Research and Development Notes

Xirasol Research and Development Notes

RandD Activities

Objective:

The objective of this project is to develop a new solar Dual Axis Tracking System for large scale solar operations that better utilises land and increases output over current mainstream single axis solar tracking systems. It is expected that this technology will be used on sites of 50 to 100KW for commercial business and utility scale solar farms of 1MW to over 1000 MW capacity. Such utility scale farms have their own sub-station for connection to the power grid.

Currently, most solar trackers available on the market work on a single axis tracking basis. Single axis trackers are normally aligned north and south, allowing the solar panels to arc from east to west. In contrast, dual access trackers have two axes of movement. Aligned with north-south and east-west, dual axis trackers have greater movement, allowing the solar panels to optimise the amount of solar energy captured throughout seasonal changes (i.e. the sun’s path changes from summer to winter and can produce 35%+ more power over a fixed mount and 15%+ more power over a single axis solar tracking system – depending on the location).

Current dual axis solar trackers are pole mounted and their first axis is rotational around the pole while the second axis tilts. Due to the height and movement of the tracking system and need to reduce shading, typically 4.5x the space of a fixed solar power mount is required.

Xirasol’s proposed new Dual Axis Solar Tracking Technology will have solar panels situated on both sides of a large horizontal beam. The solar panels can then rotate on a north south axis while tilting on an east west axis. The result of Xirasol’s proposed Dual Axis Tracker is that there are more panels in a smaller area and by taking shading into account, less land is used than fixed, single and current dual axis technology. Fitting solar farms with Xirasol’s new Dual Axis Technology will require less roads, less fences, less trenching, less cable, less labour and result in much more effective solar farms being built. It is intended that The levelized cost of electricity will be less with Xirasol’s new Dual Axis Solar Tracking Technology.

From the above proposal, Xirasol have developed the following technical objectives to guide their experimentation:

Achieve 30% material reduction via new design
Decrease required land space by 40%, while achieving increased energy output of 20%
Reduce assembly time by 20%
Allow up to 1.5MW to be tracked with only 2 motors, compared to the standard 15Kw.

Hypothesis:

From the above technical objectives, Xirasol propose the following hypotheses:

1. Via the use of spherical bearings and a 2nd axis push rod, can the new tracking system decrease required land space by 40%, while achieving increased energy output of 20%?

2. Can the new design allow material reduction of 30%, enabling assembly time decrease of 20%?

What New Knowlodge was Created:

With the increasing reliance on alternative energy sources, there is a present need to optimise energy capturing technology as to remove the pre-existing reliance on fossil fuels (ie. non-renewable resources). Current dual axis tracking technology is far too expensive and inefficient to be employed on a large-scale and Xirasol wish to bridge this gap via the development of their own dual-axis solar tracking technology.

If proven successful, the development will enable the following technical objectives:

• Achieve 30% material reduction via new design;

• Decrease required land space by 40%, while achieving increased energy output of 20%; • Reduce assembly time by 20%; &

• Allow up to 1.5MW to be tracked with only 2 motors, compared to the standard 15Kw.

These measures far exceed pre-existing technology employed internationally, and will ultimately improve land usage, product efficiency and energy outputs to a level that can relied upon for energy sources.

What was the experiment and how did it test the hypothesis?

Inherent to the complexity of the assignment, Xirasol have apportioned the experimental process of their project into the following stages (with the understanding that although some of these are integral to the progression of experimentation, they are inherently supporting activities for the purposes of this application):

Step 1: Undertake theoretical and technological research. Prior to practical experimentation, Xirasol saw it pivotal to discuss and outline the feasibility of their project with internal stakeholders.

Step 2: Project management and process design. This stage tests the practicality of theoretical concepts, and whether they would be beneficial in both technical and commercial circumstances. It is pivotal that the solution improves on the pre-existing single axis structure as to optimise solar energy capture.

