Method for the customized production of at least one photovoltaic panel

Abstract

A method for the customized production of at least one photovoltaic panel, wherein the method comprises the following steps: entering input data into a prediction module, wherein the input data comprising at least one of panel design characteristics, operational limits, desired panel characteristics, material characteristics, predicting predicted panel characteristics and / or predicted production parameters using the prediction module based on the input data, customized production of the at least one photovoltaic panel based on the input data, recording actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel during or after its production, updating the prediction module based on one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or one or more production differences between the actual production parameters and the predicted production parameters.

Description

[0001] Method for the customized production of at least one photovoltaic panel

[0002] The invention relates to a method for the customized production of at least one photovoltaic panel. The invention further relates to a photovoltaic panel production system for the customized production of at least one photovoltaic panel.

[0003] Building Integrated Photovoltaics (BIPV) are rapidly gaining popularity. BIPV offers the possibility to integrate solar panels into various building elements, for example into the structural parts of buildings, such as roofs and facades. These customized solutions are attractive due to their aesthetic and energy-efficient benefits, but they also pose challenges.

[0004] One problem with BIPV is the high degree of variability: each project is unique, making standard production methods and fixed datasheets not always applicable. This lack of standardization leads to significant uncertainties in the prediction of both the performance and production costs of new solar panels. One obstacle within the current BIPV production is, for example, the prediction of the electrical characteristics of new solar panels. Operators often have to partly calculate and partly rely on intuition to estimate what the power output of a newly designed solar panel will be. This approach results in a margin of error, making it difficult to provide customers with reliable data when preparing quotations. Moreover, it is too costly and time-consuming in practice to make a physical prototype for each project in order to test the performance.

[0005] In addition, there is the production cost calculation, in which both the production time and the required materials are predicted with uncertainty. Here too, operators base their estimates partly on mathematical models and partly on experience, which leads to deviations between expected and actual costs. This has direct consequences for financial planning and can lead to budget overruns.

[0006] Finally, the diversity of BIPV products ensures that optimizing the production process remains an ongoing challenge. Without a standardized process, production managers are often dependent on their experience to adjust production. The current management of complex processes is inefficient and limits the ability to continuously improve.

[0007] It is an objective of the invention to provide an improved method that optimizes the production process of a photovoltaic panel, or at least to provide an alternative method.

[0008] The objective of the invention is achieved by a first aspect of the invention, in which a method is provided for the customized production of at least one photovoltaic panel. The photovoltaic panel is for example a BIPV (Building Integrated Photovoltaics) panel or building- integrated photovoltaic panel. The method comprises the steps of:

[0009] - inputting input data into a prediction module, wherein the input data comprising at least one of o panel design characteristics, o operational limits, o desired panel characteristics, o material characteristics,

[0010] - predicting predicted panel characteristics and / or predicted production parameters using the prediction module based on the input data,

[0011] - customized producing of the at least one photovoltaic panel based on the input data,

[0012] - recording actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel during or after its production,

[0013] - updating the prediction module based on: o one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or o one or more production differences between the actual production parameters and the predicted production parameters.

[0014] The method provides for the customized production of at least one photovoltaic panel. Each individual photovoltaic panel must be custom-made. Each photovoltaic panel to be custom-made, for example a BIPV panel, has specific requirements, such as dimensions, shapes, colours, choice of materials, performance, and specifications. These requirements are adjusted to meet architectural and aesthetic demands.

[0015] The method according to the invention provides a step of inputting input data into a prediction module. The prediction module is preferably an automatic system. The prediction module is for example a mathematical model with a plurality of setting parameters. These setting parameters are, for example, variable. The prediction module uses, for example, a machine learning algorithm to adjust values of the setting parameters. The input data comprise panel design characteristics, operational limits, desired panel characteristics, and / or material characteristics. Panel design characteristics refer to the technical and structural properties of the photovoltaic panel as it is actually designed. This comprises properties such as dimensions, shape, choice of material, electrical characteristics (such as voltage, current, or power), colour, transparency, insulation, and the physical integration into a building structure. These are the specific features determined during a design process to enable the photovoltaic panel to function and fit within the set requirements. Desired panel characteristics refer to the properties that are desired in the photovoltaic panel. These are the desired electrical performance, such as a certain power or energy yield, a specific colour, transparency, or insulation. These desired properties form the basis for the design, but may not always be exactly realizable due to technical constraints or other factors.

