturer and / or manufacturers of the hardware components, the software components, the photovoltaic system and / or the photovoltaic modules, based on quality ratings of the hardware components, in particular the photovoltaic modules,the inverter and / or at least one other component and / or software components of the photovoltaic system, based on, in particular on the manufacturer's side, for the hardware components and / or software components of the photovoltaic system and / or, the guarantees and / or warranties granted for the component, in particular the photovoltaic modules, the inverters and / or at least one other component, as well as their exclusions, in particular where made available, based on at least one frequency of guaranteed performance and / or warranty claims against the manufacturer of hardware components, in particular the photovoltaic modules, the inverters and / or at least one other component and / or software components of the photovoltaic system and / or, based on an economic success and / or economic development of the manufacturer of the hardware components, in particular the photovoltaic modules, the inverters and / or at least one other component and / or software components of the photovoltaic system, based on an audit of the manufacturing operation of hardware components, in particular the photovoltaic modules,the inverter and / or at least one other component and / or software components of the photovoltaic system, based on a comparison with at least one environmental parameter and / or a climate zone in comparison to their suitability and / or based on a comparison of geological data and / or weather data in comparison to the suitability for the operation of a photovoltaic system and / or based on business processes, such as quality management.

6. Method according to one of claims 2 to 5, characterized in that the environmental parameter is based on • at least one installation type and / or at least one installation location of the photovoltaic system, such as ground mounting, installation integrated into agricultural land, in particular agrovoltaics, roof mounting, installation in textile architecture, building-integrated photovoltaics, floating on-water installation, floating photovoltaics, extraterrestrial installation and / or installation on a vehicle, in particular aircraft, land vehicle and / or watercraft and / or based on lighter-than-air technology (LTA) • at least one alignment of the photovoltaic system and / or the photovoltaic modules, in particular tracking the position of the sun, preferably single- or multi-axis • at least one inclination of the photovoltaic modules, in particular with the position of the sun, preferably single- or multi-axis tracking • at least one geographical location of the photovoltaic system, in particular for determining the climate zone, based on at least one component of the probability of precipitation, probability of hail, probability of snow, probability of wind strength, probability of shading, probability of pollution, based on expected solar radiation data, in particular including UV radiation components, temperature fluctuations, min. and / or max. temperature exposure and / or min. and / or max. humidity data, air pollutants and / or salinity of the ambient air at the location of the photovoltaic system, in particular a future development of one, several or all of these probabilities; • at least a distance of the photovoltaic system from industrial plants, agricultural operations, salt water bodies, in particular seas and / or oceans and / or salt accumulations, in particular salt lakes, salt works, salt mountains and / or salt pans) and / or from other natural or man-made sources of increased pollutant concentrations, preferably corrosive species, in particular in the air and / or in the water • at least one degree of environmental protection of the surroundings of the photovoltaic system, in particular by surrounding buildings or technical facilities, preferably antennas, vegetation and / or natural facilities • at least one parameter relating to the foundation of the photovoltaic system, • at least one parameter relating to drainage of the photovoltaic system, at least one parameter relating to the degree of infrastructure in the region of the photovoltaic system, in particular the distance of the photovoltaic system to supply facilities for the procurement of spare parts and / or to locations for maintenance personnel and / or • at least one parameter of the maintenance of the photovoltaic system is determined.

7. Method according to one of claims 2 to 6, characterized in that the operator characteristic is based on historical data, in particular on the qualification, preferably of employees, on damage cases, on complaint handling, preferably based on testimonials, certificates, evaluations, profit margins, efficiency, production output, specifications, documentation and / or protocols, preferably including reports and / or photo documentation, and / or reliability of existing systems of the installer and / or project participant, based on the number, size and / or types, preferably rooftop systems, field systems with or without solar tracking, of the photovoltaic systems installed by the installer and / or project participant in the past, based on insurance parameters of the installer and / or project participant, such as insurer, insurance type, coverage amount, scope of coverage, deductible, no-claims discounts, insurance exclusions and / or other insurance conditions, based on an external quality audit of the installer and / or project participant and / or at least one subcontractor of the installer and / or project participant,based on the conformity of a mounting concept with safety, electrical, and / or construction norms and / or standards, based on guarantees and / or warranties granted by the installer and / or project participants, based on a frequency of service and / or warranty claims against the installer, based on the quality testing procedures used by the installer for hardware and / or software components installed in the photovoltaic system - in the qualification phase and / or during the implementation phase, and / or based on the business processes in terms of the quality management of the installer and / or project participants, 8. Method according to one of claims 2 to 7, characterized in that the installation parameter is determined based on at least one comparison of the planned photovoltaic system with the installed photovoltaic system, and / or based on business processes in the sense of the quality management of the installation company.

