Dual-axis wind-responsive solar tracking system with load-based azimuth sensing

Abstract

A solar-tracking system uses a combination of active and passive movement to protect solar panels from wind while maximizing energy production. The system rotates the panels around a vertical axis using a drive mechanism controlled by a computing unit. The panels are mounted on horizontal hinges that allow them to tilt upward automatically when exposed to strong wind. The computing unit uses motor current, cable tension, or torque sensor data to estimate panel orientation and wind conditions, without needing separate wind sensors or angle sensors. The system also adjusts panel orientation to improve power output based on real-time electrical measurements. This approach provides reliable wind protection and efficient solar tracking with minimal hardware.

Description

[0001] DUAL-AXIS WIND-RESPONSIVE SOLAR TRACKING SYSTEM WITH LOAD-BASED AZIMUTH SENSING

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to the field of solar-energy installations, and more particularly to apparatus and control methods for orienting photovoltaic (PV) modules.

[0004] BACKGROUND OF THE INVENTION

[0005] Solar photovoltaic (PV) tracking systems are well known as a means to increase energy yield by orienting panels to follow the sun's apparent motion. Conventional single-axis trackers (see, e.g., U.S. Pat. No. 8,459,249) and dual-axis trackers (see, e.g., US Pat. No. 10,215,834) typically employ one or more rotary encoders, optical sensors, or inclinometer / tilt sensors at each row or module to measure instantaneous azimuth and elevation. While accurate, per-row encoders (e.g., U.S. Pat. No. 11,300,326) and tilt meters add significant hardware costs, require extensive wiring harnesses, and introduce additional points of failure (wiring breaks, sensor drift, calibration issues).

[0006] Wire-driven daisy-chain trackers attempt to reduce cost by using a single motor to tension a continuous cable, thereby rotating multiple rows in unison. Such systems afford lower motor count and fewer drive assemblies per hectare. However mechanical backlash, cable stretch, wheel slip, and environmental loading (wind, hail, snow, soiling) can cause each row's true azimuth to diverge from the nominal commanded position. Over time, these cumulative errors lead to misalignment with the sun, reduced irradiance capture, and— critically— failure to enter safe stow positions under high winds.

[0007] In an effort to correct drift, many prior-art approaches incorporate per-row absolute encoders (see, e.g., U.S. Pat. No. 11,280,522) or tilt sensors (see, e.g., U.S. Pat. No. 8,481,906). These methods accurately detect panel attitude but suffer from the same drawbacks of increased capital expense, wiring complexity, and additional maintenance overhead. Moreover, should any encoder or tilt sensor fail, the tracker's control logic is vulnerable to erroneous or missing data. Similarly, some systems rely on dedicated wind sensors (e.g., anemometers) collocated with the array (see, e.g., U.S. Pat. No. 12,174,644), but these typically measure bulk wind conditions rather than the localized wind-panel interaction; failure of any single wind sensor may compromise the entire wind-stow protocol.

[0008] Separately, closed-loop maximum power point tracking (MPPT) methods have been applied at the panel or string level (see for example US2022 / 0399850). Such MPPT schemes adjust module operating voltage and current to maintain peak power output but do not actively correct for misalignment of the panel plane itself. That is, while inverter-integrated MPPT ensures each string operates at its electrical maximum, it does not physically reorient the module to compensate for drift or local shading— and so cannot remediate yield loss due to tilt or azimuth errors. Furthermore, some tracker designs such as that of European Patent No. EP3 108 186 incorporate mechanical features like torsion limiters to allow photovoltaic (PV) rows to passively feather into the wind, reducing structural loads and mitigating the risk of damage during high wind events. These systems provide a level of passive protection by enabling controlled movement of the panels under excessive wind forces. However, to confirm when panels have fully feathered or returned to their operational position, such designs still often rely on positional references— typically encoders or tilt sensors. In the absence of reliable sensor data, whether due to sensor failure, icing, or electronic faults, the tracker may remain misaligned, increasing the risk of structural damage and energy loss.

[0009] Accordingly, there remains an unmet need for a low-cost, highly reliable tracker control architecture that eliminates dedicated per-row angle sensors and their associated wiring, calibration, and maintenance burdens, provides real-time, sensorless estimation of azimuth by leveraging existing drive-motor current or tension signals, integrates wind-adaptive stow logic without requiring separate wind sensors or per-row encoders, continuously optimizes actual energy yield by applying closed-loop tracking adjustments based on real power output rather than sun-position models alone, and offers redundancy, so that if any external sensor (wind, encoder, tilt) fails, the tracker can still accurately estimate orientation and stow safely under high winds.

[0010] The present invention addresses these limitations through one or more of the following features: a method of correlating drive system behaviour with known mechanical reference points; a mounting system that passively adjusts to wind conditions; and a feedback mechanism that monitors power output to guide real-time positioning. Any of these features, alone or in combination, enable multirow solar arrays with shared drive linkages to achieve accurate sun tracking, adopt protective wind stow positions, and optimise energy capture— without requiring dedicated per-row sensors to measure tilt or angular position.

[0011] SUMMARY OF THE INVENTION

[0012] According to a first aspect of the present invention, there is provided a solar tracking assembly comprising a rotatable support member configured to carry at least one solar panel and rotate about a vertical axis, a horizontal member connecting the at least one solar panel to the rotatable support member and permitting the at least one solar panel to passively pivot about a horizontal axis in response to wind, a drive mechanism operably coupled to the rotatable support member to effect controlled rotation of the solar tracking assembly about the vertical axis, the drive mechanism controlled by a computing unit, the computing unit configured to execute wind responsive orientation logic, wherein the computing unit determines current or predicted wind speed and wind direction, and causes the drive mechanism to rotate the rotatable support structure such that the rear side of the solar panel faces into the direction of incoming wind, thereby enabling the at least one solar panel to tilt upward about the horizontal member in response to wind.

[0013] According to a further aspect, the solar tracking assembly is configured such that the computing unit receives input from a wind sensor and determines wind direction and wind speed prior to executing the wind responsive orientation logic.

[0014] According to a further aspect, the solar tracking assembly further comprises a load-feedback sensor configured to measure a mechanical load associated with the drive mechanism.

[0015] According to a further aspect, the solar tracking assembly is configured such that the computing unit receives an output from the load-feedback sensor and estimates a current azimuthal position of the at least one solar panel.

[0016] According to a further aspect, the solar tracking assembly further comprises a current-sensing device for determining electrical current flowing through the drive mechanism.

[0017] According to a further aspect, the solar tracking assembly is configured such that the computing unit receives electrical current information from the current-sensing device and utilizes the current information to estimate the current azimuthal position of the at least one solar panel.

[0018] According to a further aspect, the solar tracking assembly is configured such that the computing unit further comprises calibration data mapping mechanical load to rotational orientation of the at least one solar panel.

[0019] According to a further aspect, the solar tracking assembly further comprises a power monitoring device configured to measure instantaneous electrical power output of the at least one solar panel.

[0020] According to a further aspect, the solar tracking assembly is configured such that the computing unit controls the drive mechanism to perturb the rotational orientation of the rotatable support member and compares output from the power monitoring device to determine whether to further perturb the rotational orientation of the rotatable support member.

[0021] According to a further aspect, the solar tracking assembly is configured such that the computing unit controls the drive mechanism to rotate the rotatable support member to a rotatable orientation, wherein the computing unit determines the rotatable orientation using wind responsive information including the geometry of the at least one solar panel, the horizontal member, and a predetermined wind threshold.

[0022] According to a further aspect, the solar tracking assembly is configured such that, under wind loading with the rear surface of the solar panel oriented into the wind, the at least one solar panel generates an aerodynamic moment about the horizontal hinge axis sufficient to overcome the hinge bias and cause the panel to pivot upward.

[0023] According to a further aspect, the solar tracking assembly is configured such that the computing unit determines wind direction by identifying the rotational orientation of the rotatable support member at which drive load is minimized or maximized during controlled panel rotation.

[0024] According to a further aspect, the solar tracking assembly is configured such that the computing unit determines wind speed based on the rate of change in drive load of the drive mechanism with respect to azimuthal displacement of the at least one solar panel from the wind direction.

[0025] According to a further aspect, the solar tracking assembly is configured such that the computing unit disables solar tracking and maintains a wind-stable orientation when the wind speed exceeds a survivability threshold.

[0026] According to a further aspect, the solar tracking assembly further comprises a power monitoring device including a maximum power point tracking subsystem that communicates with the computing unit to assess solar panel output.

[0027] According to a further aspect, the solar tracking assembly is configured such that the load-feedback sensor comprises a motor current sensor integrated into the drive mechanism.

[0028] According to a further aspect, the solar tracking assembly further comprises a torque sensor integrated into the drive mechanism.

[0029] According to a further aspect, the solar tracking assembly is configured such that the computing unit continuously updates the calibration data based on historical comparisons between measured power output and mechanical load estimates.

[0030] According to a further aspect, the solar tracking assembly is configured such that the drive mechanism comprises a motor coupled to the rotatable support member via a tensioned cable loop and pulley arrangement.

[0031] According to a further aspect, the solar tracking assembly is configured such that the computing unit prioritizes placing the at least one solar panel in a wind responsive orientation over a power optimization orientation when wind speed exceeds a predefined switching threshold. According to a further aspect, the solar tracking assembly further comprises a damping mechanism associated with the horizontal hinge member to reduce oscillations during high wind conditions.

[0032] According to a further aspect, the solar tracking assembly is configured such that the tensioned cable loop comprises opposing first and second cable branches, and further comprises a load-feedback sensor with respective tension sensors configured to measure cable tension in each branch, the computing unit being further configured to detect misalignment or crosswind loading based on a difference between the measured tensions.

[0033] According to a further aspect, the solar tracking assembly is configured such that the computing unit further includes calibration data defining a mapping between a mechanical load parameter and a corresponding rotational orientation of the at least one solar panel.

[0034] According to a further aspect, the solar tracking assembly is configured such that the computing unit suspends control of the drive mechanism in response to detection of at least one of a wind-load index exceeding a safety threshold, a wildlife-safe orientation, or a hail or snow-loading signal, and resumes control of the drive mechanism when the condition has cleared.

[0035] According to a further aspect, the solar tracking assembly is configured such that the drive mechanism further comprises a motor-current sensor operatively coupled to the drive mechanism, and a mechanical load-feedback sensor configured to measure cable tension in the tensioned cable loop, wherein the computing unit is configured to compare contemporaneous readings from the load-feedback sensor and the motor-current sensor, detect that one sensor is unreliable when its reading deviates from the other sensor by more than a predefined consistency threshold for a predefined time window, and thereafter substitute the remaining reliable sensor for estimating panel azimuth and for executing wind responsive orientation logic until the unreliable sensor is restored within the threshold.

[0036] According to a further aspect, the solar tracking assembly is configured such that the computing unit determines a rolling average energy-yield metric over a first predefined time window, compares the rolling average energy-yield metric with a historical baseline corrected for irradiance, and performs a calibration sweep when the deviation between the rolling average energy-yield metric and the historical baseline exceeds a predetermined drift threshold.

[0037] According to a further aspect, the solar tracking assembly further comprises a break-away clutch incorporated into the tensioned cable loop, the break-away clutch being configured to disengage the drive motor from the rotatable support member when cable tension exceeds a preset overload threshold, thereby protecting the drive train and the solar panel structure from extreme wind or mechanical jam conditions. According to a further aspect, the solar tracking assembly further comprises a wind sensor, a motorcurrent sensor operatively coupled to the drive mechanism, and a mechanical load-feedback sensor configured to measure cable tension in the tensioned cable loop, wherein the computing unit is configured to compare a wind-direction value reported by the wind sensor with a mechanically implied wind direction derived from at least one of motor current output by the motor-current sensor and cable tension data output by the load-feedback sensor, identify a persistent bias when the reported wind direction differs from the mechanically implied wind direction by more than a configurable angular tolerance for a predefined duration, and upon identifying the persistent bias, automatically apply a correction factor to the wind-direction value and or transmit an alert indicating that the wind sensor requires inspection or recalibration.

[0038] According to a further aspect, there is provided a solar-tracker plant comprising a plurality of solartracker assemblies, each assembly having a centre of rotation defined by a rotatable support member, and a ground layout in which the centres of rotation are placed at the corner points of a repeating grid made up entirely of equal-sized regular hexagons, the grid covering the site without gaps or overlaps, whereby every solar-tracker assembly is the same distance from each of its six nearest neighbours, the circular three hundred sixty-degree rotation envelopes of adjacent solarpanel structures touch but do not overlap, and the layout naturally supplies a set of row-direction families that are spaced sixty degrees apart in azimuth.

[0039] According to a further aspect, the solar-tracker plant is configured such that a computing unit assigns each solar-tracker assembly to one of the row-direction families and switches all assemblies simultaneously from one family to another in response to at least one of detection of inter-row shading, a predetermined low-sun-altitude clock event, or activation of a wildlife-safe, hail, snowmitigation, or wind-stow mode.

[0040] According to a further aspect, the solar-tracker plant is configured such that at least one tensioned drive cable that transmits rotational motion between solar-tracker assemblies is routed along a straight edge of the hexagonal grid so that successive cable spans are substantially equal in length.

[0041] According to a further aspect, the solar-tracker plant is configured such that a site-design algorithm stored in the computing unit receives a land-boundary outline and automatically fills the interior of the outline with the regular-hexagon grid, trimming partial hexagons at the boundary to maximise ground-coverage ratio while preserving non-overlapping rotation envelopes.

[0042] According to a further aspect, the solar-tracker plant is configured such that during construction, the position of each new solar-tracker assembly is established by measuring a single repeated centre-to- centre distance from two already-installed assemblies, thereby preventing cumulative angular error across the array.

[0043] According to a further aspect, the solar-tracker plant is configured such that the computing unit automatically identifies perimeter solar-tracker assemblies from the neighbour geometry of the hexagonal grid and applies lower wind-stow thresholds to those perimeter assemblies in accordance with a predefined environmental-priority routine.

[0044] According to a further aspect, there is provided a solar-tracker plant comprising a plurality of solartracker assemblies, each assembly having a centre of rotation defined by a rotatable support member, and a ground layout in which the centres of rotation are placed at the corner points of a repeating grid made up entirely of equal-sized equilateral triangles, the grid covering the site without gaps or overlaps, whereby every solar-tracker assembly is the same distance from each of its three nearest neighbours, the circular three hundred sixty-degree rotation envelopes of adjacent solarpanel structures touch but do not overlap, and the layout naturally supplies a set of row-direction families that are spaced sixty degrees apart in azimuth.