[0016] Based on the input data, the prediction module predicts predicted panel characteristics and / or predicted production parameters. Predicted panel characteristics are the expected electrical and / or physical properties of a photovoltaic panel, such as power output, voltage, current, efficiency, transparency, insulation, etc., before the panel is produced. Predicted production parameters refer to the expected values of the production process to produce the photovoltaic panel. For example, total production time, operator labour hours, production costs, production losses, or material usage.

[0017] After the input data are entered, the at least one photovoltaic panel is custom- produced based on the input data. The production is for example semi-automatic or fully automatic. During or after the production of the at least one photovoltaic panel, actual panel characteristics and / or actual production parameters are recorded. Recording preferably takes place automatically. For example, an automatic recording module records the actual panel characteristics and / or actual production parameters. The recording module is for example part of a control unit. Recording the actual production parameters ensures that each step of the panel's production is monitored. The actual production parameters are actual values measured during or after the production process. For example, the actual production time required to produce the panel, including any delays or adjustments. The actual panel characteristics refer to the measured characteristics of the final, produced panel. They provide an exact representation of the performance and physical properties of the panel. For example, the actual power output the panel generates under standard test conditions.

[0018] Subsequently, the prediction module is updated based on one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or one or more production differences between the actual production parameters and the predicted production parameters. Updating the prediction module preferably occurs automatically. By automatically updating the prediction module, the prediction module is continuously optimized during the production process. Automatically means that the prediction module is updated without human intervention or manual actions. For example, the prediction module updates itself. Alternatively, the prediction module is updated indirectly, wherein the prediction module is controlled by a control unit. The prediction module can be part of the control unit or the prediction module can be a separate unit. The control unit, for example, receives the recorded actual panel characteristics and / or actual production parameters from the recording module. The control unit processes these recorded data, being able to generate a feedback signal. This feedback signal is representative of the one or more panel differences and / or the one or more production differences. The feedback signal is sent by the control unit to the prediction module, after which the prediction module is updated.

[0019] The updating of the prediction module is not done based on a single data point (parameter or characteristic), but on a set of data comprising recorded actual panel characteristics and / or recorded actual production parameters of the at least one photovoltaic panel during or after its production. This allows for an accurate estimation of the set of parameters and characteristics of the next photovoltaic panel to be custom-produced and the corresponding production parameters.

[0020] Each newly produced photovoltaic panel is always customized and results in a new product with its own characteristics. Using the method according to the invention, a complete automation of customization in photovoltaic panels can be provided. Through the combination of automatic updating based on the recorded parameters and / or characteristics and the customization of the panels, the characteristics of new photovoltaic panels and the parameters of the production process can be predicted more accurately and more quickly.

[0021] In an embodiment, the method according to the invention comprises the step of using the updated prediction module for the customized production of a new photovoltaic panel different from the at least one photovoltaic panel. By using the updated prediction module, the production process is dynamically adjusted to produce the newly to-be-manufactured photovoltaic panel. The new photovoltaic panel differs from the previously produced photovoltaic panel. This difference may, for example, lie in the dimensions, the power output, the choice of materials, or other technical properties, depending on the specific requirements for the new photovoltaic panel.

[0022] The prediction module can be continuously improved through feedback from previous productions, resulting in an increasingly accurate and efficient production process. As a result, each new photovoltaic panel can be better tailored to the specific needs of a customer or the conditions in which it is installed.

[0023] In an embodiment, the method according to the invention comprises recording, in a recording module, actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel during or after its production. The recording module determines and collects data regarding various characteristics of the photovoltaic panel, such as dimensions, power output, and efficiency. In addition, the recording module determines and collects actual production parameters, such as material usage, production speed, and any faults that occur during the production process. The recording module comprises, for example, a plurality of sensors. These may be physical sensors that, for instance, determine the dimensions of the photovoltaic panel. Additionally or alternatively, these may be electrical sensors configured to determine, for example, the electrical properties of the photovoltaic panel. The recording module comprises, for example, a communication module that provides a connection between the recording module and another unit, such as a control unit or a central database, enabling data transfer from the recording module to the other unit. This data transfer may take place via wired connections, such as Ethernet, or wireless communication, such as Wi-Fi or Bluetooth.