9. Method according to one of claims 2 to 8, characterized in that the operator characteristic is based on historical data on damage cases, causes of damage and / or their rectification, based on profit margins, efficiency, production output and / or operator concept of existing systems of the operator, based on a number, size, availability and / or other performance indicators, in particular comprising a relation of output to forecast power and / or output in relation to the measured solar irradiation, mean time between maintenance and / or mean time between system failures of the photovoltaic systems operated by the operator, based on insurance parameters of the photovoltaic system, such as insurer, type of insurance, coverage amount, scope of coverage, deductible, no-claims discounts, insurance exclusions and / or other insurance conditions and / or based on the business processes in the sense of the operator's quality management.

10. Method according to one of claims 2 to 9, characterized in that the maintenance parameter is based on a review of a maintenance plan and / or monitoring system, in particular a comparison of the maintenance plan and / or monitoring system with standards, norms and / or best practices, based on historical data on damage cases of the photovoltaic systems maintained by the person maintaining the photovoltaic system, based on profit margins, efficiency, production output, operating times and / or causes of damage and / or their rectification, existing systems of the person maintaining the system, based on a number, size, concept and / or the photovoltaic systems maintained by the person maintaining the system, based on processes and / or response times of the person maintaining the system, based on insurance parameters of the person maintaining the system, such as insurer, type of insurance, coverage amount, scope of coverage, deductible, no-claims discounts, insurance exclusions and / or other insurance conditions,based on an external quality audit of the waiting party and / or at least one of the waiting party's subcontractors, based on the waiting party's response times, based on a minimum uptime guarantee given by the waiting party, and / or based on the business processes in terms of the quality management of the waiting party, in particular a maintenance company.

11. Method according to one of the preceding claims 2 to 10, characterized in that the operating parameter is based on a degree of completeness of the documentation necessary for the construction, planning, purchasing of components, installation, operation and / or maintenance of the photovoltaic system, based on a test and / or qualified self-declaration of data sheets of the materials, components, hardware components used in the photovoltaic system, in particular the photovoltaic modules, inverters and / or at least one other hardware component and / or software components, based on a test and / or qualified self-declaration of construction plans, cabling plans and / or diagrams, based on an examination and / or qualified self-declaration of handling and / or installation protocols, based on the examination and / or qualified self-declaration of acceptance protocols, based on an examination and / or qualified self-declaration of the conformity of the realized photovoltaic system with the planning of the photovoltaic system, based on an examination and / or qualified self-declaration of the supporting structure, the hardware components, in particular comprising the photovoltaic modules, the inverters and / or at least one further hardware component, the software components, and / or the design, based on at least one examination and / or qualified self-declaration for hotspots and / or based on at least one electroluminescence measurement, preferably for cracks, breaks, defects,inactive cells and / or cell parts and / or other suitable criteria, in particular using an infrared measuring method, an electroluminescence recording device and / or at least one other suitable measuring or documentation device, based on at least one test and / or qualified self-declaration of the photovoltaic system, at least one photovoltaic module, at least one hardware component, based on a comparison of the serial number of the hardware components with the installed position and / or the installation location of the installed photovoltaic system, in particular including a position planned for installation and / or geodesics, based on a test of the traceability of the supply chain for the hardware components, in particular the photovoltaic modules, the inverters and / or at least one other hardware component,and / or the software component and / or based on the quality of the connections and / or links of the hardware and software components and / or the quality of the substructure, in particular including at least one tracking system, preferably on stability, susceptibility to corrosion, natural frequency and / or other essential quality characteristics and / or the quality of the cabling, in particular based on a quality of the cable specification and / or a quality of a laying.

12. Method according to one of the preceding claims, characterized in that the determination of at least one variable and / or a temporal progression of the variable, using a neural network, using artificial intelligence, using Use of machine learning and / or using links derived empirically and / or from experience and / or assumed values, wherein the variable comprises one or more variables selected from the group comprising the parameter, the degree of aging, the power output, the efficiency, the developer parameter, the hardware component parameter, the parameter of the manufacturer of hardware or software components, the planning parameter, the environmental parameter, the operator parameter, the installer parameter, the maintenance parameter, the assembly parameter, the operating parameter and / or other relevant parameters relating to the manufacture of hardware and / or software components, to the construction and / or operation of a photovoltaic system.