[0045] According to a further aspect, the solar-tracker plant is configured such that a computing unit assigns each solar-tracker assembly to one of the row-direction families and switches all assemblies simultaneously from one family to another in response to at least one of detection of inter-row shading, a predetermined low-sun-altitude clock event, or activation of a wildlife-safe, hail, snowmitigation, or wind-stow mode.

[0046] According to a further aspect, the solar-tracker plant is configured such that at least one tensioned drive cable that transmits rotational motion between solar-tracker assemblies is routed along a straight edge of the triangular grid so that successive cable spans are substantially equal in length.

[0047] According to a further aspect, the solar-tracker plant is configured such that a site-design algorithm stored in the computing unit receives a land-boundary outline and automatically fills the interior of the outline with the equilateral-triangle grid, trimming partial triangles at the boundary to maximise ground-coverage ratio while preserving non-overlapping rotation envelopes.

[0048] According to a further aspect, the solar-tracker plant is configured such that during construction, the position of each new solar-tracker assembly is established by measuring a single repeated centre-to- centre distance from two already-installed assemblies, thereby preventing cumulative angular error across the array.

[0049] According to a further aspect, the solar-tracker plant is configured such that the computing unit automatically identifies perimeter solar-tracker assemblies from the neighbour geometry of the triangular grid and applies lower wind-stow thresholds to those perimeter assemblies in accordance with a predefined environmental-priority routine.

[0050] According to a further aspect, there is provided a computer-implemented method of operating a solar tracking assembly, the method comprising executing, by a computing unit, control logic to cause a rotatable support member to carry at least one solar panel and rotate about a vertical axis, wherein a horizontal member connects the at least one solar panel to the rotatable support member and permits the at least one solar panel to passively pivot about a horizontal axis in response to wind, the method further comprising determining, with the computing unit, current or predicted wind speed and wind direction, and rotating, with the computing unit, the rotatable support member such that a rear side of the solar panel faces into the direction of incoming wind, thereby enabling the at least one solar panel to tilt upward about the horizontal member in response to wind.

[0051] According to a further aspect, the computer-implemented method further comprises receiving, with the computing unit, wind direction and wind speed data from a wind sensor, and determining, with the computing unit, the wind direction and wind speed prior to rotating the rotatable support member.

[0052] According to a further aspect, the computer-implemented method further comprises measuring, with a load-feedback sensor, a mechanical load associated with a drive mechanism operably coupled to the rotatable support member, and providing the measured mechanical load to the computing unit.

[0053] According to a further aspect, the computer-implemented method further comprises estimating, with the computing unit, a current azimuthal position of the at least one solar panel based on the measured mechanical load.

[0054] According to a further aspect, the computer-implemented method further comprises measuring, with a current-sensing device, electrical current flowing through the drive mechanism, and providing the measured electrical current to the computing unit.

[0055] According to a further aspect, the computer-implemented method further comprises estimating, with the computing unit, the current azimuthal position of the at least one solar panel based on the measured electrical current.

[0056] According to a further aspect, the computer-implemented method further comprises accessing, with the computing unit, calibration data defining a mapping between mechanical load and rotational orientation of the at least one solar panel, and applying the calibration data to the estimation of azimuthal position.

[0057] According to a further aspect, the computer-implemented method further comprises measuring, with a power monitoring device, instantaneous electrical power output of the at least one solar panel, and providing the power measurement to the computing unit.

[0058] According to a further aspect, the computer-implemented method further comprises perturbing, with the computing unit, the rotational orientation of the rotatable support member, comparing, with the computing unit, successive power measurements, and determining, with the computing unit, whether to further perturb the rotational orientation based on the comparison.

[0059] According to a further aspect, the computer-implemented method further comprises determining, with the computing unit, a rotatable orientation of the rotatable support member using wind responsive information, the information including at least geometry of the solar panel, geometry of the horizontal member, and a predetermined wind threshold, and rotating, with the computing unit, the rotatable support member to the determined rotatable orientation.

[0060] According to a further aspect, the computer-implemented method further comprises generating, under wind loading, an aerodynamic moment about the horizontal axis of the solar panel sufficient to overcome a hinge bias and cause the panel to pivot upward.

[0061] According to a further aspect, the computer-implemented method further comprises identifying, with the computing unit, a rotational orientation of the rotatable support member at which drive load is minimized or maximized during controlled panel rotation, and determining, with the computing unit, wind direction based on the identified orientation.

[0062] According to a further aspect, the computer-implemented method further comprises determining, with the computing unit, wind speed based on a rate of change in drive load of the drive mechanism with respect to azimuthal displacement of the at least one solar panel from the wind direction.

[0063] According to a further aspect, the computer-implemented method further comprises disabling, with the computing unit, solar tracking, and maintaining, with the computing unit, a wind-stable orientation when wind speed exceeds a survivability threshold.

[0064] According to a further aspect, the computer-implemented method further comprises communicating, with the computing unit, with a maximum power point tracking subsystem to assess solar panel output. According to a further aspect, the computer-implemented method further comprises measuring, with a motor current sensor integrated into the drive mechanism, mechanical load associated with the drive mechanism, and providing the measured mechanical load to the computing unit.

[0065] According to a further aspect, the computer-implemented method further comprises measuring, with a torque sensor integrated into the drive mechanism, drive shaft torque, and providing the measured torque to the computing unit.

[0066] According to a further aspect, the computer-implemented method further comprises continuously updating, with the computing unit, the calibration data based on historical comparisons between measured power output and mechanical load estimates.

[0067] BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a perspective view of a solar-tracking assembly according to one embodiment of the present invention.

[0069] Figure 2 is a schematic perspective view of two linked solar-tracking assemblies driven by a tensioned cable loop according to one embodiment of the present invention.

[0070] Figure 3 is a perspective view of a solar-tracking assembly with a compact lower stop and rear brace according to one embodiment of the present invention.

[0071] Figure 4 is a perspective view of the solar-tracking assembly of Figure 1 with its panels lifted to a wind-relief position.

[0072] Figure 5 is a perspective view of a solar-tracking assembly that employs a stop wire to limit downward panel rotation.

[0073] Figure 6 is a flow-chart illustrating a sensorless heading-and-wind estimation routine executed by the controller.

[0074] Figure 7 is a graph illustrating wind-induced yaw torque as a function of yaw angle.

[0075] Figure 8 is a flow-chart illustrating a perturb-and-observe power-optimisation routine.

[0076] Figure 9 is a flow-chart illustrating a learn-and-recall routine for optimal azimuth set-points.

[0077] Figure 10 is a flow-chart illustrating environmental-priority logic for wind, hail, snow and wildlife modes.

[0078] Figure 11 is a plan view of a photovoltaic plant layout in a repeating hexagonal grid. Figure 12 is an enlarged plan view of three neighbouring tracker posts forming an equilateral triangle and their rotation envelopes.

[0079] Figure 13 is a plan view of multiple trackers on the hexagonal grid showing module rectangles, straight cable runs and one selected row-direction family.

[0080] Figure 14 is a block diagram illustrating an example of a control-computing subsystem of the computing unit and / or system, or modules of the system, according to at least some embodiments herein.

[0081] Figure 15 shows a solar tracking assembly with a rotatable vertical support member, upper and lower solar panel support members, and a motor for driving rotation.

[0082] Figure 16 shows a solar tracking assembly with a motor-driven rotatable vertical support member and a solar panel rotated by wind force.

[0083] Figure 17 shows a solar tracking assembly with a motor-driven rotatable vertical support member and solar panels.

[0084] Figure 18 shows a solar tracking assembly with a rotatable vertical support member, upper and lower solar panel support members, and a motor for driving rotation.

[0085] Figure 19 shows a solar tracking assembly with a rotatable vertical support member, lower and upper solar panel support members, and at least one solar panel.

[0086] Figure 20 shows a solar tracking assembly with multiple support members and a rotatable vertical support member.

[0087] Figure 21 shows a solar tracking assembly with a rotatable vertical support member, upper and lower solar panel support members, and a foundation.

[0088] Figure 22 demonstrates a solar tracking assembly with a vertical support member, a solar panel support member, and counterweights attached to the solar panels.

[0089] Figure 23 demonstrates a solar tracking assembly with a vertical support member, a solar panel support member, and a counterweight mechanism.

[0090] Figure 24 demonstrates a solar tracking assembly with a vertical support member, a solar panel support member, and a counterweight mechanism.

[0091] Figure 25 demonstrates a solar tracking assembly with a vertical support member, solar panel support member, rotatable solar panel, and counterweights. Figure 26 shows a solar tracking assembly with a rotatable vertical support member, a solar panel support member, and a solar panel attached via connection members.

[0092] Figure 27 shows a solar tracking assembly with a rotatable vertical support member, a solar panel support member, and a solar panel attached via a connection member.

[0093] Figure 28 shows a system diagram of a wind-responsive solar panel system with wire-based rotational control.

[0094] Figure 29 shows a locking member securing a wire onto a wheel in a wire-based rotational control system for solar panels.

[0095] Figure 30 shows a system diagram of multiple solar panel arrays driven by a single motor through a wire-based transmission system.

[0096] Figure 31 shows a release mechanism securing a wire within a wheel in a wire-based rotational control system for solar panels.

[0097] Figure 32 shows a fastening mechanism for independent rotation of a rotatable vertical member in a solar tracking system.

[0098] Figure 33 shows a system diagram of dual motors with wire loops for counteracting torque in a solar panel tracking system.

[0099] Figure 34 shows a system diagram of a dual-motor configuration with wire loops for solar panel rotation control.

[0100] Figure 35 shows a system diagram illustrating the rotation of solar panels around two axes in a wind- responsive solar panel system.

[0101] DETAILED DESCRIPTION

[0102] Throughout the present disclosure, the following terms and definition may be used.

[0103] As used herein, a solar cell, aka a photovoltaic cell, is capable of generating electricity when illuminated by light. Solar cells are commonly available on the market, such as crystalline solar cells, monocrystalline solar cells, polycrystalline solar cells, th in-fil m solar cells, or the like.

[0104] As used herein, the term "solar panel" refers to a panel comprising one or more photovoltaic cells, aka solar cells.

[0105] As used herein, the term "azimuth" refers to the horizontal angular position of a solar panel or tracker assembly relative to a fixed reference direction (typically true north). In practice, the azimuth angle is measured in the plane parallel to the ground and is defined as the clockwise angle between the reference direction and the projection of the panel's normal vector onto that horizontal plane. For example, an azimuth of 0° indicates that the panel faces true north, 90° indicates east, 180° indicates south, and 270° indicates west. In a single-axis tracker, rotation about the vertical support member changes the panel's azimuth, allowing the module to follow the sun's apparent motion from dawn to dusk. Accurate knowledge of a panel's azimuth is critical both for maximizing harvested irradiance and for orienting the panel edge-on to prevailing winds during stow or wind-protection modes. Consequently, mapping mechanical load readings (motor current or cable tension) to the corresponding azimuth angle enables sensorless estimation of true heading without encoders or optical sensors.

[0106] As used throughout this disclosure, the term "sensorless" specifically refers to the absence of dedicated angle sensors (such as rotary encoders, tilt meters, or optical position sensors) and dedicated wind sensors (such as anemometers or vane-type direction sensors) at each tracker row. The invention may still utilize internal system signals— such as motor current, cable tension, or inverter power output— which serve as indirect indicators of orientation or environmental load. These signals are not considered "external sensors" in the conventional sense, and their use is consistent with the "sensorless" approach described herein.

[0107] Although bifacial modules are capable of converting light on both of their broad faces and may appear optically symmetrical, they are still assigned a functional "front" and "rear" in the context of the present invention. The distinction is installation-dependent: the face that, in the tracker's normal power-tracking orientation, presents the larger solid angle toward the sky and thus receives the predominance of direct and diffuse solar irradiance is defined as the front, while the opposite face is defined as the rear. Accordingly, even a glass-glass bifacial module mounted in a portrait frame will have a front (sky-facing) surface and a rear (ground-facing) surface as those terms are used herein. All references in this description to wind acting on the "rear surface," or to orienting the module so its "rear surface" confronts the wind vector, should therefore be interpreted with respect to this installation-based front / rear assignment rather than intrinsic cell architecture.

[0108] A central tenet of the herein described invention is the concept of passive lift, tilt, or rotation. This refers to a self-acting (in other words, not driven via power such as a motor) mechanical behavior built into each solar panel's mounting hardware, whereby wind forces alone cause the panel to "feather" or pivot to a safer angle. In practical terms, each panel is supported on a horizontal hinge that provides a controlled resisting torque— often via a calibrated spring, torsion bar, or internal friction element— so that under light or moderate wind the panel remains fixed at its normal suntracking tilt. When wind speed increases and the aerodynamic influence acting on the large, flat panel (preferably rear) surface exceeds this resisting torque, the hinge yields and the panel automatically pivots upward. As the panel lifts, its angle of attack relative to the oncoming wind decreases, which rapidly reduces the wind-induced torque. The pivot continues until a new equilibrium is reached— either at a partially open angle where hinge resistance balances diminished wind moment or, in higher gusts, until the panel contacts a mechanical stop.

[0109] The present invention addresses the technical problem of improving the wind protection, structural safety, and energy optimisation of solar tracking assemblies, while minimising hardware complexity, sensor count, and maintenance requirements. Conventional solar trackers require multiple powered axes, dedicated tilt actuators, encoders, and external wind sensors to maintain safe operation under varying wind conditions, increasing system cost and reducing reliability. The invention overcomes these limitations by combining passive tilt mechanisms, load-based sensorless control logic, and a simplified mechanical drive architecture, providing a robust, low-complexity solution for wind- responsive solar tracking.

[0110] In configurations where a calibrated return spring or torsion element is employed, gravity and spring force work together to bring the panel back down once wind forces drop below the hinge's preload threshold. In other embodiments, no return spring is required: the panel simply rests against a support bar, cable, or rigid stop at its normal tilt. In that case, gravity alone causes the panel to swing back down; the support member limits further downward rotation, ensuring the panel returns precisely to its intended tracking angle. Whether using a spring-loaded hinge or a simple gravitybased rest stop, the critical feature is that no active power or external command is needed to move the panel during gusts or to return it afterward.