[0024] In an embodiment, the panel design characteristics or desired panel characteristics comprise at least one of electrical characteristics, such as voltage, current, or power, colour, transparency, insulation. The electrical characteristics, such as voltage, current, or power, determine the performance and efficiency of the photovoltaic panel. These characteristics can be customized to meet specific installation requirements. Visual characteristics such as colour and transparency are important when the panel is integrated into architectural applications, such as with BIPV panels. The structural characteristics, such as insulation, contribute to the durability, safety, and energy efficiency of the panel. The advantage is that the panel characteristics can be flexibly and custom-tailored so that they not only meet the technical requirements for energy yield but also fit aesthetically and structurally with the specific application, such as the integration of solar panels into buildings.

[0025] In an embodiment, the operational limits comprise at least one of maximum and / or minimum dimensions of the at least one photovoltaic panel, maximum length of a series of solar cells of the at least one photovoltaic panel, minimum and / or maximum spacing between solar cells and / or series of solar cells. The operational limits refer to the specific technical and operational constraints related to the equipment and systems used for manufacturing, moving, and assembling the photovoltaic panels. For example, the maximum and minimum dimensions of the photovoltaic panels define the size limits that the production machines can handle.

[0026] In an embodiment, the material characteristics comprise at least one of solar cell specifications, connection element specifications, glass specifications, layer specifications. The material characteristics refer to the specific properties and technical requirements of the various components used in the production of photovoltaic panels. The solar cell specifications are the technical properties of the solar cells. This concerns the type of solar cell, such as monocrystalline, polycrystalline, or thin-film, as well as their efficiency. The connection elements refer to, for example, the interconnections between the solar cells, with specific requirements including conductivity, strength, and corrosion resistance. The glass used as the protective top layer for the solar cells must meet certain criteria, such as thickness, light transmission, scratch resistance, and impact resistance. Layer specifications concern the properties of the different layers in the panel, such as the anti-reflective coating, the rear glass, the backsheet, and any additional protective layers.

[0027] In an embodiment, the production parameters comprise at least one of production time, operator labour hours, production losses, material usage. The production parameters are variables that characterize the production process of photovoltaic panels.

[0028] In an embodiment, the method according to the invention comprises the step of storing the actual panel characteristics and the actual production parameters. These data are, for example, stored in a (central) database or in a local storage of the recording module. The data may, for example, be stored temporarily.

[0029] In an embodiment, the step of predicting the predicted panel characteristics and / or the predicted production parameters using the prediction module comprises predicting the predicted panel characteristics and / or the predicted production parameters using the prediction module based on the stored actual panel characteristics and / or the stored actual production parameters. The prediction module preferably uses stored actual panel characteristics and production parameters to generate accurate predictions. For example, the prediction module analyses the stored data and identifies patterns, possibly using statistical and machine learning techniques.

[0030] In an embodiment, the step of updating the prediction module comprises updating setting parameters of the prediction module such that future predictions take into account actual panel characteristics and actual production parameters. Setting parameters are specific values or configurations within the prediction module that determine the operation of the prediction module. These setting parameters determine how the prediction module interprets and uses (stored) data and / or input data to generate future predictions. Updating the setting parameters of the prediction module may comprise, for example, adjusting certain coefficients of the underlying algorithm, such as linear regression.

[0031] In an embodiment, the method according to the invention comprises the step of predicting a cost price for the customized production of the at least one photovoltaic panel using the prediction module based on the input data. The input data used for predicting the cost price comprise, for example, current material prices, such as the cost of solar cells. Additionally, production parameters such as production time, required labour, and expected production losses are taken into account. The desired panel characteristics, including size, colour, and technology (e.g., monocrystalline or polycrystalline), are also considered. During the prediction process, the prediction module performs calculations based on the collected input data. This may include the prediction module calculating production costs per unit by taking into account total material costs, hourly labour costs, and other indirect costs such as overhead and energy consumption. Based on these calculations, the module generates an estimate of the total cost price for customizing the solar panels. This makes it possible to minimize the deviation between expected and actual costs. This enables manufacturers to achieve more accurate budgeting.