Step 3: Prototype development. This stage attempts to replicate the theoretical concepts in a physical solution to enable the development of a suitable product to achieve the technical objectives. Such involved the selection of materials based on theoretical study that would be both able to withstand the harsh natural environmental conditions as well as still being able to achieve the technical objectives (ie. have rigidity and malleability).

Step 4: Undertake controlled environment experimentation. This stage involves the testing of the prototype in controlled environments where Xirasol can control that conditions the prototype is within and see whether the rigidity of the product satisfies the technical requirements. This must be undertaken prior to large-scale uncontrolled experimentation to determine its efficacy in small-scale experimentation environments.

Step 5: Undertake uncontrolled environment experimentation. After successfully progressing through controlled environmental experimentation, the prototype will be tested in uncontrolled environments. Specifically, Xirasol will be positioning their prototype units in the real world, and testing their efficacy in both the tracking of the sun (to optimise solar energy capturing) and having a rigidity to withstand harsh wind, rain, salt and solar conditions. This will be compared to theoretical results to deduce its efficiency overtime.

Step 6: Further prototype development. Inherent to the infancy of solar tracking technology and the novelty of Xirasol’s proposal, the applicant expects adjustments to be made to the prototype’s design dependent on experimental results.

Specifically, work undertaken in FY23 is positioned within steps 3-6, in which Xirasol is currently undertaken real- world experimentation and making amendments to their design. Xirasol continued experimentation on their phase 1 design, and have made improvements to their design based on initial experimentation results. This led to the development of the phase 2 design, which will be discussed further below.

To enable diversity in the real-world experimentation, Xirasol have positioned their tests in unique environmental conditions. The differing locations provides contrast and variability in many parameters including:

Terrain;
Soil types;
Geographic location;
Grid operator requirements;
Grid operating conditions; &

How did you evaluate or plan to evaluate results from your experiment?

Xirasol installed prototypes up to 400 kW (~4x larger than FY21) and begun the installation of a 640kW system. We are pushing prototype sizing up towards 1MW, which is hoped to be achieved in FY23. Such configurations have uncovered issues in relation to installation/deployment techniques that were protected in the 100kW systems, requiring some design amendments. Specifically, this was in relation to the centralised motor system and the interactions with the related software package. Such amendments have not been in the field long enough yet to deliver a large cache of data, albeit, data is now accruing and analysis is ongoing.

Xirasol deployed systems across different Australian climatic conditions. We have installed systems in areas of varying solar resource across Victoria (Melbourne for low and Mildura for high), in tropical climates in North Queensland and in higher wind settings in coastal WA. We hope the differing climatic conditions will provide differing performance data sets. In relation to the Mildura installation, Xirasol will compare our actual output data against theoretical performance data for other tracking systems and hopefully verify that the Xirasol system is indeed producing greater power output under high solar irradiance conditions.

The Melbourne plant, when completed, will give us contrasting results for a much lower solar irradiance. We want to verify if the advantage of Xirasol dual axis tracking over other conventional single axis trackers is constant or if the output is different in some way due to higher/lower solar irradiance. Deployment in high wind and high/tropical rainfall will also give us valuable performance data as and when we can completed these installs.

Further, Xirasol to test practical efficacy, Xirasol installed a number of prototypes to power agricultural water pumps, and general electricity supply.

FY23 Results:

- Continued with installation of 600 kW project in Melbourne, providing a chance to investigate tracker performance under lower solar irradiance. However, completion has been delayed due to the host landowner, with full installation for testing anticipated to be completed in FY24. The tracking system from Xirasol / Sunflower was completed in the 23 year and only contractor works outside Xirasol to be completed in the 24 year.
- Continued with installation of 2,700 kW tracker systems at locations in WA, VIC and QLD, with systems varying in size from 53 kW to 600 kW and providing insights into varied climate and location types. Installation is completed in VIC, with installation underway in WA and QLD. There have been some delays to these installations due to covid lockdowns, border restrictions, and international shipping delays. Works are now nearing completion. All works by Xirasol / Sunflower were completed in the 2022/23 year with only external contractor works / customer works to do with grid after this period.
- There have been other installation challenges due to very soft ground at the sites. Some design changes were also needed to accommodate different solar panels (as manufacturers continue to update design of solar panels, which must be reflected in the design of the Xirasol system).
- While the regional VIC installations are complete, power surges on the electricity grid (caused by off site, high capacity water pumps turning on and off) created control issues on the tracker installations. Eventually the issue was diagnosed, and improvements were made to electrical, electronics and software to better protect the trackers from electrical interference from users on the power grid.
- Further analysis has indicated that the wind speed gauge and data collection must be improved. This is important as at certain wind speeds the tracker must be placed in a safe position to prevent wind damage. The safe position moves the solar panels to a horizontal aspect, thus presenting the least surface area to gusting and or high winds.

If you reached conclusions from your experiments in the selected income period, describe those conclusions:

Ultimately, the results from internal testing are indicative of a proven hypothesis via the following:

• Reduced land use. Specifically, initial experimentation indicates that employing the dual-axis structure will enable Xirasol to position the solar tracking product more closely together, with minimal impact of shading on the efficacy of solar tracking and capture. Results indicate that 8m seems to be the most effective distance between rows in order to enable enough space for the capture of solar energy during times the sun is at its lowest point;

• Optimised capture of solar irradiation. Experimental results indicate that the dual-axis solar tracking units drastically improves on the capture of solar irradiation, which in turn improves the efficacy of solar energy technology; &

• Decreased overrun on material use. Inherent to the dual-axis structure and the ability for the technology to track the sun during its natural arc, there is less need to develop solar panels facing in opposite directions. This in turn reduces the amount of materials used in the development of solar capture farms.

The phase 2 design is undoubtedly an improvement on the initial design, recorded significantly greater results in both structural and malleability measures. Testing in more practical environments was undertaken in FY22 and continued into FY23, which saw the installation of various size and configuration of solar panels across Australia. It is hoped such installations will provide a diverse range of results, allowing Xirasol to quantify the efficacy of their design in varying practical environments and determine the relationship to temperature, solar radiance, wind, rainfall, etc. While further prototype installations have been completed, there have been some delays due to issue with landholders, however, these issues are expected to be resolved in the coming months.

Analysis of long term data, collected from the initial installations along with field observations is allowing us to identify and implement incremental gains in terms of both electrical output and also structural stability / longevity. We are also identifying similar gains in terms of installation techniques and protocols. One of the key examples of this is various improvements in both design and installation techniques that help to make the tracker more resistant and tolerant to wind load and wind induced tortional loading.

For completeness, Xirasol note the experimentation philosophy has been successful to date, uncovering various technical inefficiencies in the phase 1 design and enabling the redevelopment of technical objectives and hypothesis. Such philosophy will continue in future experimentation, with a hope the aforementioned objectives will be confirmed by December 2024.

Briefly describe supporting activities:

Development of novel componentry for Spherical Bearings, 2nd Axis Push Rods and Dual Axis Movement Arc’s. Work performed as part of this supporting activity is as follows:

Crossbar – changes and additional strengthening
Saddle – updated design and additional strengthening
Upgraded large Standa’s to steel from Aluminium
New Centre drive shafts
Urethane flex disc with a fabricated yolk.
A bearing alignment and mounting method has been developed to suit the low tolerance drive tube material.
Updated centre structure with crescent
Software development
Ongoing repairs and maintenance
Changes to wiring looms for ease of installation and protection of cables.
The following activities were required to support the core activity:
Create a new welding cell for the fabrication of both crescent and drive module frames.
This is and continues to be a major development project in terms of improving the manufacturing of the prototypes. In a nutshell, we are converting and merging two manual welding fixtures into one full robotic welding operation. In order to facilitate robotic welding slight design changes have been made to both components. Sub-assemblies are also incorporated on satellite fixtures to the mane assembly fixture.