[0111] Because passive lift engages instantaneously— without waiting for an electronic controller to detect wind conditions or for a motor to spool up— the system reduces peak structural loads and greatly speeds response to sudden gusts. The only active control required is rotation about the vertical axis: once each panel is oriented edge-on to the wind by the azimuth motor, any remaining aerodynamic moment is handled passively by the hinge. When wind levels subside, panels drop back to full powerproducing orientation on their own. By relying on this purely mechanical, self-regulating tilt relief, the invention minimizes hardware complexity, eliminates extra actuators and wiring for the tilt axis, and provides a built-in, failsafe wind-protection function that both preserves energy yield and enhances structural durability.

[0112] The term "wind responsive," as used in the present invention, refers to a solar tracking architecture in which wind-related protection and survivability are achieved not through powered actuators or real-time sensing at the tilt axis, but instead through an intelligent combination of active vertical-axis rotation and passive horizontal-axis response. In this configuration, the system does not attempt to rigidly resist wind loads across two axes with sensors and motors, but rather accommodates aerodynamic forces through controlled structural behaviour. The tracker may be controlled to actively rotate around its vertical post to orient the panels, while the panels themselves are mounted on horizontal hinges or mounting members that allow them to feather upward under wind induced lift forces. This passive mechanical tilt— requiring no powered drive or control signal— automatically reduces aerodynamic torque as wind intensity increases, and then restores the panels to their original position once conditions subside. Such a system must be capable of detecting wind direction and reacting promptly, while tolerating the fact that horizontal tilt is neither directly sensed nor motorized. Wind responsive systems therefore introduce unique technical challenges: they must infer panel orientation and wind loading indirectly through drivetrain feedback, manage nonuniform tilt responses across multiple rows, and operate with minimal hardware while maintaining structural safety and energy yield. The present invention addresses these challenges through a suite of control, sensing, and fault-tolerant innovations tailored specifically to the demands of wind responsive architectures employing vertical active rotation, horizontal passive lift, and distributed cable-driven actuation.

[0113] The present invention utilises a dual-rotation solar tracker in which a photovoltaic panel assembly is supported for movement about two mutually orthogonal axes. A first, substantially vertical axis is defined by a post that can be turned through at least three-hundred degrees by any convenient actuator— such as a wire-and-wheel transmission, a direct-drive gear-motor, a hydraulic rotary actuator, or an equivalent mechanism capable of controlled azimuth rotation. A second, substantially horizontal axis is defined by a support member that carries the panel and is biased toward a nominal operating tilt by a resilient or damped element; the bias is calibrated so that the panel can lift or feather when the aerodynamic overturning moment imposed by wind exceeds a predetermined threshold.

[0114] By rotating the vertical post, the controller turns the panel so its narrow edge faces the wind, whichever direction the wind comes from, thereby reducing exposed area and torsional load. As wind pressure increases, the panel support lifts, raising the leading edge, lowering the angle of attack and damping aerodynamic effects such as flutter. When the gust passes, a spring or damper returns the panel to its normal tilt without any powered drive.

[0115] This dual-axis arrangement enables the control algorithms disclosed herein to: align perimeter trackers edge-on to wind while allowing interior trackers to continue power production under higher thresholds; • introduce small azimuth perturbations for real-time power optimisation without compromising structural safety;

[0116] • shed wind load automatically, avoiding premature full-array stow commands; and

[0117] • reach elevated or steep orientations, when required, for hail, snow removal or wildlife clearance, using only the single active drive on the vertical post.

[0118] In consequence, the invention achieves wind protection across the full 360-degree horizon while reducing peak structural loads and drive-train complexity, all with minimal additional sensor or actuator cost.

[0119] The primary subject matter of the present disclosure lies in the control logic used to operate solartracker arrays. The mechanical arrangements illustrated in Figures 1-5 are therefore presented only as non-limiting examples that demonstrate how the algorithms can be integrated with representative tracker hardware. A person of ordinary skill will appreciate that the same control concepts can be implemented on a wide variety of tracker architectures, including but not limited to:

[0120] • single-row or multi-row vertical-axis trackers of the type shown;

[0121] • horizontal single-axis trackers (HSAT) with torque-tube drive lines;

[0122] • dual-axis gimbal or pedestal trackers;

[0123] • slew-drive, linear-actuator, hydraulic, geared, belt-and-pulley, or cable-and-capstan actuation systems; and

[0124] • distributed or centralised drive configurations, whether direct-drive, indirect, manual, or hybrid.

[0125] Accordingly, unless expressly stated otherwise, references in the description to specific structural elements— such as posts 102, upper support members 105, stop members 104 / 104B, wheels 200, or panels 500— are intended solely to provide illustrative context. The inventive control algorithms may be executed on any hardware capable of imparting controlled rotational or translational motion to a solar-collector surface and of sensing, estimating, or receiving the operational parameters referenced herein. Variations in geometry, material selection, actuation technology, sensing modality, power supply, communication protocol, and installation environment do not depart from the scope of the invention as defined by the appended claims.

[0126] In one aspect of the present invention, as shown in Figure 1, there is a solar panel mounting system 100. A rotatable vertical support member 102 extends upward from a foundation or ground anchor (not shown) and is arranged to rotate about its substantially vertical longitudinal axis. Rotation of support member 102 is effected by a suitable drive means— such as a direct-drive gear-motor, a hydraulic rotary actuator, or a remote motor acting through a tensioned wire-and-wheel transmission— none of which is shown in this view for clarity.

[0127] Rigidly attached to the upper end of vertical support member 102 is an upper solar-panel support member 105, depicted in this example as a tube that defines a substantially horizontal hinge axis. A series of connection members 400, illustrated in this example as sleeve-type hinge brackets, are spaced along support member 105 and pivotally couple respective solar panels 502 to the upper solar panel support member 105. Collectively solar panels 502 form a solar panel structure 500. Each connection member 400 allows its associated panel 502 to swing about the horizontal axis so that the panel can feather under wind load. In the embodiment shown three connection members 400 secure two side-by-side panels 502, but any spacing, module count or bracket style may be used.

[0128] Optionally, located beneath the upper solar panel support member 105 is a lower solar panel support member 104 which spans between the solar panels 502 and provides a mechanical stop that arrests downward rotation of the panels when wind subsides. Member 104 may carry bumpers, dampers or spring elements (not illustrated) to cushion contact.

[0129] The rear surface 500B of each panel faces away from incident sunlight during normal powerproducing operation. When a gust is detected or forecast, the drive means rotates support member 102 so that rear surfaces 500B present a narrow edge to the oncoming wind. If wind pressure rises further, the aerodynamic moment acting below the hinge axis causes panels 502 to pivot upward about member 105, thereby reducing frontal area and relaxing tension in any drive cables. When the gust passes, gravity or a torsion-bias element within each connection member 400 returns the panels to their nominal tilt against the lower support member 104.

[0130] Although Figure 1 depicts a single post carrying two rectangular modules, the invention encompasses any number of panels, hinge brackets and support-member geometries, as well as alternative drive arrangements capable of rotating the vertical support member about its axis.

[0131] Figure 2 illustrates, in schematic perspective, a portion of a solar tracking system according to one aspect of the present invention, in which two vertical axis tracker assemblies are driven in unison by a single tensioned wire loop.

[0132] A first tracker includes a rotatable vertical support member 102 surmounted by an upper solar panel support member 105. Support member 105 defines the horizontal hinge axis on which the panel structure 500 is mounted. A wheel 200 is fixed to the lower end of support member 102, which in turn accepts a wire 202 around a portion of the circumference of the wheel 200 to facilitate rotation of the vertical support member 102. Wind flowing in direction J contacts the rear surface 500B of the panel structure 500, causing the panel structure to passively rotate upwards about the connection members 400. Such rotation may take the form of any combination of pivoting, hinging or feathering motion permitted by the connection members 400, and can be accommodated by, for example, one or more of: a simple pin or sleeve hinge; a pre-loaded torsion spring; an elastomeric or viscous damper; a cam-follower or sliding bearing; a flexible composite joint; or any other mechanical arrangement that yields under the aerodynamic moment to allow controlled lift of the panel, and subsequently returns the panel to its rest position either by gravity, spring bias, or damping action once wind pressure subsides.

[0133] A second, identical tracker is arranged down-loop to the right. Both trackers are coupled by a continuous drive wire 202 that runs from a motor unit 300. The motor unit contains a power train (for example, a gear-motor) that rotates a drive wheel 302; the drive wheel grips the wire and, when energised, advances the loop in either direction, thereby turning wheels 200 at each post and rotating the trackers in unison.

[0134] Located inline with the wire loop is a motion-transmission system 204. System 204 may house a spring tensioner, shock absorber, shear pin, magnetic breakaway, or any equivalent component that moderates transient loads and, if wind torque exceeds a preset threshold, decouples the wire so the trackers can re-orient freely without overstressing the motor or cable.

[0135] In operation the computing unit (not shown) commands motor 300 to rotate a wheel 302 attached thereto, to pay out or reel in wire 202, thus rotating both vertical support members 102 as required. The conditions, instructions and method for commanding the motor 300 will be further described with reference to the present invention.

[0136] Although Figure 2 depicts two trackers driven by a single loop, additional posts may be added in series, further motion-transmission devices 204 may be inserted as required, and any equivalent motor, wheel or cable arrangement may be substituted without departing from the scope of the invention.

[0137] The motor 300 and mechanical transmission such as wheel 302, cable 202, and pulleys 200 may be considered a drive mechanism, this may further include any associated power-conversion or control electronics— regardless of whether those electronics are mounted locally at the post or remotely in a control cabinet or the like.

[0138] In another aspect of the present invention, illustrated in Figure 3, the solar-panel mounting system 100 of Figure 1 is adapted to employ a wire-and-wheel drive arrangement and a compact stop geometry. A rotatable vertical support member 102 again projects upward from a ground anchor (not shown) and is arranged to turn about its substantially vertical longitudinal axis under the action of a suitable drive means. In this embodiment the drive means may include a tensioned wire (not visible in this view) that is wrapped about a wheel 200 mounted at the base of support member 102; wheel 200 provides one illustrative form of the "wire-receiving wheel" referenced elsewhere in the specification and may incorporate a circumferential groove, capstan ribs or a tapered drum profile to grip the wire during bidirectional rotation.

[0139] Rigidly secured to the upper end of support member 102 is the upper solar-panel support member 105, again defining the horizontal hinge axis about which a plurality of panel frames 500 can pivot. To prevent the panels from over-rotating past their nominal rest position when wind loading diminishes, a stop member 104B is provided. Stop member 104B constitutes a downsized variant of the lower support member 104 described with reference to Figure 1; it is located adjacent upper support member 105— preferably attached to the torque-tube or an adjacent bracket— and presents a fixed abutment surface that limits further downward travel of the associated panel 500 once a preset angular threshold is reached.

[0140] An auxiliary support member 106 spans between the rear surface 500B of the panel and the upper support member 105 (or a neighbouring structural element) to stiffen the module frame against torsion and cross-wind loads. Support member 106 may be a removable brace, strut or tie-rod and can incorporate quick-release couplers to facilitate panel replacement.

[0141] Except for the modified wire-drive wheel 200, compact stop 104B and auxiliary brace 106, the materials, dimensions and modes of operation of the components shown in Figure 3 correspond to those already described for Figure 1. Any optional features disclosed elsewhere herein— such as elastomeric bumpers on the stop surface, tension sensors in the wire, or torsion-bias elements in the hinge brackets— may likewise be incorporated into the embodiment of Figure 3 where compatible.

[0142] In multi-row wire-driven systems, particularly those spanning longer cable lengths, mechanical compliance— including wire stretch, pulley backlash, and post inertia— can introduce control lag or nonuniform motion across tracker rows. To mitigate these effects, the system may implement one or more of the following strategies:

[0143] (i) Tension management subsystems (e.g., spring-loaded tensioners or dynamic cable take-ups) that reduce slack and maintain consistent preload throughout the loop;

[0144] (ii) Dampening or delay compensation algorithms within the control logic, which adjust drive commands based on expected propagation delay between upstream and downstream posts;

[0145] (iii) Distributed sensing, in which torque or tension sensors at multiple points in the loop provide localized feedback to the controller, allowing per-zone adjustments to ensure uniform motion; and (iv) Limit stops or alignment markers at known azimuth angles that allow periodic resynchronization of row positions based on mechanical contact or inferred load profiles.

[0146] These features help maintain synchronized tracking performance and improve overall azimuth accuracy, even under varying load or wind conditions.

[0147] Figure 4 depicts the system of Figure 1 in a wind-reacted state. A gust acting on the rear surfaces 500B has generated an aerodynamic moment sufficient to lift each solar panel 502 about the horizontal hinge axis defined by upper support member 105. As a result, the panels have pivoted upward from their nominal rest position.

[0148] Throughout this motion the connection members (or equivalent hinge brackets) permit free rotation, while the wire-and-wheel drive— represented in this embodiment by wheel 200 at the base of the vertical support member 102— maintains tension without impeding panel feathering. Stop member 104B remains spaced from the underside of the adjacent panel frame; once the gust subsides, gravity or any torsion-bias elements in the hinges will cause each panel 502 to swing downward until it again contacts stop 104B, thereby returning the array to its power-producing orientation.

[0149] In a further aspect of the invention, illustrated in Figure 5, the tracker assembly 100 of the foregoing figures is provided with a cable-suspended stop arrangement that constrains downward panel rotation without requiring a full-width lower rail. A rotatable vertical support member 102 extends upward from a ground anchor (not shown), or a wheel, and is driven about its substantially vertical longitudinal axis in the manner previously described. Rigidly coupled to the upper end of support member 102 is an upper solar-panel support member 105 that defines the horizontal hinge axis for a solar panel 502; the panel's rear surface is identified at 500B Pivotable attachment of the module to the upper support member may be provided by at least one connection member 400.

[0150] Instead of the rigid lower support of Figure 1, downward travel of panel 502 is arrested by a stop member 104 in cooperation with a tensile stop wire 103. Stop member 104 is secured to, or integrally formed with, the lower edge of the panel frame and projects outward so as to engage the stop wire 103 once a predetermined tilt limit is reached. The opposite end of stop wire 103 is anchored to the vertical support member 102 at a selected height and radial offset, thereby defining the maximum allowable pivot angle of the panel about hinge axis 105.