[0032] In an embodiment, recording the actual panel characteristics comprises measuring electrical characteristics of the at least one photovoltaic panel. During measuring, panel characteristics such as voltage, current, and power are recorded using sensors and / or measuring instruments. These measurements can be performed both in laboratories and in the production environment.

[0033] In an embodiment, recording actual panel characteristics comprises inspecting the at least one photovoltaic panel. This means that the panel is visually and technically checked to determine whether it complies with the design specifications and quality standards. The inspection can take place during or after the production process and comprises both physical and functional evaluations of the photovoltaic panel.

[0034] The inspection is performed, for example, using an advanced image recognition system, such as a line scanner, which scans the entire surface of the photovoltaic panel and collects data. This data is used to automatically detect defects, such as air bubbles between the glass, broken photovoltaic cells, colour differences, or cracks.

[0035] The image recognition software uses, for example, machine learning algorithms trained on a broad dataset of produced photovoltaic panels to accurately identify and classify deviations. The system can distinguish between critical and non-critical defects, and the results of the analysis are stored in a central database. This database contains data relating to dimensions, glass type, colour, and other relevant panel characteristics. This enables statistical analyses to be performed and predictions to be made about the likelihood of defects per square meter.

[0036] The output of the machine learning algorithms can then be validated by operators, who review the data and make corrections if necessary. This validated data can be used to further optimize quality control, inform customers about the quality of their specific product, or generate predictions for future productions with the aim of reducing defects and improving production processes. Furthermore, this data can contribute to process improvements by providing insight into common defects and their causes. In an embodiment, the method according to the invention comprises updating a material stock based on the actual production parameters. When the actual quantity of materials used, such as solar cells, glass, or connection elements, is recorded, this information can be transmitted to the stock management system. This system then updates the available stock, so that there is always an accurate overview of what is still available for future production. By recording material consumption in real-time, the stock can be updated automatically. This helps to order new materials in time, prevent material shortages, and minimize overstock.

[0037] In a second aspect of the invention, a photovoltaic panel production system is provided for the customized production of at least one photovoltaic panel, the system comprising:

[0038] - a user interface for inputting input data comprising at least one of: o panel design characteristics, o operational limits, o desired panel characteristics, o material characteristics,

[0039] - a prediction module configured to predict predicted panel characteristics and / or predicted production parameters based on the input data,

[0040] - a production machine configured to produce the at least one photovoltaic panel based on the input data,

[0041] - a control unit connectable to the user interface, wherein the control unit is configured to control the prediction module and the production machine, wherein the production system is configured to:

[0042] - record the actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel during or after its production,

[0043] - update the prediction module based on: o one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or o one or more production differences between the actual production parameters and the predicted production parameters.

[0044] The user interface is, for example, an interactive platform through which users can input input data required for the production of customized photovoltaic panels. This user interface may contain various input fields and options for specifying relevant parameters, such as panel dimensions, electrical characteristics (such as voltage, current, and power), material specifications (e.g., type of solar cells or glass), and production requirements. After inputting the input data into the user interface, the information is passed to the control unit, for example via an input signal, after which the control unit is configured to control the prediction module and the production machine. The control unit may comprise the prediction module. The input data is then processed by the prediction module to predict predicted panel characteristics and / or predicted production parameters based on the input data. The production machine receives the required production specifications from the control unit. This comprises details such as the exact dimensions of the panel, the positioning and assembly of the solar cells, the electrical connections, and finishes. Based on this data, the production machine starts customizing the photovoltaic panel. During and after the production of the photovoltaic panel, the actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel are recorded. The recording is performed, for example, by a recording module. The recording module is, for example, part of the control unit. The control unit is configured to update the prediction module based on one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or one or more production differences between the actual production parameters and the predicted production parameters. By updating and thereby optimizing the prediction module, future predictions become more accurate for producing a new customized photovoltaic panel.

