Designing an optimal solar system is complex and time-consuming, involving the selection and configuration of panels, inverters, and Module Level Power Electronics (MLPEs).
But it doesn’t have to be.
AutoStringer is a powerful automation tool that can help solar designers achieve optimal results quickly and efficiently. This not only results in optimal energy production and accuracy, but it delivers time and cost savings to boot.
AutoStringer uses a range of inputs to achieve that, including panel characteristics (voltage, current, manufacturer specifications), inverter and DC optimizer specifications (such as input voltage, input current, output voltage, and the number of physical inputs), and panel layout (location, orientation, tilt).
Mixed integer linear programming (MIP) techniques are used to determine the best configurations for maximum energy production and minimal cost. It considers a number of constraints, including the electrical parameters of the available devices, site temperature, panel characteristics, and compatibility of the components. It also considers string power, number of strings, and total panel power to select the optimal type and number of inverters and optimizers. Once the optimal configuration is determined, AutoStringer optimizes the string path to reduce wiring cost by using a combination of Traveling Salesman and Snake algorithms — which we’ll explain later in this blog.
AutoStringer dramatically cuts down the time required for designing solar panel stringing and configurations. AutoStringer will return results in under 2 seconds on average and under 2.5 seconds 90% of the time. This allows you to focus on other crucial aspects of your projects, leading to faster turnaround times and increased productivity.
Let’s take a closer look at how it all works.
How AutoStringer works
AutoStringer’s design process consists of several steps:
1) Component validation and compatibility check
AutoStringer begins by checking that the components’ properties are valid and compatible with the other components in the system. For instance, if an MLPE device lacks sufficient power to handle the specified panel type or if an inverter is not compatible with MLPEs, AutoStringer will flag the issue.
2) Configuration generation
Following component validation, AutoStringer generates possible configurations for a solar panel system based on the inverters, optimizers, and panels selected. AutoStringer creates all possible combinations of string lengths using the panel type and considers the provided inverter and optimizer types that meet compatibility criteria. This step narrows down the search domain for the integer linear programming (ILP) solver, leading to faster optimization.
3) Panel grouping
AutoStringer uses a graph-based approach and the breadth-first search algorithm to group panels into sets suitable for stringing. If required based on the design, AutoStringer ensures that all panels on a given string have the same azimuth to optimize the system’s energy output.
4) BOS component selection
In the BOS (Balance of System) component selection stage, AutoStringer identifies the optimal combination of inverters and MLPE devices, such as DC optimizers or microinverters. The goal is to minimize cost and the number of unstrung panels.
AutoStringer begins by evaluating different design configurations based on the types of BOS components available. For systems using inverters with MPPTs, it considers panels in series to form strings, which are then linked to the input ports of the inverters’ MPPTs. This approach requires careful consideration of factors such as the number of MPPTs, their input voltage and current ratings, and how they align with the panel strings.
For systems equipped with DC optimizers or microinverters, each device may cover one or multiple panels or have multiple inputs, allowing for several strings to be connected to each optimizer or microinverter. AutoStringer accounts for these variations by optimizing the number of panels per optimizer or microinverter and ensuring that each string is appropriately matched to the device’s capacity. For microinverters, the panels are connected directly to the appropriate microinverter ports. In designs that use DC optimizers, the panels are first connected to the DC optimizers, and then the strings from the DC optimizers are connected to the appropriate ports on the inverters.
AutoStringer also supports designs that incorporate external MPPTs. In this case, multiple strings of panels can be connected to the input ports of the external MPPTs, and the external MPPTs are then connected to the input ports of the inverters. Compared to multi-input DC optimizers, external MPPTs offer more flexibility regarding panel coverage in some scenarios
Using a set of solar panels (layout), the available types of BOS and MLPE components that can be used in the system, and other auxiliary information, such as the temperature variation at the site, AutoStringer determines which BOS and MPPT components to choose. It accounts for the minimum and maximum string length, the minimum and the maximum number of strings, and the maximum total power derived from the electrical data of the inverters. These constraints ensure that the design is feasible and meets various electrical and mechanical requirements, while minimizing the total cost of the selected inverters and the number of unstrung panels.
The output of the selection stage is the optimal combination of BOS and MLPE components (inverters and optimizers) and the corresponding configurations, which are used in the subsequent stage.
5) String routing
The string routing stage of AutoStringer follows the completion of the previous stage. It aims to create the optimal stringing configuration by connecting individual panels to the inverters/DC optimizers in a way that is consistent with the selected stringing configurations. It’s worth noting that the string routing process differs slightly for designs with microinverters. In these designs, there is no stringing among the microinverters themselves. Instead, the microinverters are connected to individual panels, and the stringing is limited to the panels. The string routing algorithm accounts for this difference and constructs the strings accordingly.
The string routing stage of AutoStringer follows the completion of the previous stage. It aims to create the optimal stringing configuration by connecting individual panels to the inverters/DC optimizers in a way that is consistent with the selected stringing configurations.
An initial tour passing through the panels in each panel group is constructed per panel group.
Once the tour is constructed for each panel group, AutoStringer breaks up the tour into strings and assign the strings to the corresponding inverters.
When DC optimizers are used in a design, the first step is to select the best DC optimizer type based on the minimum cost per panel coverage by that DC optimizer type. Once the DC optimizer type is selected, the panels are strung together and attached to the DC optimizer input ports using a simple greedy algorithm. In this case, the strings may be composed of multiple panels and multiple strings may be connected to a single DC optimizer.
Once the panels are strung and attached to the corresponding optimizers, the stringing among the DC optimizers is handled in the same way as the stringing among the panels in the absence of DC optimizers (dealing with strings of DC optimizers instead of strings of panels). In this case, multiple strings of DC optimizers may be connected to an inverter, and the strings are assigned to inverters using a simple greedy algorithm.
Overall, the string routing stage of AutoStringer aims to create the optimal stringing configuration for the solar panels and MLPEs while adhering to the stringing configurations selected by the ILP solution from the former stage.
It’s worth noting that the string routing process differs slightly for designs with microinverters. In these designs, there is no stringing among the microinverters themselves. Instead, the microinverters are connected to individual panels, and the stringing is limited to the panels. The string routing algorithm accounts for this difference and constructs the strings accordingly.
6) Design output and visualization
The final step involves creating a list of inverters with their attached panel strings or DC optimizer strings which is used to create a graphical representation of the solar panel layout that is then visualized in Aurora.
For more information
Want to learn more? Reach out and schedule a quick, personalized demo to get all your specific questions answered.