[0151] During normal power-producing operation the stop wire 103 remains slack or lightly tensioned, and the panel follows commanded tracking motions imparted by rotation of support member 102. When wind loading subsides after a gust event, gravity and / or any torsion-bias elements in the connection members 400 cause the panel 502 to swing downward until stop member 104 contacts the stop wire 103, at which point further rotation is mechanically limited. The tensile nature of wire 103 permits a lightweight implementation and accommodates manufacturing tolerances, while still providing a positive, repeatable stop action. Except for the cable-based stop arrangement just described, the materials, dimensions and modes of operation of the components shown in Figure 5 may correspond to those already detailed with respect to Figures 1, 2, 3 and 4, and any optional features disclosed elsewhere herein can be incorporated where compatible.

[0152] In an example of the present aspect of the invention, as shown in Figure 6, the computing unit carries out a continuous, sensorless process for estimating both the heading of each solar-panel structure 500 and the wind load acting upon it. The sequence is identified by reference numerals 1200 through 1209.

[0153] Step 1200: the routine starts and enters its repeating loop.

[0154] Step 1201: the controller reads two kinds of live data— (i) drive-train load, obtained as motor current and / or wire tension in the motion-transmission system 204, and / or (ii) the present wind-speed value, taken from an anemometer if available or from a previously stored drag estimate if not. In some embodiments, both types of sensors are used in combination to improve heading estimation and wind inference. In other embodiments, either sensor may be used independently, depending on system architecture and available data.

[0155] Step 1202 forms a normalised load metric by comparing the opposing wire tensions (or successive motor-current samples). Because aerodynamic torque is lowest when the panel is either edge-on or square-on to the wind and reaches a single peak at an intermediate offset, this metric shows a characteristic double-well profile. During commissioning the system swept the full azimuth range at several wind speeds and stored that load-versus-heading surface. Step 1203 now maps the current load metric— and the measured or inferred wind speed— onto that surface to output a heading estimate. Step 1204 uses the same data to derive the instantaneous drag, which can serve as a backup wind reading if the anemometer is unavailable.

[0156] Step 1205 checks whether the combination of heading and drag fits inside an expected operating window based on the row's commanded position and its zone-specific wind limits.

[0157] If the load is outside the window the routine moves to step 1206, where it raises a diagnostic flag indicating a possible fault such as a jammed hinge, wire slip or unusually strong cross-wind.

[0158] If the load is inside the window the routine follows the normal path to step 1207.

[0159] In either case the routine proceeds to step 1208, packaging the current heading estimate, wind figure and any diagnostic flag for transmission to the site controller or SCADA system.

[0160] Step 1209 ends the present cycle; after a short, fixed delay the algorithm returns to step 1201 and repeats. Running continuously in this manner, steps 1200-1209 give each tracker row a self- calibrating "compass and wind gauge" without adding separate angle sensors or extra weather hardware, while also providing real-time fault detection.

[0161] Figure 7 plots the relationship between the azimuth offset from the instantaneous wind direction (labelled "yaw angle") on the horizontal axis and the resulting torsional load about the tracker's vertical post (labelled "yaw moment") on the vertical axis. Because aerodynamic moment (and therefore wire tension or motor torque) varies approximately with the square of wind speed, the ordinate of Figure 7 can be interpreted as the product ’M(0) x v2', where 'M(0)' is the normalised moment coefficient taken from the curve and 'v' is the measured wind speed. In operation the controller reads the real-time wind vector from the site anemometer, applies the corresponding 'v2' factor, and consults the stored 'M(0)' curve to predict the tension that would arise at any candidate azimuth offset from the wind direction. The actual tension in the drive— sensed by a tensiometer or inferred from motor current— is then compared with these predictions. This yields a small set of plausible panel headings that are consistent with the measured load.

[0162] To resolve the remaining ambiguity the controller issues a slow, low-energy rotation command. If the observed tension falls the system infers it is moving toward the nearest minimum and continues in that direction; if tension rises, it reverses. By iterating in this way the tracker converges on the azimuth offset that produces the lowest load for the present wind speed, conclusively identifying whether the wind is striking the module faces or the module backs, and thereby determining the true azimuth of the row without any rotary encoder. At high wind speeds the same routine automatically drives the tracker to the edge-on (load-minimum) position, providing an inherent wind-relief function while simultaneously limiting torsional stress on the drive components and preserving the structural integrity of the tracker array.

[0163] The system's capability to infer wind speed and direction from drive-train load has been confirmed through empirical testing, demonstrating that motor current and cable tension provide reliable indicators of aerodynamic loading in real-world conditions.

[0164] As a further example of this aspect of the present invention, demonstrated in Figure 8, whereby the computing unit runs a repeating power-optimisation loop made up of steps 1400 to 1412. At 1400 the loop starts. In 1401 the tracker is set to its normal angle for electricity production. Step 1402 vertically rotates the tracker by a small angle. After that move, step 1403 waits briefly so the electrical output can stabilise, and step 1404 reads the actual AC or DC power coming from the modules.

[0165] Step 1405 checks whether this new power reading is higher than the previous one. If power went up, step 1406 tells the tracker to keep stepping vertical rotation in the same direction. If power did not go up, the logic goes to 1407. There, if power actually dropped, step 1408 makes a single step the other way; if the change is almost zero, step 1409 holds the present position and the loop ends at 1412 until the next scheduled run.

[0166] Before any fresh step is taken— whether called for by 1406 or by the reversal in 1408— the safety check at 1410 looks at the current wind load, this may be by any method described in this disclosure. If wind is below a preset safe limit, the loop returns to 1402 for another test step. If wind is too high, step 1411 pauses the optimisation and follows the site's wind-protection rules (for example going into stow mode). When conditions are safe again the loop resumes at 1402.

[0167] The cycle ends at 1412 either because the "hold" condition of 1409 was reached or because some outside rule (high wind, hail, snow, wildlife mode) stopped further optimisation. After a fixed wait the controller restarts the process at 1400. In this way each tracker keeps edging toward the real power peak while automatically backing off whenever wind becomes a concern, all without extra light sensors or complex forecasts.

[0168] The loop terminates at step 1412 either because the hold condition of step 1409 was reached or because an external rule (wind, hail, snow, wildlife) overrides further optimisation. After a fixed interval the controller restarts the sequence at step 1400, ensuring the tracker continually nudges itself toward the local maximum power point while respecting real-time wind constraints. By relying solely on incremental motor commands and on module-level power readings, the routine achieves micro-tracking accuracy without additional irradiance sensors or complex predictive modelling.

[0169] In one embodiment of the present invention, the power-optimisation loop of steps 1400-1412 is executed in concert with the sensorless orientation-and-wind routine of steps 1200-1209. During each optimisation cycle the computing unit first invokes the heading-estimation routine; the numeric load metric and inferred wind-pressure value produced at step 1208 are cached for the duration of the power test. The cached heading confirms the tracker is pointing where the controller believes, while the cached wind index supplies the "wind safe?" decision at step 1410. In this way the perturb- and-observe algorithm can be suspended, resumed, or amplitude-limited in real time without relying on separate encoders or dedicated anemometers.

[0170] Conversely, each small adjustment issued at step 1402 or 1408 also gives the controller a fresh data point: the system sees how much extra motor load that move caused and adds the result to the table used in steps 1201-1204. Over time these extra samples keep the load-to-heading calibration up to date, even as wires stretch or hardware ages. The two routines therefore reinforce each other— loadbased heading estimates keep the power search safe in windy conditions, and the power search keeps the heading estimates current— all without adding new sensors beyond the existing motor- current or wire-tension reading and, optionally, a single site anemometer for initial wind-speed as a reference.

[0171] In an additional control mode the computing unit continuously compares the measured electrical output of each solar-panel structure 500— obtained, for example, from string-current sensors, module-level DC optimisers, or inverter telemetry— with the expected output calculated from the sun's position and the last commanded azimuth angle. If the measured output of a given tracker falls more than a predetermined margin (e.g., eight percent) below its expected value while an adjacent tracker remains within the margin, the controller infers that the adjacent tracker is casting shade. To correct the condition, the controller applies a small azimuth offset (typically two-to-five degrees) to one or both of the involved trackers so that their panel edges separate and the shaded tracker regains full irradiance. After a short verification interval (e.g., sixty seconds) the controller re-checks power: if the shortfall has fallen below a lower margin (e.g., two percent) the new headings are retained, otherwise further two-degree corrections are applied up to a safety limit (e.g., fifteen degrees from nominal). Because individual panels may rest slightly off their theoretical tilt after wind-induced feathering, this closed-loop routine compensates for real-world misalignment without requiring additional tilt sensors and maximises energy yield under dynamic conditions.

[0172] In a further control mode the computing unit monitors the agreement between the wind-direction sensor and the drive-train response. For each control cycle it logs the wind vector reported by the sensor and the azimuth that produces the lowest wind-induced load as inferred from existing feedback signals— such as motor current, cable-tension balance, or the symmetry of passive hinge lift. If the reported direction differs from the mechanically implied direction by more than a configurable tolerance (e.g., five-to-ten degrees) for a sustained period, the controller flags the discrepancy as a probable wind-sensor bias or mounting shift. Upon detecting such a persistent offset the controller can, without adding hardware:

[0173] • apply an automatic software correction so that subsequent wind responsive decisions use a compensated wind vector; and / or

[0174] • send an alert via SCADA or other telemetry indicating that the wind sensor may require inspection or manual re-zeroing.

[0175] The correction factor is stored in non-volatile memory and re-evaluated whenever conditions change, allowing the tracker field to maintain reliable wind protection even if the physical wind sensor drifts, vibrates loose, or is partially obstructed.

[0176] A further example of one aspect of the present invention is shown in Figure 9, where the computing unit executes an adaptive power-optimisation method that can operate in two modes: learn mode for previously unseen conditions and optimise-from-learning mode when a condition has been encountered before.

[0177] At step 1500 the routine begins a new optimisation cycle. In step 1501 the controller records an environmental signature that may include time-of-day, irradiance class, ambient temperature, wind class, and any site-specific shading indices. Step 1502 compares that signature with a lookup table stored in non-volatile memory.

[0178] If the signature is already present, the algorithm enters optimise-from-learning mode. At step 1503 the tracker is rotated directly to the best mechanical angle previously learned for the same conditions. Step 1504 performs a live wind-safety test using the load-and-drag estimator disclosed elsewhere. If wind load exceeds the safe limit the controller executes step 1505, invoking the appropriate wind-protection routine, after which the cycle terminates at step 1506. If conditions are safe, the cycle likewise ends at step 1506 and normal energy production continues.

[0179] If no matching signature exists the algorithm branches to learn mode at step 1507. The tracker is first set to its nominal sun-table position in step 1508, then nudged by a small azimuth increment in step 1509. After a brief dwell (step 1510) to allow electrical output to stabilise, step 1511 samples the AC or DC power produced by the modules.

[0180] Step 1512 compares the new power reading with the previous reading. If power has risen, the controller follows step 1513 and continues stepping in the same direction. If not, decision node 1514 determines whether the power has actually fallen. A decrease causes a single reverse step at 1515; a negligible change causes the tracker to hold its current position at step 1516.

[0181] Before each fresh test step the controller executes the wind-safety check at step 1517. If wind load remains within the allowable band, control returns to step 1509 for another perturb-and-observe cycle. If the load is too high, step 1519 pauses the learn process and calls the wind-protection strategy, resuming the loop once conditions are again safe.

[0182] When the perturb-and-observe process detects no further power gain (step 1516), the controller stores the final optimal angle together with the environmental signature in step 1518, adding a new entry to the lookup table. The cycle then ends at step 1506. Subsequent cycles that encounter the same conditions can therefore bypass learn mode and move directly to the proven best angle, minimising production loss while still allowing the system to learn whenever it sees a new operating scenario. In this manner the power-optimisation routine continuously improves its own performance database while automatically deferring to wind-protection measures whenever required, all without additional irradiance sensors or manual recalibration. Various improvements to the herein disclosed aspects and embodiments can include performing a calibration sweep during commissioning in which the drive motor rotates the wire loop through its full azimuth range under low-wind conditions. During that sweep, the controller records steady-state values from one or more load-feedback sensors, including but not limited to motor current and mechanical tension, and stores the resulting torque-versus-heading data in non-volatile memory.. In normal operation, live tension or current measurements can then be interpolated against this stored calibration curve to estimate each row's true heading without requiring any encoder or inclinometer. Over time, additional calibration sweeps may be conducted automatically or manually to account for cable stretch, wheel wear, or other mechanical drift, ensuring that the load-to-angle mapping remains accurate throughout the system's lifetime.

[0183] In certain embodiments, the controller may compare tension on opposing sides of a given loop (or successive tension samples when a single-loop sensor is used) so as to detect azimuth misalignment relative to the wind or crosswind loading. Whenever the absolute difference between left and right wire tensions exceeds a predetermined threshold, a diagnostic flag is raised and corrective action may be taken— such as issuing a small azimuth adjustment or generating an alert to maintenance personnel. This left-versus-right tension comparison provides a continuous indication of crosswinds or mechanical slip without adding any per-row sensors, and serves to maintain safe stow functionality even if a primary wind sensor is offline.

[0184] Further embodiments can leverage the known panel angle (derived from load-to-angle calibration) together with the measured drive tension to invert a simplified aerodynamic model and thereby infer local wind speed and, in some cases, wind direction. By solving for the wind speed that would generate the observed tension at a given panel angle, the controller effectively transforms the drivetrain data into a proxy anemometer reading. In this way, no dedicated wind sensor is strictly required for stow logic; instead, either the mechanical load-feedback sensor (e.g., wire tension) or the motor current sensor— or both— may serve as a proxy wind gauge by indicating aerodynamic loading on the panel. The control logic may operate using either input alone, or may combine both for improved robustness and redundancy in wind-stow decision-making.

[0185] Because the drive-load mapping yields an independent heading estimate, the tracker can continue normal tracking and wind-stow operations even in the event of failure of any external angle sensor or wind sensor. For example, if a per-row encoder stops reporting or an inclinometer becomes stuck, the controller seamlessly switches to the load-based heading estimate to maintain accurate sun alignment. Similarly, if the site's anemometer goes offline, the system relies on the inferred wind speed from motor current or wire tension to execute safe stow logic. In this manner, the system remains robust and reliable, with sensorless redundancy ensuring that neither structural safety nor energy yield is compromised by single-point failures.