[0045] In an embodiment, the control unit is further configured to generate a feedback signal, wherein the feedback signal is representative of the one or more panel differences and / or the one or more production differences, and to send the feedback signal to the prediction module, wherein the prediction module updates itself based on the feedback signal. In this way, an automatic feedback loop is established between the control unit and the prediction module.

[0046] The feedback signal generated by the control unit and sent to the prediction module may be added automatically or manually. Automatic feedback occurs without operator intervention. The prediction module is updated directly based on collected data and the analyzed one or more panel differences and / or one or more production differences. This results in continuous and optimized self-adjustment of the prediction module, which increases efficiency and minimizes human errors.

[0047] Alternatively or additionally, the feedback may be added manually, whereby an operator reviews the collected data and decides whether and how the prediction module should be updated. This may be useful in situations where human expertise is needed to assess the validity of the data or when exceptional circumstances arise that a control unit may not adequately process. The manual feedback can be inputted via the user interface, allowing the operator to suggest or implement specific adjustments to the prediction module. The invention will be further explained below with reference to the figures, which illustrate, in a non-limiting manner, embodiments of the invention. The figures show in:

[0048] Fig. 1: a flowchart of an embodiment of the method according to the invention for the customized production of at least one photovoltaic panel;

[0049] Fig. 2: a schematic representation of an embodiment of a photovoltaic panel production system according to the invention for the customized production of at least one photovoltaic panel.

[0050] Fig. 1 illustrates a flowchart of an embodiment of the method according to the invention for the customized production of at least one photovoltaic panel. The method provides for the customized production of at least one photovoltaic panel. Each individual photovoltaic panel must be custom-made. Each photovoltaic panel to be custom-made, for example a BIPV panel, has specific requirements, such as dimensions, shapes, colours, and choice of materials. These requirements are adjusted to meet architectural and aesthetic demands.

[0051] The method provides a first step (100) of inputting input data into a prediction module. The prediction module is an automatic system. The prediction module is, for example, a mathematical model with a plurality of setting parameters. These setting parameters are, for example, variable. The prediction module, for example, uses a machine learning algorithm to adjust values of the setting parameters. The input data comprises panel design characteristics, operational limits, desired panel characteristics, and / or material characteristics. Panel design characteristics refer to the technical and structural properties of the photovoltaic panel as it is actually designed. This comprises aspects such as the dimensions, shape, choice of materials, electrical characteristics (such as voltage, current or power), colour, transparency, insulation, and physical integration into a building structure. These are the specific features determined during the design process to ensure the photovoltaic panel functions and fits within the set requirements. Desired panel characteristics refer to the properties one wishes to achieve in the photovoltaic panel. These are the desired electrical performance, such as a specific power output or energy yield, a specific colour, transparency, or insulation. These desired properties form the starting point for the design, but due to technical limitations or other factors, they cannot always be exactly achieved.

[0052] The method provides a second step (101) of predicting, based on the input data, predicted panel characteristics and / or predicted production parameters using a prediction module. Predicted panel characteristics are the expected electrical and / or physical properties of a photovoltaic panel, such as power output, voltage, current, efficiency, transparency, insulation, etc., before the panel is produced. Predicted production parameters refer to the expected values of the production process to produce the photovoltaic panel. For example, the total production time, operator labour hours, production costs, production loss, or material consumption.

[0053] After the input data has been inputted, the at least one photovoltaic panel is custom- produced based on the input data (step 102). The production step (102) and the prediction step (101) may occur simultaneously. Production occurs, for example, semi-automatically or fully automatically. During or after the production of the at least one photovoltaic panel, actual panel characteristics and / or actual production parameters are recorded (103). Recording occurs automatically. For example, an automatic recording module records the actual panel characteristics and / or actual production parameters. The recording module is, for example, part of a control unit. Recording the actual production parameters ensures that each step of the panel’s production is monitored.