What was the hypothesis?

In addition to the structural component of the R&D, there is also a software component that helps enable the system to maximise energy generation, predict potential mechanical failures and help increase the overall asset life.

The software component of the R&D centres on the integration of the Programmable Logic Controller (PLC) with the several pieces of disparate hardware that are needed to drive / position the tracker to optimise solar energy capture and to collect different data sets used in system control and monitoring. A PLC is a type of industrial digital computer that can be used to control various processes and machines. This level of integration is extremely complex and has not been achieved before.

The below explanation provides an overview of how the software component attempts to integrate with the physical infrastructure of the solar tracking system being developed in Core Activity 1.

•The solar panels must be able to accurately track the sun all day. The active tracking is implemented in the first layer of integration via the use of motors turning transmissions that are driven with Variable Speed Drives (VSD). The VSD’s are a standard industrial device that not only allow freely selectable speed and direction of the motors but also provide data on how much power is being used to implement the movement. This power value is logged and will be used to eventually develop predictive maintenance regimes along with fault conditions and status.

• Accurate closed loop positioning of the array is achieved in 2 axis of movement by incorporating a multi- axis MEMS position sensor fixed to the moving structure. This allows a set-point to be followed.

• The second layer of control is the determination of the actual sun position relative to the array. Giving the function block the inputs of latitude/ longitude, time of day and elevation, the sun position is calculated continuously. The function block primary outputs are given as rotation of azimuth and elevation in degrees absolute. Secondary outputs are numerous like sun rise and set times and length of day. Absolute accuracy here is crucial to enable the technical objectives to be met.

• The third layer brings in the mechanical structure that operates in a rotation of north and south, and east and west, but this movement does not correlate to the suns positioning. This is where the core activity work begins. Significant research and development is needed to create a new tool that converts the detailed sun position data into angular degrees across an additional 2 planes. However, physical constraints of the mechanics mean that the solar panels can’t directly aim at the sun 100% of the time. This is where back-tracking algorithms are needed.

• Back-tracking uses the solar array axes to maintain the solar panels in the full sunlight close to 100% of the time. This involves the panels being positioned at an ever increasing oblique angle to the sun. Back tracking is deployed during the times of day when the sun is lowest in the sky.

From the above proposal, Xirasol have developed the following technical objectives to guide their experimentation:

• Development of a series of algorithms to allow remote automated control of solar panel positioning for seasonal and daily sun curvatures to increase energy output by 20%

• Automatically predict and subsequently detect severe weather patterns and automatically adjust the position of the panels for real time protection

From the above technical objectives, Xirasol propose the following hypotheses:

• Via the development of the accompanying software, can the new tracking system follow the sun’s natural curvature, irrespective of seasonal or daily changes, to increase energy output by 20%?

• Can the software automatically predict and detect severe weather conditions and adjust the solar panels to a position that does not subject them to damage to increase lifespan by 20%?

What new knowledge was this core activity intended to produce?

To accompany hardware, Xirasol developed an algorithm that controls solar panel positioning. Due to the novelty of the tracker, no proprietary software could be aligned with the Xirasol system. A new algorithm was developed to suit Xirasol’s specific needs.

Further, due to different environmental conditions, the forever changing climate and sun curvature, a universal algorithm cannot be applied. There needs to exist a level of difference to accommodate the specific environment conditions. Specifically, the variability in the algorithm accounts for:

Weather conditions including wind speed, rainfall and ambient temperatures; &
Relative sun curvature.

New knowledge will be generated via the development of an algorithm that will:

• Control the solar panel positioning (with regard to the specific weather conditions and sun curvature) to improve energy output by 20%; &

• Automatically predict and subsequently detect weather patterns and adjust the position of the panels in real- time.