[0186] The closed-loop power-optimisation routine described above operates solely on module-level power measurements obtained from existing inverters or DC / AC meters, without the need for any dedicated irradiance sensors or sun-tracking cameras. By comparing actual AC or DC power before and after each small ±1° azimuth perturbation, the controller identifies the local maximum-power orientation in real time. Because no external light sensor or tilt meter is required, the hardware complexity is kept to a minimum while still adapting to unmodeled shading, soiling, and panel drift.

[0187] Whenever an external condition such as a wildlife-safe interval (e.g., nighttime or motion-detector event) or a hail, snow-stow requirement becomes active, the power-optimisation loop automatically pauses and reverts the panels to a predefined safe orientation. Once the wildlife event ends or the hail, snow load subsides, the perturb-and-observe routine seamlessly resumes from its last tested angle. This approach ensures that wildlife and hail, snow protection are integrated into the same control framework as wind-stow and power optimisation, all without requiring additional hardware beyond the existing drive motor, tension sensor, and inverter power meter.

[0188] An example of an environmental priority routine such as a hail, snow-stow implementation is shown in Figure 10. At step 1300 the environmental-priority routine is initialised and enters its repeating control cycle. The controller immediately evaluates, at decision node 1301, the live wind-load index— this index being derived either (i) from the site anemometer or (ii) from the proxy drive-train metric (motor-current or wire-tension differential) already defined in the present disclosure. If the index meets or exceeds the stored wind-stow threshold for the current tracker row, control advances to action step 1302: the drive mechanism 300, 202, 200 rotates the vertical support member 102 until the narrow edge of the solar-panel structure 500 is aligned with the incoming wind vector, thereby placing the row in its wind-aligned stow orientation. The routine then terminates the current pass at terminal 1308.

[0189] When the wind-load index is below the wind-stow threshold, the algorithm proceeds to the second environmental priority, decision node 1303. Here the controller tests whether (a) precipitation is occurring and (b) ambient temperature lies below a predefined hail, snow limit held in non-volatile memory. If both conditions are true, the routine branches to action step 1304, rotates the tracker so that the module faces point away from the prevailing wind (rear surfaces 500B windward), and exits via terminal 1308. This hail, snow-mitigation attitude promotes sliding of accumulating hail, snow from the front glass while allowing the passive horizontal hinge to feather further should gusts develop. In an alternative embodiment, the same control logic may be applied for hail mitigation. During conditions of elevated hail risk— such as active site alerts, weather forecasts, or inferred atmospheric cues— the controller may execute a hail-mitigation routine in place of or in addition to hail, snow mitigation. In this case, the tracker is rotated such that the rear surfaces (500B) of the solar panels face the prevailing wind or hail vector. This orientation reduces the effective angle of hail impact on the exposed surfaces, leveraging the panel's rear-side construction to minimize potential damage from hailstones. Unlike hail, snow mitigation, which primarily promotes hail, snow shedding via gravitational sliding, the hail-mitigation strategy prioritizes structural protection by altering panel orientation to reduce impact severity without relying on active tilt control.

[0190] If the precipitation-and-temperature test is negative, the routine evaluates its third priority at decision node 1305: the controller checks whether (i) the installation is within a nighttime interval determined by the real-time clock, or (ii) wildlife activity has been asserted by motion sensors or other site telemetry. A positive result causes a branch to action step 1306, in which the controller commands the tracker assemblies into a predefined wildlife-safe orientation. In one embodiment, this orientation comprises arranging perimeter tracker rows in continuous straight lines— typically aligned along one of the row-direction families of the polygon grid— thereby forming a visual and physical perimeter barrier that discourages nocturnal fauna from entering the array interior. In other embodiments, all trackers may be aligned into straight rows across the site to simplify detection of wildlife intrusions or to avoid creating gaps between tracker assemblies. Once the wildlife-safe orientation is established, the cycle ends at terminal 1308..

[0191] Should none of the protective triggers be active, the algorithm follows the default path to step 1307, allowing the tracker to remain in, or resume, its ordinary sun-tracking and power-optimisation sequence previously described with respect to steps 1400-1412. The cycle then terminates at 1308. After a fixed scheduler interval (for example, ten seconds) the routine restarts at step 1300, thereby ensuring that wind protection always overrides hail, snow mitigation, which in turn overrides wildlife-safe positioning, while normal energy-harvesting operation is maintained whenever environmental conditions permit. Because the only powered axis is the vertical rotation about member 102 and the horizontal hinge is passive, all three protective modes are achieved without additional tilt actuators or per-row environmental sensors, preserving the low-hardware, sensor-less philosophy of the invention.

[0192] In a further, independent yet complementary aspect of the present disclosure the inventors have recognised that the sensor-less, wind-responsive trackers disclosed above perform best when the surrounding plant geometry also minimises mutual shading, cabling distance and surveying error. To that end, Figures 11-13 illustrate preferred non-rectangular lattice arrangements— specifically polygon meshes such as regular-hexagon and equilateral-triangle meshes— not previously adopted in utility-scale single-axis tracker farms.

[0193] Unlike the conventional square grid, a polygon lattice more closely matches the circular rotation envelope of each vertical-axis tracker tree 100. The result is a higher ground-coverage ratio, shorter and straighter cable corridors for the tensioned wire 202 and DC cabling, and six (hexagon) or three (triangle) built-in row-direction "families" spaced 60 ° apart. These multiple natural alignments dovetail with the context-aware algorithms already described:

[0194] Wind protection: perimeter rows remain unambiguously identifiable in the lattice, allowing the environmental-priority routine of Fig. 10 to apply zone-specific stow thresholds without additional surveying data.

[0195] Sensor-less heading estimation: the equal-leg geometry fixes each tracker's neighbours at known bearings and distances, simplifying the shielding-factor term used in the load-to-heading map of Fig. 6.

[0196] Power-feedback micro-tracking: the controller can switch entire family groups between the six (or three) row directions at dawn and dusk to avoid edge shading, exactly as the perturb-and-observe loop of Fig. 8 requires, without widening row spacing.

[0197] Drive-cable logistics: straight polygon edges act as rail-beds for the distributed wire loop, reducing accumulated backlash and easing the fatigue-tracking scheme described later in this specification.

[0198] Because these geometric advantages are achieved purely at the civil-layout level, no extra sensors, actuators or control hardware are needed; the approach therefore preserves the low-cost, low- maintenance philosophy that underpins every other subsystem of the invention.

[0199] Illustrative Control Logic Pseudo-Code

[0200] To further aid in understanding the present invention, the following sections provide illustrative examples of control logic in pseudo-code form. These examples are intended to demonstrate one possible implementation of the control routines described in this specification and are not limiting as to the scope of the invention.

[0201] Example 1: Sensorless Heading Estimation Routine

[0202] Function EstimateTrackerHeading():

[0203] Loop: load_metric = ReadLoadFeedback() / / Obtain motor current or cable tension wind_speed = ReadWindSensor() OR EstimateWindSpeedFromLoad() normalizedjoad = NormalizeLoad(load_metric, wind_speed) heading_estimate = MapLoadToHeading(normalized_load, wind_speed)

[0204] StoreHeadingEstimate(heading_estimate)

[0205] If LoadOutOfExpectedRange():

[0206] RaiseDiagnosticFlag()

[0207] TransmitHeadingAndDiagnostics()

[0208] Wait(Samplelnterval)

[0209] Example 2: Wind Responsive Orientation Logic

[0210] Function WindResponsiveControl(): wind_direction = EstimateWindDirection() wind_speed = EstimateWindSpeed()

[0211] If wind_speed > WindProtectionThreshold: target_azimuth = wind_direction + 180 / / Orient rear side toward wind

[0212] RotateTrackerTo(target_azimuth)

[0213] Else:

[0214] MaintainNormalTracking()

[0215] Example 3: Power Optimisation Loop (Perturb-and-Observe Method)

[0216] Function PowerOptimisation(): previous_power = MeasurePowerOutput()

[0217] StepAzimuth(SmallAngle)

[0218] WaitForStabilisation() current_power = MeasurePowerOutput()

[0219] If current_power > previous_power:

[0220] ContinueSteppingSameDirection()

[0221] Else If current_power < previous_power:

[0222] StepAzimuth(OppositeDirection)

[0223] Else:

[0224] HoldPosition()

[0225] Example 4: Load-Based Wind Estimation

[0226] Function EstimateWindSpeedFromLoad(): load_metric = ReadLoadFeedback() heading = GetCurrentEstimatedHeading() wind_speed = lnvertLoadModel(load_metric, heading)

[0227] Return wind_speed

[0228] Figure 11 is a schematic plan view showing but one example of a representative portion of a photovoltaic plant in which the tracker foundations are disposed on a regular hexagonal lattice. Each solid dot marks the centre-of-rotation of a respective solar-tracking assembly or tree 100, while the surrounding hexagon depicts the notional unit cell used for civil set-out. Because every tracker centre lies at a vertex shared by three adjoining hexagons, each tracker has six equidistant neighbours separated by 60 ° radial lines, and the circular rotation envelopes of the module structures 500 (not illustrated for clarity) may touch but do not overlap. This geometry preserves a full 360 ° sweep for every row yet raises ground-coverage ratio by roughly 12 - 14 % compared with a conventional square grid.

[0229] Figure 12 is an enlarged fragment extracted from the lattice of Figure 11, illustrating three neighbouring trackers. Dashed lines join the tracker centres C to form an equilateral triangle T whose side length equals the centre-to-centre pitch. The three circles represent the rotation envelopes of the associated module structures; they are mutually tangent at the mid-points of the triangle sides, so any further increase in tracker diameter would cause immediate interference. During construction, once two trackers establish one side of the triangle, the position of the third tracker is fixed by a single length measurement, eliminating cumulative surveying error even over undulating terrain.

[0230] Figure 13 depicts a larger sector of the same hexagonal field with the module structures 500 drawn as elongated rectangles mounted on their respective solar tracking assemblies 100. Opposed straight conductors or motion-transmission members— such as the tensioned drive cables 202 and a return span 200— follow alternate hexagon edges, thereby remaining parallel, of equal length and free of abrupt bends. The module rectangles are shown aligned in one of the six discrete orientation families inherent in the lattice; selection of a different family (offset by 60 °) would simply rotate every rectangle about its post without altering the centre spacing. This built-in choice allows the control system to adopt the row alignment that minimises dawn-and-dusk shading while maintaining the tighter packing density delivered by the hexagonal grid.

[0231] Additional advantages of the polygon lattice can be appreciated with reference again to FIGS. 11-13.

[0232] Because every solar tracking assembly 100 is equidistant from its fixed set of neighbours, civil tolerances are relaxed: once the first two posts are placed, the position of the third is prescribed by a single, repeated centre-to-centre dimension. The layout therefore propagates length error rather than angular error, so cumulative mis-alignment does not grow with array size. In rocky or undulating terrain this equal-leg geometry allows each local triangle (or hexagon) to "float" slightly in height without breaking row continuity, reducing earth-works relative to a rigid rectangular grid. Conductor runs, hydraulic hoses, or data fibres can be clipped to the straight edges of the polygons; because consecutive spans are equal, voltage-drop, pressure-loss, and signal-latency calculations all simplify to a one-dimensional repetition. The polygon mesh also lends itself to phased construction: each completed hexagon (or triangle) is a self-contained unit that can be commissioned, energised and revenue-generating while neighbouring cells are still under build-out. The polygon mesh need not be perfectly regular. So long as the tracker centres are positioned on the corner points of a repeating network of straight-edged polygons— i.e. a tessellation in which adjacent edges lie on common straight lines— the key benefits (self-registration of post spacing, straight cable corridors, multiple discrete row-direction "families", and low cumulative survey error) are realised. In practice this means that the interior angles at the tracker centres need not all be exactly 60 °: for example, a sequence of polygons having angles of 70 °, 70 ° and 40 ° around each centre still provides the same continuous straight-edge rows and hence most of the geometric advantages, albeit with slightly less symmetry. Accordingly, references herein to "regular hexagons", "equilateral triangles" or "60-degree spacing" are illustrative only; comparable results are obtained with any tiling whose polygons share common edge lines, even when those polygons are merely equi-sided or otherwise nearly regular.

[0233] For exemplary purposes, to illustrate how the control logic exploits the multiple inherent row directions of a triangular lattice, consider a plant in which the tracker pivots occupy the vertices of congruent equilateral triangles, as schematically suggested by the three-envelope contact drawing of FIG. 12. In this example one side of every unit triangle runs on a strict north-south bearing, and the six azimuth "families" therefore lie at 30°, 90°, 150°, 210°, 270° and 330° as measured clockwise from true north. The computing unit stores these six headings as shade-avoiding set-points and switches between them under the real-time power-optimisation scheme already disclosed.

[0234] The above sequence shows that only three step events (07:00, 09:00, 16:00) and two mirror events (12:00, 19:00) are required throughout the day; all other motion is micro-tracking of a few degrees. Because each family heading is shared by the entire lattice, no row-to-row phasing is needed, and mutual shading remains below 1 % of active aperture even at extreme solar declinations. The same rule set is reversible for southern-hemisphere sites simply by exchanging the family headings in the look-up table.

[0235] In one aspect of the present invention, a site-design algorithm is provided for automatically generating an optimized ground layout for a solar-tracking plant. The algorithm receives as input a boundary outline defining the available land area for installation and computationally fills the interior of the boundary with a repeating polygonal grid, such as a regular hexagonal or equilateral triangular mesh. The algorithm automatically positions the centres of rotation of the solar-tracker assemblies at the corner points of the generated grid, ensuring uniform spacing, mechanical clearance between adjacent tracker assemblies, and maximized land-use efficiency. In particular, the site-design algorithm is configured to automatically trim or omit partial grid cells that intersect the site boundary, thereby avoiding tracker placements that would extend beyond the usable land area or result in incomplete rotation envelopes. The algorithm may further incorporate technical constraints such as prescribed minimum clearances between tracker assemblies, cabling corridors, or access pathways. By automating the layout process in this manner, the algorithm improves construction efficiency, reduces surveying error, and ensures that the deployed solartracking plant maintains optimal ground-coverage ratio while preserving reliable, interference-free mechanical operation.