[0054] Next, the prediction module is updated (104) based on one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or one or more production differences between the actual production parameters and the predicted production parameters. The updating of the prediction module occurs automatically. By automatically updating the prediction module, the prediction module is continuously optimized during the production process. Automatically means that the prediction module is updated without human intervention or manual actions. For example, the prediction module updates itself. Alternatively, the prediction module is updated indirectly, wherein the prediction module is controlled by a control unit. The prediction module may be part of the control unit or may be a separate unit.

[0055] The updating of the prediction module is not done based on a single data point (parameter or characteristic), but based on a set of data comprising recorded actual panel characteristics and / or recorded actual production parameters of the at least one photovoltaic panel during or after its production. This allows for an accurate estimation of the set of parameters and characteristics of the next photovoltaic panel to be custom-produced and the associated production parameters.

[0056] Each newly produced photovoltaic panel is always custom-made and results in a new product with own characteristics. Using the method according to the invention, full automation of the customization of photovoltaic panels is enabled. By combining automatic updating based on the recorded parameters and / or characteristics and the customization of the panels, the characteristics of new photovoltaic panels and parameters of the production process can be predicted more accurately and more quickly.

[0057] The method comprises the optional step (105) of using the updated prediction module for the customized production of a new photovoltaic panel different from the at least one photovoltaic panel. By using the updated prediction module, the production process is dynamically adjusted to produce the new photovoltaic panel. The new photovoltaic panel differs from the previously produced photovoltaic panel. This difference may, for example, lie in the dimensions, the power output, the choice of materials, or other technical properties, depending on the specific requirements for the new photovoltaic panel. The prediction module is continuously improved through feedback from previous productions, leading to an increasingly accurate and efficient production process. As a result, each new photovoltaic panel can be better adapted to the specific needs of a customer or the conditions in which it is to be installed.

[0058] Fig. 2 shows a schematic representation of an embodiment of a photovoltaic panel production system according to the invention for the customized production of at least one photovoltaic panel. The photovoltaic panel production system (200) provides for the customized production of at least one photovoltaic panel. Each individual photovoltaic panel must be custom-made.

[0059] The photovoltaic panel production system (200) comprises a user interface (201). The user interface (201) is, for example, an interactive platform through which users can input input data (202) needed for the production of customized photovoltaic panels. The input data (202) comprises panel design characteristics, operational limits, desired panel characteristics, and / or material characteristics. The user interface (201) may include various input fields and options for specifying relevant parameters, such as panel dimensions, electrical characteristics (such as voltage, current, and power), material specifications (e.g., type of solar cells or glass), and production requirements.

[0060] After inputting the input data (202) into the user interface (201), the information is sent to a control unit (203) by means of an input signal (204). The control unit controls a prediction module (205) and a production machine (206). The control unit (203) controls the prediction module (205) by means of a control signal (207). In Fig. 2, the prediction module (205) is a separate unit. The control signal (207) is processed by the prediction module (205) to predict predicted panel characteristics and / or predicted production parameters based on the input data (202). The predicted panel characteristics and / or predicted production parameters are fed back to the control unit (203) by means of a prediction signal (208).

[0061] The control unit (203) controls the production machine (206) by means of a production signal (209). The production signal (209) comprises details such as the exact dimensions of the panel, the positioning and assembly of the solar cells, the electrical connections, and finishes. Based on this data, the production machine (209) starts customizing the photovoltaic panel.

[0062] During and after the production of the photovoltaic panel, the actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel are recorded by a recording module (210). The recording module (210) is, for example, part of the control unit (203). The recording module (210) comprises, for example, a plurality of sensors. These may be physical sensors that determine, for example, the dimensions of the photovoltaic panel. Additionally or alternatively, these may be electrical sensors configured to determine, for example, the electrical properties of the photovoltaic panel. The recording module (210) comprises a communication module that provides a connection between the recording module (210) and the control unit (203), wherein data transfer (211) can take place from the recording module (210) to the control unit (203). This data comprises the actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel recorded by the recording module (210). This data transfer (211) may occur via wired connections, such as Ethernet, or wireless communication, such as Wi-Fi or Bluetooth.