Please explain what sources were investigated, what information was found, and why a competent professional could not have known or determined the outcome in advance:

Xirasol undertook a literature review, followed by consultations with experts (eg. academics, software engineers, etc.) to deduce current uncertainties. Xirasol deduced there was no publicly accessible knowledge that would achieve the technical objectives, and further:

• Where tracking algorithms exists, they are unable to integrate seamlessly to the dual-axis structure.

• No tracking system was flexible enough to deduce the positioning of the sun and adjust both the N-S and E-W axes simultaneously to optimise ray capture.

• Current tracking systems are quite passive in that they can gather data but not provide a large level of remote control functionality and provide limited ability to process and predict weather condition information. Based on the above and inherent to the novelty of the project, Xirasol deduced there was no certainty their algorithm would successfully achieve the technical objectives and be able to optimise solar capture via tracking the sun curvature.

What was the experiment and how did it test the hypothesis?

Step 2: Project management and process design. This stage tests the practicality of theoretical concepts, and whether they would be beneficial in both technical and commercial circumstances. It is pivotal that the algorithm integrates with the dual-axis system and is able to optimise the capture of solar radiation. Inherent to the target audience, the algorithm must work in both an automatic and manual environment.

Step 3: Initial code/algorithm development. This stage attempts to replicate the theoretical concepts in an algorithmic solution to enable the development of a suitable product to achieve the technical objectives. Such involved the integration of the Programmable Logic Controller (PLC) with the several pieces of disparate hardware that are needed to drive / position the tracker to optimise solar energy capture and to collect different data sets used in system control and monitoring. A PLC is a type of industrial digital computer which can be used to control various processes and machines. This level of integration is extremely complex and has not been achieved before.

Step 4: Undertake controlled environment experimentation. This stage involves the testing of the algorithm in controlled environments where Xirasol can control that conditions the software is subject to and deduce whether the algorithm can successfully place the dual-axis structure such that the capture of solar radiation is optimises. This must be undertaken prior to large-scale uncontrolled experimentation to determine its efficacy in small-scale experimentation environments.

Step 5: Undertake uncontrolled environment experimentation. After successfully progressing through controlled environmental experimentation, the algorithm will be tested in uncontrolled environments. Specifically, Xirasol will be testing their algorithm in real-world environments, and testing the efficacy of the software to track the natural sun curvature. This will be compared to theoretical results to deduce its efficiency overtime.

Step 6: Further algorithmic development. Countless updates are required during the developmental phase to improve the effectiveness of the algorithm. Errors are a part of the natural lifecycle of the algorithm’s development, and Xirasol understand they will need to employ an iterative and complex algorithm development regime until the technical objectives are achieved.

Specifically, work undertaken in FY23 is positioned within steps 4-6, in which Xirasol is currently undertaken real- world experimentation and making amendments to their algorithmic structure. Xirasol undertook experimentation on the algorithm simultaneously to the dual-axis prototype, and have made changes based on initial results.

To enable diversity in the real-world experimentation, Xirasol have tested their algorithm in various environmental conditions, with differences in the following parameters:

Geographic location;
Grid operator requirements;
Grid operating conditions; &
Weather conditions including wind speed, ambient temperature and temperature range.

All of these parameters will have an impact on the structure, rigidity and malleability of the dual-axis prototype, and consequently, the algorithm must be able to override their impact to position the system at the most optimal position.

How did you evaluate or plan to evaluate results from your experiment?

The software has been implemented along with the physical prototypes and is currently working in real world conditions.

Specifically, the importance of the software was realised immediately, with such providing additional asset protection and management capabilities to maximise efficacy. Consequently, the algorithm has evolved into a practical functionality that is yielding good solar panel efficiency and protection. In the long-term, it is hoped machine learning and artificial intelligence methodologies will adjust the algorithm automatically to suit the panel’s specific environment. The software and platform have had wholesale changes due to the following technical complications:

Local controls of the system were impractical and inefficient with users preferring the automatic controls.

Remote control inputs from radio-based communications were only instantaneous with limited range. This was deemed to be a hardware problem with the software, in that it was unable to detect inputs over a certain distance.