13. The method according to claim 12, characterized in that training data is used to determine the variables by means of the neural network, artificial intelligence, machine learning, wherein this training data preferably comprises data on damage cases, insurance data, development of the system parameters, the variables, the characteristics and / or the performance data and / or the status data of the photovoltaic system, in comparison with the parameters previously determined for other systems with the actual parameters, preferably for existing photovoltaic systems.

14. The method according to claim 12 or 13, characterized in that the use of a neural network, the use of artificial intelligence, the use of machine learning and / or the use of empirically and / or empirically assumed values ​​of links derived from the use of at least one dimensionality reduction, at least one principal component analysis, PCA (principal component analysis), at least one clustering, at least one classification, at least one multivariate regression analysis, at least one, preferably statistical, time series analysis, and / or at least one evaluation criterion with regard to conformity with relevant norms, standards and / or best practices, in particular for processing and / or editing training data.

15. Method according to one of the preceding claims, characterized in that hardware components comprise at least one photovoltaic module, at least one photoactive cell for converting photons into electrons or vice versa, at least one inverter, at least one converter, at least one connecting line, at least one cabling, at least one distribution box and / or at least one device for merging and / or separating and / or distributing cables and / or cable harnesses, at least one plug connection, at least one holding element, at least one supporting element, at least one supporting structure, at least one tracking system (“tracker system”), at least one drive unit, in particular comprised by the tracker system, at least one control unit, at least one sensor, at least one light sensor, at least one precipitation sensor, at least one sun position sensor and / or at least one optimizer, preferably for at least one of the Photovoltaic modules and / or operatively connected to the photovoltaic module.

16. Method according to one of the preceding claims, characterized in that the software component comprises at least one control software, at least one tracker software, at least one inverter software and / or an operating and / or maintenance software.

17. Method according to one of the preceding claims, characterized in that the first period of time is determined by at least a first time and / or start time and a second time and / or a further time and / or an end time.

18. Method according to claim 17, characterized in that the first point in time is the start of a preliminary planning phase, the second point in time is the end of the preliminary planning phase, the end of a planning phase, the end of a development phase, the end of a planning phase of the installation of the photovoltaic system, the start of the assembly phase and / or the Notice to Proceed (NTP) point in time, the end of an assembly phase, the start of a commissioning phase, the end of a commissioning phase, the end of the preliminary and / or final acceptance phase, in particular the preliminary and / or final commissioning, the certification, a point in time during regular operation, a point in time during a change of ownership and / or after completion of a maintenance phase or repair phase of the photovoltaic system, the first point in time is the commissioning of the photovoltaic system and the second point in time is the end of a predetermined start-up phase of the photovoltaic system, the first point in time is the Commissioning of the photovoltaic system and the second point in time is the end of a predetermined simulation interval, the first point in time is the end of a previous simulation interval and the second point in time is the end of the predetermined simulation interval, the second point in time is the point in time of a change in the operator of the photovoltaic system, the second point in time is the point in time of a change in insurance for the photovoltaic system, and / or the first point in time is the point in time of a new financing of the photovoltaic system and / or the point in time of a partial or complete revision of the photovoltaic system, in particular a repowering.

19. Method according to one of the preceding claims, characterized in that the system parameter is determined based on a system parameter determined for a previous second period, the system parameter for the second period is compared with the system parameter determined for the first period, the system parameter for the first period is determined at least partially based on a modification of the system parameter and / or the operating parameter, the maintenance parameter and / or at least one maintenance parameter for the second period, wherein preferably the first period and the second period have approximately identical lengths, in particular one year,the second period is shorter than the first period or the second period is longer than the first period and / or the interval between the determinations is arbitrarily determined and / or the assessment of the parameters of a photovoltaic system is determined and determined for the first time after any period of operation.

20. A data processing device comprising at least one processor configured to carry out a method according to any one of the preceding claims.

21. A computer program comprising instructions which, when the program is executed by at least one computer, cause the computer to carry out a method according to one of claims 1 to 19.

22. A computer-readable (storage) medium comprising instructions which, when executed by a computer, cause the computer to carry out a method according to any one of claims 1 to 19.

Images