[0063] The control unit (203) subsequently updates the prediction module (205) based on one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or one or more production differences between the actual production parameters and the predicted production parameters. Specifically, the control unit (203) generates a feedback signal (212). The feedback signal (212) is representative of the one or more panel differences and / or the one or more production differences. The feedback signal (212) is sent to the prediction module (205), wherein the prediction module (205) updates itself based on the feedback signal (212). In this way, an automatic feedback loop is created between the control unit (203) and the prediction module (205). By updating and thereby optimizing the prediction module (205), future predictions become more accurate for producing a new customized photovoltaic panel.

Claims

CLAIMS1. A method for the customized production of at least one photovoltaic panel, wherein the method comprises the steps of:- inputting input data into a prediction module, wherein the input data comprising at least one of o panel design characteristics, o operational limits, o desired panel characteristics, o material characteristics,- predicting predicted panel characteristics and / or predicted production parameters using the prediction module based on the input data,- customized producing of the at least one photovoltaic panel based on the input data,- recording actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel during or after its production,- updating the prediction module based on: o one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or o one or more production differences between the actual production parameters and the predicted production parameters.

2. Method according to claim 1, further comprising the step of using the updated prediction module for the customized production of a new photovoltaic panel different from the at least one photovoltaic panel.

3. Method according to any one of the preceding claims, comprising recording, in a recording module, actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel during or after its production.

4. Method according to any one of the preceding claims, wherein the panel design characteristics or desired panel characteristics comprise at least one of- electrical characteristics, such as voltage, current or power,- colour,- transparency,- insulation.

5. Method according to any one of the preceding claims, wherein the operational limits comprise at least one of- maximum and / or minimum dimensions of the at least one photovoltaic panel,- maximum length of a series of solar cells of the at least one photovoltaic panel,- minimum and / or maximum spacing between solar cells and / or series of solar cells.

6. Method according to any one of the preceding claims, wherein the material characteristics comprise at least one of- solar cell specifications,- connection element specifications,- glass specifications,- layer specifications.

7. Method according to any one of the preceding claims, wherein the production parameters comprise at least one of- production time,- operator labour hours,- production losses,- material usage.

8. Method according to any one of the preceding claims, further comprising the step of storing the actual panel characteristics and the actual production parameters.

9. Method according to claim 8, wherein predicting the predicted panel characteristics and / or the predicted production parameters using the prediction module comprises predicting the predicted panel characteristics and / or the predicted production parameters using the prediction module based on the stored actual panel characteristics and / or the stored actual production parameters.

10. Method according to any one of the preceding claims, wherein updating the prediction module comprises updating setting parameters of the prediction module such that future predictions take into account actual panel characteristics and actual production parameters.

11. Method according to any one of the preceding claims, further comprising the step of predicting a cost price for the customized production of the at least one photovoltaicpanel using the prediction module based on the input data.

12. Method according to any one of the preceding claims, wherein recording actual panel characteristics comprises measuring electrical characteristics of the at least one photovoltaic panel.

13. Method according to any one of the preceding claims, wherein recording actual panel characteristics comprises inspecting the at least one photovoltaic panel.

14. Method according to any one of the preceding claims, further comprising updating a material stock based on the actual production parameters.

15. Photovoltaic panel production system for the customized production of at least one photovoltaic panel, the system comprising:- a user interface for inputting input data comprising at least one of: o panel design characteristics, o operational limits, o desired panel characteristics, o material characteristics,- a prediction module configured to predict predicted panel characteristics and / or predicted production parameters based on the input data,- a production machine configured to produce the at least one photovoltaic panel based on the input data,- a control unit connectable to the user interface, wherein the control unit is configured to control the prediction module and the production machine, wherein the production system is configured to:- record the actual panel characteristics and / or actual production parameters of the at least one photovoltaic panel during or after its production,- update the prediction module based on: o one or more panel differences between the actual panel characteristics and the predicted panel characteristics, and / or o one or more production differences between the actual production parameters and the predicted production parameters.

16. Photovoltaic panel production system according to claim 15, wherein the control unit is further configured to generate a feedback signal, wherein the feedback signal is representative of the one or more panel differences and / or the one or more productiondifferences, and to send the feedback signal to the prediction module, wherein the prediction module updates itself based on the feedback signal.

Images