The algorithm would not correctly adjust for severe weather conditions instantaneously with delays of up to 5 minutes. In severe circumstances, this could cause significant damage. Xirasol recognise the algorithm is in its infancy and is unlikely to work without undergoing an iterative development process. In doing so, Xirasol have developed a portal to gather public data to trigger system protections with adequate warning.
• Incorrect angling in overcast and suboptimal conditions that impeded optimised performance if positioning was derived from sun sensors. Sun sensors do not provide any form of backtracking functionality, and subsequently, were not providing the depth of information desired. The algorithm was subsequently refined to incorporate an additional stage to override this functionality.
• Several compatibility issues have been discovered with multiple devices. This triggers the use of additional vendor specific interface devices that add cost and lower functionality and reliability. Xirasol are progressing towards structuring the interface on a web-based application to avoid this issue.

Xirasol note that as the hardware component changes to reflect experimental results, the software must subsequently be adjusted as well (the software is integrated into the hardware of the particular prototype, and must correctly account for the malleability, rigidity and strength of the hardware).

FY23 has continued to focus on the consolidation of software activity (ie. data collection test data to verify previous program changes). We have continued to collect data from 100kW installations and have some operational data from the larger installations.

The bulk of the data collected is showing that previous program changes were correct and little further modification has been required. However, we noticed some significant issues in some settings where both hardware and software could not cope with power surges.

Software changes were required to better protect hardware and the physical tracker, changing the way the control system reacted to such power surges. Since implementing the changes, the prototypes seem to have stabilised and we have not had further issues from power surges.

Other key work completed in FY23 includes:

• Identified an issue with the type of weather / wind speed monitor being using. The issue has been diagnosed and an amendment to future designs has been implemented. Retro- fitting of installed units will be implemented. • Grouping of systems for wind front with key systems in areas showing wind fronts coming in and enabling following systems to go flat and reduce wind issues.
• Updated Wind Sensors, and discovered UPS are required due to issues with power and reliability. Retro- fitting of installed units will be implemented.
• Updated Modems and 4G connection for reliability
• Changed moment envelope and torque settings to handle different winds in different area.

If you reached conclusions from your experiments in the selected income period, describe those conclusions:

Inherent to the infancy of experimentation, Xirasol are currently unable to determine the satisfaction of their hypotheses and advise that further experimentation will need to be undertaken. Specifically, although initial experimentation is indicative of optimistic results, Xirasol presently do not have enough data of a diverse range of conditions to certify that the software improves the structural rigidity and technical efficiency.

In particular, Xirasol advise the following:

The tracking system was able to track the natural curvature of the sun in neutral conditions (ie. with minimal clouds, wind and rain); complications did arise where the weather conditions changed as indicated above. The results do not indicate a 20% increase in the energy capture, albeit, there is a clear increase. Further experimentation needs to be undertaken to ensure the algorithm can more appropriately track the sun in diverse conditions.
The algorithm was not able to correctly override tracking system when severe environmental conditions causing some damage to the prototype units. This has been improved overtime, and some overrides have arose during experimentation, albeit, Xirasol is unable to determine, with certainty, that the algorithm can improve the lifespan of their units to the required amount. It is estimated lifespan is presently being increased by 5-10%.

Work will continue into FY24 to complete the required retro-fitting of existing prototypes to enable further in-field testing in different weather conditions. A range of repairs and maintenance activities will also need to be carried out due to address damage to panels and inverters.

Xirasol still has a particular interest in the ability of the software to predict severe weather conditions and adjust its positioning accordingly, given this was a technical uncertainty of the initial algorithm. Experimentation will continue on the abovementioned hypotheses and technical objectives under the same philosophy in an attempt to extend the current technical capacity of the algorithm, with further testing to be undertaken in more practical environments concurrently with the prototype installations. It is hoped such experimentation will increase the dataset for the algorithm, enabling movements to be supported by more distinct and complete data.