[0236] In various embodiments, the site-design algorithm configured to generate the repeating hexagonal or triangular grid is not merely an abstract geometric construction, but provides a technical effect by optimizing the spatial arrangement of solar tracker assemblies within irregular site boundaries. Specifically, the algorithm maximizes ground-coverage ratio while preserving critical mechanical clearances between adjacent tracker rotation envelopes, ensuring collision-free operation throughout the full 360-degree azimuthal rotation range of each assembly. This results in a tangible improvement in system performance and installation efficiency compared to conventional rectangular or manually laid out grids.

[0237] The algorithm further contributes to technical field implementation by automatically trimming partial grid shapes at the boundary of the site, thereby eliminating the need for manual adjustment or post-layout surveying corrections. By systematically applying these geometric constraints during the automated layout process, the system reduces cumulative alignment errors, simplifies tracker placement during construction, and facilitates straight cable routing along grid edges, which collectively enhance mechanical reliability and electrical efficiency of the deployed plant.

[0238] The layout algorithm disclosed herein provides a technical contribution beyond mere abstract geometry. It generates a tracker placement pattern that inherently ensures mechanical compatibility between adjacent tracker assemblies, avoids rotational envelope interference, simplifies construction alignment, and facilitates straight, uniform cable routing. These effects produce practical, real-world advantages during site installation and operation, including improved land-use efficiency, reduced installation errors, and enhanced mechanical and electrical system integrity.

[0239] When wind-responsive, wildlife-safe, hail, or hail, snow-mitigation modes are invoked, the shadeavoidance logic yields to the environmental-priority routine of FIG. 10; once the temporary constraint is cleared, the plant re-enters the family schedule at the next appropriate clock event. In this manner the triangular (or hexagonal) lattice interlocks seamlessly with the load-derived heading estimator, adaptive wind stow, and power-feedback optimiser already taught, while achieving a tighter ground-coverage ratio and materially shorter cabling than rectangular designs— all without adding sensors, encoders or extra actuators. In various embodiments, each hinge bracket or connection member includes a torsion-bias element— such as a spring, torsion bar, or friction-damped pivot— configured to resist upward panel rotation under wind load. The bias torque is selected to maintain the panel in its normal tracking tilt under light wind, while yielding under higher wind pressure to allow passive feathering.

[0240] Typical preload values may range from 5 to 50 Nm per hinge, depending on panel area, weight, and local wind design speeds. The resisting torque may be factory-calibrated or field-adjustable, allowing tuning for different module types or structural standards. In some configurations, the element provides nonlinear stiffness, increasing torque resistance as angular deflection increases, thereby improving stability during gusts while still allowing reliable stow return under gravity.

[0241] In various embodiments, the system includes both a motor current sensor and a mechanical loadfeedback sensor (such as a wire tension sensor). These sensors may be used independently or together. The motor current sensor provides electrical load data representative of drive resistance, while the load-feedback sensor offers direct mechanical load measurements. The control system can use either sensor as the basis for estimating azimuth, inferring wind conditions, or triggering wind- stow actions. This modular sensor strategy allows for adaptation to diverse hardware configurations and supports graceful degradation under sensor fault conditions.

[0242] In certain embodiments, the drive mechanism further comprises a torque sensor operatively coupled to the drive shaft or transmission components. The torque sensor is configured to directly measure rotational torque applied to the rotatable support member, providing real-time data representative of mechanical load, aerodynamic resistance, or drivetrain performance. Suitable torque sensors may include, but are not limited to, strain-gauge-based shaft torque sensors, torsional load cells, magnetoelastic torque transducers, or optical torque sensors.

[0243] The output from the torque sensor may be provided to the computing unit as part of the system's load-feedback architecture. In some embodiments, the torque sensor operates in conjunction with or as an alternative to other load-feedback devices, such as motor-current sensors or cable-tension sensors. The computing unit may utilize the torque sensor data to estimate the current azimuthal orientation of the solar panel structure, infer wind direction and magnitude, detect drivetrain anomalies (e.g., jamming, excessive mechanical resistance), and execute wind-responsive orientation logic.

[0244] In systems incorporating both a torque sensor and other load-feedback sensors, the computing unit may apply cross-validation techniques to enhance measurement reliability. For example, discrepancies between the torque sensor reading and the motor current or cable tension may trigger diagnostic routines, sensor substitution, or conservative safety modes as described elsewhere herein. The use of a direct torque sensor provides an additional, independent data source for assessing mechanical loading, which is particularly advantageous in configurations where drivetrain elasticity, backlash, or passive tilt mechanisms introduce uncertainties in indirect load measurements. By capturing torque data at the drive shaft or load path itself, the system can achieve improved accuracy in azimuth estimation and wind-response control, contributing to enhanced structural safety and energy optimization under dynamic environmental conditions.

[0245] Because motor-current data arise only when the drive is actively energised, those readings are most informative during commanded motion. In contrast, the tension sensor yields a continuous load signal— even with the drive locked— thereby enabling static wind-load estimation and fault detection while the array is fully stationary. The controller therefore prefers motor-current feedback for low- cost installations and supplements it with tension sensing where continuous load visibility is required.

[0246] By eliminating per-row encoders, tilt sensors, and additional anemometers, the present architecture significantly reduces both capital expenses and ongoing maintenance burdens. The only additional hardware elements required are a tension sensor or a current sensor at each drive motorcomponents that can be housed in a single, weatherproof enclosure per loop instead of at every tracker post. This reduction in sensor count and wiring harness length directly lowers the bill of materials and simplifies field installation, preserving the cost advantage of multi-row, wire-driven tracker systems while still delivering robust, sensorless control and real-time feedback.

[0247] "Motor current", as used herein, denotes the instantaneous electrical current flowing through the drive motor (300). It is measured on the motor's power leads by a current-sensing device (e.g. Halleffect sensor or shunt resistor) and is sampled by the computing unit as an indirect indicator of mechanical torque, aerodynamic load and panel azimuth.

[0248] "Drive load" denotes a quantitative parameter that is representative of the mechanical resistance encountered by the drive mechanism while rotating the tracker about the vertical axis. The parameter may be obtained from one or more of:

[0249] (a) the motor-current value defined above, which is proportional to drive torque;

[0250] (b) a shaft-torque signal delivered by an in-line torque sensor; or

[0251] (c) a cable-tension value measured in the tensioned cable loop that transmits motion to the rotatable support member.

[0252] The term therefore covers any measurable quantity whose magnitude rises or falls with the torsional load acting on the drive. The expression "calibration sweep" denotes a controlled rotation of the tracker through essentially its entire usable azimuth range, preferably from one mechanical end-stop to the opposite end-stop, at a deliberately slow angular speed— for example about two degrees per second or less— while site wind is calm, for instance below about four metres per second. During the sweep the normal suntracking routine is paused. At regular angular intervals, typically every half-degree to five degrees, the control system samples the chosen drive-load parameter and stores each azimuth-and-load pair in non-volatile memory. The stored map is then used for sensorless azimuth estimation and windload inference. When mechanical drift causes mapping error to exceed a preset tolerance, the sweep can be repeated automatically to refresh the calibration.

[0253] In one aspect of the present invention there is provided a sensorless azimuth-estimation system specifically designed to address the technical challenges introduced by the use of wind responsive tracking systems comprising vertical-axis rotation in a wire-and-wheel drive architecture, with passive horizontal lift. Unlike conventional horizontal single-axis trackers, systems with vertical rotation driven by distributed cable loops must contend with mechanical backlash, cable elasticity, and variable tension across multiple posts. These effects make precise angular tracking difficult using motor commands alone. At the same time, the presence of passive horizontal lift renders traditional tilt sensors ineffective, as panel angle varies dynamically with wind loading and is not motor- controlled. The disclosed system overcomes these problems by calibrating a load-to-angle map during a one-time commissioning sweep in low-wind conditions, recording motor current and / or wire tension at known headings. During operation, real-time drive load samples are used to estimate each row's azimuth without the need for rotary encoders or tilt meters. This is particularly effective in our configuration because the vertical drive axis is the only controllable rotation, while horizontal tilt remains unpowered and reactive to environmental conditions.

[0254] In another aspect, the invention addresses the unique wind-stow requirements of wind responsive vertical-axis, passively-tilted tracker systems. Unlike traditional designs that depend on active control over both tilt and azimuth, the disclosed system must achieve reliable wind relief using only the vertical drive axis and passive mechanical elements. The challenge is that any active tilt response is mechanically decoupled from the controller, relying instead on aerodynamic forces to initiate motion. One aspect of the present invention solves this by rotating the tracker azimuthally so that the rear surface of the panel faces the wind, allowing the panel to feather upward on its horizontal hinge. This combination of vertical active positioning and passive tilt yields effective 360-degree wind relief using a single actuator and no real-time wind sensor, an outcome not achievable with traditional two-axis or fixed-tilt architectures. To address long-term energy loss due to mechanical degradation specific to wire-driven systems, particularly those in wind responsive systems with dual rotation axis— such as cable stretch, wheel slip, and hinge wear— one aspect of the present the invention further provides a closed-loop azimuthal power-optimisation routine. Because the panels can passively rotate around the horizontal axis, the actual irradiance on each module cannot be inferred from sun models alone. Instead, the controller introduces small azimuth perturbations and compares the resulting changes in real-time electrical output. This perturb-and-observe loop allows the tracker to converge on the true maximum-power orientation despite mechanical drift, shading, or passive tilt deviations— conditions that are especially prevalent in the vertical-horizontal configuration of the present wind responsive system.

[0255] In another aspect of the present invention, the invention offers a load-based wind detection capability tailored to wind responsive tracker architecture. Since the only controlled motion is vertical-axis rotation and panel tilt is passive, changes in wind load manifest as changes in drive torque or cable tension. By comparing live load readings to a stored map of torque versus azimuth angle, the controller estimates both wind direction and speed without the need for a dedicated anemometer. This proxy wind-sensing approach is particularly well-suited to the presently disclosed wire-driven system, where each loop inherently senses the distributed mechanical effects of aerodynamic forces across multiple rows.

[0256] The invention also provides a dual-branch tension-difference diagnostic that responds to problems unique to long cable loops and passive panel motion. In such systems, small asymmetries— due to stuck panels, uneven hail, snow accumulation, or partial feathering— can result in substantial differences in left / right cable tension. By continuously sampling both sides of the cable and identifying sustained mismatches, the system can flag faults, initiate corrective rotation, or trigger stow behaviour before damage occurs. This real-time load comparison is essential for maintaining alignment across multi-row systems without per-row sensors.

[0257] A further aspect of the invention includes dual-sensor redundancy and automatic fail-over logic. Because the architecture of the present invention depends on real-time load feedback to infer both orientation and wind conditions, the system monitors both motor current and cable tension as independent sources of mechanical load. If one channel becomes unreliable— due to electrical noise, mechanical hysteresis, or degradation— the controller seamlessly switches to the other input and continues operation, preserving control integrity even under partial sensor failure.

[0258] To ensure that azimuth estimation remains accurate over time despite mechanical changes such as hinge loosening, cable aging, or structural settling, one aspect of the present invention features an on-line machine-learning calibration process. Each new drive load and estimated heading pair is used to refine the torque-angle surface via regression, replacing the original calibration once model error falls below a threshold. This self-correcting process is especially important for passive-tilt, wind responsive architectures, where panel behaviour can evolve due to repeated wind cycling.

[0259] In addition, one aspect of the present invention includes a self-triggering recalibration routine based on rolling energy yield. If the system detects that delivered energy per irradiance unit has dropped significantly from baseline, it automatically schedules a recalibration sweep during low-wind conditions. This helps correct for mechanical drift in systems where drive mechanics and passive panel behaviour may degrade asymmetrically over time.

[0260] One aspect of the present invention incorporates a fatigue-aware monitoring system for the drive cable, which is particularly vulnerable in long-loop, multi-row configurations. By tracking the amplitude and frequency of tension cycles and applying cumulative fatigue models, the controller estimates remaining cable life and informs predictive maintenance, reducing the risk of unexpected failure in high-cycle passive systems.

[0261] These integrated subsystems— sensorless azimuth estimation, passive wind responsive tilt, closed- loop power optimisation, torque-based wind gauging, cable fault detection, dual-sensor redundancy, adaptive calibration, automatic recalibration, and fatigue tracking— are designed specifically to overcome the challenges introduced by wind responsive vertically actuated, horizontally reactive, wire-driven architecture. The resulting system provides robust performance with reduced hardware complexity and increased long-term reliability.

[0262] Further examples of suitable solar tracking assemblies for use with the present invention will now be described with reference to Figures 15 to 22.

[0263] According to one such example, as shown in Figures 15 and 16, there is provided a solar tracking system 100 comprising a rotatable vertical support member 102, a lower solar panel support member 104, an upper solar panel support member 105, and at least one solar panel 500 having a front surface 500A and a rear surface 500B. Optionally, the rotatable vertical support member 102 may be connected to a motor 300 for driving rotation of the rotatable vertical support member 102.

[0264] The at least one solar panel 500 is attached to one or more connection members 400, which in turn are connected to the upper solar panel support member 105. The connection member 400 may be in the form of a hinge or alternative mechanism that allows relative rotation between components. In this case, the solar panel 500 rotates relative to the upper solar panel support member 105, while the upper solar panel support member 105 remains static. The motor 300 rotates the rotatable vertical support member 102 to place the solar tracking system into a desired orientation relative to the position of the sun in the sky or the direction of wind force present. One desired position may be such that the rear surface 500B faces the direction of any wind force present; upon the wind force contacting the rear surface 500B, the rear surface 500B rotates around the upper solar panel support member 105 via the connection member(s) 400. Figure 16 illustrates an example of the solar tracking system 100 of Figure 15, where the solar panel 500 is rotated in the presence of wind from direction X.

[0265] Figure 17 shows the solar tracking system of Figures 15 and 16, where the motor 300 has rotated the solar tracking system via the rotatable vertical support member 102. This Figure clearly shows the front surface 500A of the at least one solar panel 500.

[0266] In another example of a suitable solar tracking assembly, shown in Figure 18, there is provided a solar tracking system 100 comprising a rotatable vertical support member 102, a lower solar panel support member 104, an upper solar panel support member 105, and at least one solar panel 500 having a front surface 500A and a rear surface 500B. Optionally, the rotatable vertical support member 102 may be connected to a motor 300 for driving rotation.

[0267] The at least one solar panel 500 is attached to one or more connection members 400, which in turn are connected to the upper solar panel support member 105. Provided on the lower solar panel support member 104 are one or more dampening members 110, which function to cushion contact between the one or more solar panels 500 and the lower solar panel support member 104. The dampening members 110 may comprise foam, rubber, springs, or any suitable cushioning element as would be readily understood by a person of skill in the art.

[0268] The motor 300 rotates the rotatable vertical support member 102 to place the solar tracking system into a desired orientation relative to the location of the sun or the direction of wind force present.

[0269] A further example of a solar tracking assembly suitable for use with the present invention is shown in Figure 19. The system 100 comprises a rotatable vertical support member 102, a lower solar panel support member 104, an upper solar panel support member 105, at least one vertical solar panel support member 106, and at least one solar panel 500 having a front surface 500A and a rear surface 500B.

[0270] The at least one solar panel 500 is attached to one or more connection members 400, which in turn are connected to the vertical solar panel support member 106. The motor 300 rotates the rotatable vertical support member 102 to place the solar tracking system into a desired orientation relative to wind direction. A further suitable example is illustrated in Figure 20, where the system 100 additionally includes diagonal solar panel support members 107 to provide increased rigidity. The solar panels 500 are again attached via connection members 400 to the upper solar panel support member 105, allowing the panels to rotate relative to the support structure in response to wind forces.

[0271] In yet another example, shown in Figure 21, the solar tracking system 100 includes a horizontal support member 109 located at one distal end of the rotatable vertical support member 102, and arms 108 connecting the horizontal support member to the upper and lower support members 105, 104. This configuration allows rotation of the horizontal support member 109 to raise the solar panels 500 in response to wind.

[0272] Figure 22 demonstrates a solar tracking assembly 100 according to a further example of the present invention, where there is provided a vertical support member 102 supporting a solar panel support member 105, rotatably attached to the solar panel support member 105 is at least one solar panel 500. The rotatable attachment between the solar panel 500 and the solar panel support member 105 is achieved by at least one connection member 400. Weights 450 are attached to the solar panel 500, or to the solar tracking assembly 100 such that the weights 450 provide a counterweight to the weight of the solar panel 500. In other words, the weights 450 bias the solar panel 500 about the solar panel support member 105 to a resting position when there is no, or little, wind present against the solar panel 500.

[0273] Figure 23 demonstrates a solar tracking assembly 100 according to a further example of the present invention, where there is provided a vertical support member 102 supporting a solar panel support member 105, rotatably attached to the solar panel support member 105 is at least one solar panel 500. The rotatable attachment between the solar panel 500 and the solar panel support member 105 is achieved by at least one connection member 400. Further provided are additional solar panel support members 107 supporting the solar panel 500, attached to which is a weight 450. In this example, the weight 450 is in the form of one or more cylinders; however, it should be understood by a person of skill in the art that any form of weight or heavy material may be suitable. The weight 400 provides a counterweight to the weight of the solar panel 500. In other words, the weight 450 biases the solar panel 500 about the solar panel support member 105 to a resting position when there is no, or little, wind present against the solar panel 500.

[0274] Figure 24 demonstrates a solar tracking assembly 100 according to a further example of the present invention, where there is provided a vertical support member 102 supporting a solar panel support member 105, rotatably attached to the solar panel support member 105 is at least one solar panel 500. The rotatable attachment between the solar panel 500 and the solar panel support member 105 is achieved by at least one connection member 400. Further provided are additional solar panel support members 107 supporting the solar panel 500; these additional solar panel support members 107 form a prism-shaped cross section, attached to which is a weight 450. In this example, the weight 450 is in the form of one or more cylinders; however, it should be understood by a person of skill in the art that any form of weight or heavy material may be suitable. In this example, the vertical support member 102 attaches to the solar panel support member 105. The weight 450 provides a counterweight to the weight of the solar panel 500. In other words, the weight 450 biases the solar panel 500 about the solar panel support member 105 to a resting position when there is no, or little, wind present against the solar panel 500.

[0275] Figure 25 demonstrates a solar tracking assembly 100 according to a further example of the present invention, where there is provided a vertical support member 102 supporting a solar panel support member 105, rotatably attached to the solar panel support member 105 is at least one solar panel 500. The rotatable attachment between the solar panel 500 and the solar panel support member 105 is achieved by at least one connection member 400. Further provided are additional solar panel support members 107 supporting the solar panel 500; these additional solar panel support members 107 form a prism-shaped cross section, attached to which is a weight 450. In this example, the weight 450 is in the form of one or more cylinders; however, it should be understood by a person of skill in the art that any form of weight or heavy material may be suitable. In this example, the vertical support member 102 attaches to the solar panel support member 105 by a connection member 400. The weight 450 provides a counterweight to the weight of the solar panel 500. In other words, the weight 450 biases the solar panel 500 about the solar panel support member 105 to a resting position when there is no, or little, wind present against the solar panel 500.

[0276] Figure 26 illustrates a solar tracking assembly 100, highlighting the connection between the rotatable vertical support member 102 and the solar panel support member 105. The rotatable vertical support member 102 is designed to support the solar panel support member 105, allowing it to rotate and adjust the orientation of the attached solar panel 500. The connection between the vertical support member 102 and the solar panel support member 105 is crucial for enabling the rotational movement necessary to optimize the panel's position in response to wind forces. The solar panel 500 is attached to the solar panel support member 105 via connection members 400, which allow the solar panel 500 to rotate relative to the support member 105. This configuration ensures that the solar panel 500 can adjust its orientation to minimize wind resistance and optimize energy production, while the robust connection between the vertical support member 102 and the solar panel support member 105 provides stability and support for the entire system. Figure 27 illustrates a solar tracking assembly 100, demonstrating the effect of a strong wind Z coming from the right. The system comprises a rotatable vertical support member 102, a solar panel support member 105, and a solar panel 500 attached via a connection member 400. In this scenario, the strong wind Z exerts force on the solar panel 500, causing it to rotate upwards around the connection member 400. The connection member 400 allows the solar panel 500 to pivot relative to the solar panel support member 105, enabling the panel to adjust its orientation in response to the wind force. This rotation helps to minimize the wind resistance and reduce the stress on the solar panel and its mounting structure. The robust connection between the vertical support member 102 and the solar panel support member 105 ensures stability and support for the entire system, even under strong wind conditions.

[0277] In a further example of a solar tracking assembly according to the present invention, a resilient member may be added to the solar tracking assembly to hold the solar panel 500 in a rest position (i.e., a position where there is little or no wind). The resilient member may be a spring, elastic, string, rope, or the like. The resilient member may extend between the weight 450 and a solar panel support member 105, and in the rest position, the resilient member is in a substantially taut state. When force is exerted against the solar panel 500, in one direction, the resilient member changes to a substantially non-elastic state and allows movement of the solar panel 500; in the other direction, the resilient member stretches and extends in a taut state, biasing the solar panel 500 against the force, back to a rest position.

[0278] In a further example of the present invention, the weight 450 is placed at 90-120 degrees from the solar panel 500. At 90 degrees, the weight 450 has no effect when the panel 500 is at a horizontal position. When there are panels three times as wide as they are high, the optimal angle is approximately 110 degrees.

[0279] In a further improvement of the present invention, according to any of the examples herein described, a measurement device (not shown) may be used to drive rotation of a solar panel 500 via a motor 300. The measurement device may measure properties of wind or other external forces present; these properties include, but are not limited to, speed and direction. Upon the measurement device measuring a property at a predetermined threshold, the motor may be directed to rotate the solar panel 500 to a determined orientation. This determination of the required orientation and control of the motor may be performed by a computing unit (not shown). Ideally, the solar panel 500 is rotated such that the rear surface 500B faces the direction of wind force present, such that the wind force causes the solar panel 500 to rotate to a position such as that demonstrated in Figure 27, whereby the solar panel 500 is substantially parallel to the direction of the wind. In this manner, in the presence of high wind or other external forces, the solar panel 500 may rotate to a position whereby wind or other external force is permitted to pass by the solar panel

[0280] 500 with minimal resistance.

[0281] In all embodiments and examples of the present invention, the solar tracking assembly 100 may be secured to the ground via a ground anchor (not shown) or similar. The ground anchor may be attached to the vertical support member 102. The ground anchor may be in the form of a ground screw as would be understood by a person of skill in the art; preferably, the ground screw extends above ground by approximately 20-80 cm and contains a hollow aperture within which the vertical support member 102 may be placed. In this manner, the ground anchor is embedded within the ground, and the vertical support member 102 is secured within it. This provides a solid and stable foundation for the solar tracking assembly 100. Other forms of ground anchors would also be suitable as would be understood by a person of skill in the art, for example, multiple ground anchors 103 could be embedded in the ground at different angles, or alternative ballast or weight could be placed atop the ground in place of a ground anchor. The ground anchors, for example, may be in the form of a screw which, upon rotation into the ground, secures the ground anchor within the ground, or they may be in the form of a metal bar having a cross-section designed to provide strength, which may include a "C" cross-section, "H" cross-section, "I" cross-section, or the like, which may be driven into the ground by force. Further, the ground anchor may have a substantially flat bottom portion.

[0282] Further solar tracking assembly designs and improvements are shown in Figures 28 to 35. According to one aspect of the present invention as shown in FIG. 28, there is provided a wheel for rotating a solar panel, where the wheel is driven via a wire and second wheel attached to a motor.

[0283] A rotatable vertical member 102 supports a solar panel support member 105. The solar panel support member 105 holds one or more solar panels 500. A wheel 200 is positioned below the solar panel support member 105 and is connected to a wire 202. The wire 202 wraps around the wheel 200 and extends to a second wheel 200, which is connected to a motor 300. The motor 300 drives the second wheel 200, which in turn controls the tension and movement of the wire 202.

[0284] The wire 202, by being wrapped around the wheel 200, enables the rotation of the solar panel support member 105 and the attached solar panels 500 around the vertical axis of the rotatable vertical member 102. The motor 300, through the second wheel 200, provides the necessary force to adjust the orientation of the solar panels 500 by controlling the wire 202.

[0285] The configuration allows for the solar panels 500 to adapt to strong winds by rotating around the horizontal axis, thereby reducing the torque exerted on the panels. The wire 202 ensures that the panels can be rotated in any direction, providing flexibility in panel orientation. The placement of the wire 202 and the wheel 200 below the point of the solar panels 500 ensures that the panels can rotate freely without interference.

[0286] To further elaborate on the above, the solar panels 500 may rotate freely around the solar panel support member 105 which reduces force exerted on the panels 500 and thus the torque exerted on the vertical member 102. One possible method by which the solar panels 500 are rotated around the solar panel support member 105 is via wind force on the solar panels 500. The solar panels 500 may be rotatably connected to the solar panel support member 105 via a rotatable connection such as a hinge, or they may be fixedly connected to the solar panel support member 105 which may then rotate itself. Any form of rotation of the solar panels 500 around the solar panel support member 105 is contemplated and should be understood by a person of skill in the art.

[0287] The rotatable vertical member 102 is rotatable, typically via mechanical means but may be by any form of force. It may be directly rotated through connection to a motor, which may be achieved via gears or the like as would be understood by a person of skill in the art. One possible method for achieving rotation of the rotatable vertical member 102 is via a wheel 200 as is further described in this document. Utilising a wheel 200 to rotate the vertical member 102 allows the solar panels 500 to be fixed in a certain orientation to allow for reduced torque on the vertical member 102 in the presence of certain wind conditions. In an embodiment comprising multiple solar panel systems according to the present invention, each system comprises solar panels 500, a vertical member 102, and a solar panel support member 105, each system may have one or more wheels 200 to orient the solar panels 500 to reduce torque exerted on the respective vertical members 102. Driving of the wheels 200 to cause rotation of the vertical members 102 will be described further herein.

[0288] In an alternative embodiment, the solar panels are fixed in the horizontal direction but rotate around the rotatable vertical support member 102. This configuration allows for the solar panels 500 to adapt to strong winds by rotating around the vertical axis, thereby reducing the torque exerted on the panels. The wire 202 ensures that the panels can be rotated in any direction, providing flexibility in panel orientation. The placement of the wire 202 and the wheel 200 below the point of the solar panels 500 ensures that the panels can rotate freely without interference.

[0289] As shown in FIG. 35, the rotatable vertical member 102 rotates around axis A, while the solar panels 500 attached to the solar panel support member 105 rotate around axis B. Rotation of the rotatable vertical member 102 may be driven by a wheel 200 and wire 202 as herein described, or by any other suitable method, including direct rotation via the wire 202. Rotation of the solar panels 500 around axis B is driven by the wind. Additionally, the system can connect multiple wheels 200 to the same wire 202, allowing for the control of multiple solar panels 500 simultaneously. The motor 300 can control multiple wire loops, each loop managing one or more solar panels 500, enhancing the scalability and efficiency of the system.

[0290] According to a further aspect of the present invention as shown in FIG. 29, there is provided a locking member for locking the wire into place on the wheel.

[0291] The wheel 200 is a component in the wire-based rotational control system for wind-responsive solar panels. The wheel 200 facilitates the movement and positioning of the solar panels by interacting with the wire 202. The wire 202 wraps around the wheel 200, enabling the transmission of rotational force from a motor to the solar panels.

[0292] The wire 202 is secured onto the wheel 200 using a locking member 270. The locking member 270 ensures that the wire 202 remains fixed in position on the wheel 200, preventing any slippage that could disrupt the control and orientation of the solar panels. The locking member 270 engages with the wire 202 at a specific point, providing a secure attachment that maintains the wire's tension and alignment.

[0293] The locking member 270 is designed to hold the wire 202 firmly against the wheel 200, ensuring consistent and reliable operation of the solar tracking system. By locking the wire 202 in place, the locking member 270 allows the system to accurately control the rotation of the solar panels, even under varying wind conditions. This configuration enhances the stability and responsiveness of the solar tracking system, contributing to the overall efficiency and effectiveness of the solar tracking system.

[0294] According to another aspect of the present invention, the locking member for securing the wire onto the wheel can be implemented using a screw mechanism. This mechanism allows for the wheel to be adjustable even after the wire has been tightened, providing fine-tuning capabilities for the orientation of the solar panels.

[0295] In one embodiment, the screw mechanism includes screws that can be released slightly to allow the wheel to turn. When the screws are loosened, the wheel can be rotated, which in turn adjusts the orientation of the attached solar panels. The screws are positioned within grooves on the wheel, allowing for free movement within the groove when the screws are loosened.

[0296] If the fine-tuning provided by the groove is not sufficient, the wheel can be further adjusted by utilizing additional holes positioned along the groove. Once the wheel is turned all the way to the end of the groove, another hole becomes visible. The screw can then be unscrewed from its current position and inserted into the next hole, allowing for further adjustment of the wheel and the attached solar panels.

[0297] This screw mechanism is crucial for ensuring that all solar panels face the same direction after the wire has been tightened. During the initial installation and tightening of the wire, the panels may end up pointing in slightly different directions, leading to a loss of production. The ability to fine-tune the direction of the panels using the screw mechanism ensures optimal alignment and maximizes energy capture.

[0298] The locking member 270, as shown in FIG. 29, is designed to hold the wire 202 firmly against the wheel 200, ensuring consistent and reliable operation of the solar tracking system. By locking the wire 202 in place, the locking member 270 allows the system to accurately control the rotation of the solar panels, even under varying wind conditions. This configuration enhances the stability and responsiveness of the solar tracking system, contributing to the overall efficiency and effectiveness of the solar tracking system.

[0299] According to one aspect of the present invention, the wire makes at least one and a half full loops around the wheel. This configuration, as shown in FIG. 28, enables the rotation of the solar panel support member 105 and the attached solar panels 500 around the vertical axis of the rotatable vertical member 102. By making at least one and a half full loops (i.e., at least 540 degrees) around the wheel 200, the system ensures that the panels can be rotated in any direction, providing flexibility in panel orientation. This design allows the panels to be rotated an additional 90 degrees in either direction, regardless of their initial position, thereby enhancing the system's adaptability to varying wind conditions and sun positions.

[0300] Another aspect of the present invention involves the placement of the wire and wheel below the lowest point of the solar panels. As depicted in FIG. 28, the wire 202 and the wheel 200 are positioned below the solar panel support member 105, which holds the solar panels 500. This strategic placement ensures that the panels can rotate freely without interference from the wire 202. By positioning the wire and wheel below the panels, the system avoids potential obstructions and allows for smooth, unobstructed rotation, thereby enhancing the overall efficiency and reliability of the solar tracking system.

[0301] Additionally, the present invention includes at least one wire tensioner configured to adjust the tension of the wire. As shown in FIG. 29, the wire tensioners are integrated into the system to maintain optimal wire tension, which is crucial for precise control of the solar panel orientation. The wire tensioners allow for adjustments to be made to the wire's tension, ensuring that the wire remains taut and responsive to the motor's movements. In an embodiment containing two wire tensioners, by shortening one wire tensioner and lengthening the other, the direction of the panels can be adjusted between the wire tensioners, providing fine-tuned control over the panel orientation. This feature enhances the system's ability to adapt to varying wind conditions and ensures consistent, reliable operation.

[0302] According to a further aspect of the present invention as shown in FIG. 30, there is provided multiple arrays of solar panels 500, each array driven by a single wheel 200. The wheel 200 is driven by a wire 202 from a wheel 200 attached to a motor. The motor drives several wheels 200 using the wire 202, which each drive their own solar panel array 500.

[0303] The solar panels 500 are arranged in multiple arrays, each supported by a solar panel support member. The support members are connected to the wheel 200, which facilitates the rotation of the solar panels 500 around a vertical axis. The wire 202 wraps around the wheel 200, enabling the transmission of rotational force from the motor to the solar panels 500.

[0304] The wire 202 extends from the wheel 200 connected to the motor to the wheels 200 of each solar panel array 500. This configuration allows the motor to control the orientation of multiple solar panel arrays 500 simultaneously. The wire 202 ensures that the panels can be rotated in any direction, providing flexibility in panel orientation and allowing the system to adapt to varying wind conditions.

[0305] The placement of the wire 202 and the wheel 200 below the point of the solar panels 500 ensures that the panels can rotate freely without interference. The system can connect multiple wheels 200 to the same wire 202, allowing for the control of multiple solar panels 500 simultaneously. The motor can control multiple wire loops, each loop managing one or more solar panels 500, enhancing the scalability and efficiency of the system.

[0306] In various aspects of the present invention, the solar panel system may comprise or be in communication with a computing unit, also referred to as a controller, control system, or processing device. The computing unit may be configured to monitor environmental inputs, analyze system feedback, execute control logic, and issue commands to one or more act

[0307] As used herein, the term "module" may refer to one or more functional modules, each of which may be implemented as one or more hardware modules and / or one or more software modules and / or a combined software / hardware module in a node. In some examples, the module may represent function realized as software and / or hardware of the node.

[0308] As used herein, the term "computer program carrier", "program carrier", or "carrier", may refer to one of an electronic signal, an optical signal, a radio signal, and a computer readable medium. In some examples, the computer program carrier may exclude transitory, propagating signals, such as the electronic, optical and / or radio signal. Thus, in these examples, the computer program carrier may be a non-transitory carrier, such as a non-transitory computer readable medium.

[0309] As used herein, the term "processing module" may include one or more hardware modules, one or more software modules or a combination thereof. Any such module, be it a hardware, software or a combined hardware-software module, may be a determining means, estimating means, capturing means, associating means, comparing means, identification means, selecting means, receiving means, sending means or the like as disclosed herein. As an example, the expression "means" may be a module corresponding to the modules listed above in conjunction with the Figures.

[0310] As used herein, the term "software module" may refer to a software application, a Dynamic Link Library (DLL), a software component, a software module, a software object, a React component, an object according to Component Object Model (COM), a software function, a software engine, an executable binary software file or the like.

[0311] The terms "processing unit" or "processing circuit" may herein comprise one or more processors, an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or the like. The processing circuit or the like may comprise one or more processor kernels.

[0312] As used herein, the expression "configured to / for" can refer to that a processing circuit can be configured to, such as adapted to or operative to, by means of software configuration and / or hardware configuration, perform one or more of the actions described herein.

[0313] As used herein, the term "action" may refer to an action, a step, an operation, a response, a reaction, an activity or the like. It shall be noted that an action herein may be split into two or more subactions as applicable. Moreover, also as applicable, it shall be noted that two or more of the actions described herein may be merged into a single action.

[0314] As used herein, the term "memory" may refer to a hard disk, a magnetic storage medium, a portable computer diskette or disc, flash memory, random access memory (RAM) or the like. Furthermore, the term "memory" may refer to an internal register memory of a processor or the like.

[0315] As used herein, the term "computer readable medium" may be a Universal Serial Bus (USB) memory, a Digital Versatile Disc (DVD), a Blu-ray disc, a software module that is received as a stream of data, a Flash memory, a hard drive, a memory card, such as a MemoryStick, a Multimedia Card (MMC), Secure Digital (SD) card, etc. One or more of the aforementioned examples of computer readable medium may be provided as one or more computer program products. As used herein, the term "computer readable code units" may be text of a computer program, parts of or an entire binary file representing a computer program in a compiled format or anything there between.

[0316] As used herein, "computing unit", or "controller" refers broadly to any hardware and / or software combination configured to monitor environmental inputs, execute wind-responsive control logic, and issue actuation commands to the tracker drive. In particular, the computing unit may comprise one or more processors (e.g., microcontrollers, ASICs, FPGAs, or general-purpose CPUs), memory (volatile or non-volatile), and any input / output interfaces needed to communicate with sensors (anemometers, motor-current or wire-tension sensors, cameras, etc.) and actuators (motors, clutches, locks). The computing unit may be implemented as a single on-board controller per tracker, a distributed network of edge devices, a centralized field controller, or any hybrid arrangement, and includes all software or firmware modules (e.g., sampling, threshold-comparison, adaptationuators such as motors (300), release mechanisms, or other rotational drive components.

[0317] The computing unit may be:

[0318] On-site: Physically integrated into each solar panel structure (500), mounted to the vertical support member (102), or housed in a local control enclosure (e.g., within or adjacent to an inverter or junction box).

[0319] Distributed: A system of interconnected microcontrollers, edge processors, or embedded computing devices deployed across multiple solar panel structures, each responsible for local control and coordination.

[0320] Centralized: A central computing unit located within a site control cabinet or operator building, managing communication and control for a plurality of solar panel structures across the installation.

[0321] Remote: A cloud-based or off-site computing system communicatively linked to the solar panel system via a network connection, such as cellular, satellite, or internet-based protocols. In such embodiments, local controllers may act as gateways or edge processors, relaying data between the solar panel system and the remote computing unit.

[0322] It will be appreciated that the wind-responsive control logic described herein, including the detection of wind conditions, estimation of azimuth, and automated rotation of the tracker assemblies to a wind-stable orientation, may be implemented by any of the disclosed computing unit configurations. Whether the computing unit is embodied as an on-site controller integrated with the tracker assembly, a distributed network of edge processors, a centralized site controller, or a remote / cloud- based system communicatively linked to the trackers, each configuration is capable of executing the wind-responsive orientation logic in real-time. The specific computing architecture may be selected based on site requirements, available infrastructure, or operational preferences, without departing from the scope of the present invention. In various aspects of the invention, the computing unit is configured to implement failsafe behavior in the event of missing, invalid, or inconsistent sensor input. For example, if a wind-speed sensor fails to report data within a predefined interval, or if the reported value remains excessively high for an implausibly long duration, the controller may switch to a secondary input source— such as a site-level weather forecast retrieved via network connection or a predictive model based on historical site conditions. If no valid wind data is available from either real-time or predicted sources, the system defaults to a conservative wind-protection mode, placing the solar panel structure into a known safe orientation that minimizes aerodynamic loading. This logic ensures that structural safety is preserved even during sensor fault, communication failure, or data corruption, and may be applied in conjunction with any of the control routines described elsewhere in this disclosure. In one embodiment, a hybrid architecture may be employed, wherein local control logic is executed on a low-power on-site microcontroller, while high-level decision making (e.g., optimization strategies, weather model integration) is performed remotely and transmitted back to the system.

[0323] The computing unit may communicate with the motor (300), sensors, and other system components through a variety of interfaces, including but not limited to: * Wired connections: Serial bus (e.g., RS-485), Ethernet, Modbus, CAN bus, or direct GPIO / relay triggering.

[0324] • Wireless connections: Wi-Fi, LoRa, ZigBee, Bluetooth Low Energy (BLE), 4G / 5G cellular, satellite uplinks, or proprietary RF systems.

[0325] • Powerline communication (PLC): Signals transmitted over existing power cabling infrastructure. In one embodiment, each motor (300) or actuator may include an embedded controller or motor driver that receives commands via serial or digital signal from the computing unit. In another embodiment, the computing unit may use PWM, analog voltage, or current-based commands to drive actuation directly.

[0326] The computing unit may receive data from one or more sensors as previously described, including, but not limited to, environmental sensors such as wind speed and direction (anemometers), temperature, irradiance, precipitation (rain / hail, snow), humidity, electrical sensors such as voltage, current, or power sensors positioned at the inverter or module level, and mechanical sensors such as tension sensors in the wire (202), rotation encoders on the vertical support member (102) or solar panel support member (105), torque sensors integrated into the drive shaft or drivetrain, orlimit switches, accelerometers, or gyroscopes.

[0327] In some embodiments, the computing unit is configured to log sensor data, execute predictive algorithms (e.g., based on weather forecasts or historical performance), and adjust panel orientation or system behaviour accordingly.

[0328] The computing unit may execute one or more control algorithms, including position control algorithms which calculate and control optimal rotational position for energy production or wind safety, stow mode algorithms which are based on wind or hail, snow thresholds, tension values, or time-of-day triggers, feedback loops which adjust position in real time based on energy output, and error detection and diagnostic algorithms which identify overcurrent conditions, motor stalls, communication loss, or mechanical resistance anomalies.

[0329] In some embodiments, the computing unit may include a real-time clock (RTC), enabling time-based operations such as nighttime repositioning (e.g., wildlife-safe stow mode) or time-of-day sun tracking.

[0330] The computing unit may further include user interface components, such as a touchscreen, status LEDs, or keypad, remote access capability, allowing operators to view, control, or configure the system via a web portal, mobile app, or SCADA interface, and APIs or integration layers for interoperability with energy management systems (EMS), building automation systems, or utility dashboards.

[0331] In some embodiments, the computing unit may be configured to receive software updates over-the- air (OTA), and may store configuration profiles for various operational modes (e.g., wind-sensitive vs. energy-maximizing behavior).

[0332] With reference to Figure 14, there is depicted an example control-computing subsystem 1500 of the present solar-tracker controller. The subsystem 1500 comprises a Processing Module 1501, a Memory 1502, an I / O Module 1506, and is supplied with executable instructions via a Computer Program 1503 delivered by a Program Carrier 1505. Processing Module 1501 may be embodied as one or more hardware circuits— such as a microprocessor, ASIC, or FPGA— or as a combination of hardware and software, and is responsible for executing the wind-responsive orientation routines, perturb-and-observe power-optimization loop, and environmental-priority stow logic. Memory 1502 is a non-transitory computer-readable medium that stores Computer Program 1503 along with configuration parameters— such as load-to-angle calibration maps, wind-load thresholds TT and Tz, perturbation step sizes, and mode parameters— and Program Carrier 1505 provides those instructions into Memory 1502 via any suitable medium (for example, a flash memory device or firmware image). I / O Module 1506 facilitates bidirectional communication with field hardware and supervisory systems: it acquires raw inputs such as motor-current samples, cable-tension signals, inverter power telemetry, optional anemometer readings, and remote weather forecasts, and forwards them to subordinate functional modules within Processing Module 1501 while also transmitting computed load indices, azimuth estimates, and actuation commands back to the drive mechanism and SCADA interface. Within Processing Module 1501, the Sensor Input Module 1510 acquires and buffers all incoming sensor and forecast data via I / O Module 1506; the Load and Azimuth Estimation Module 1520 computes a unified wind-load index W and maps live load measurements to an azimuth estimate using the stored calibration surface; the Control Logic Interface Module 1530 packages those indices, headings, and mode flags for delivery to the wind- stow and power-optimization routines; and the Fallback and Calibration Module 1540 monitors the validity and consistency of all environmental inputs— invoking conservative, edge-on stow positions upon detecting missing, saturated, or implausible values and scheduling automatic calibration sweeps when rolling energy-yield drift exceeds predefined thresholds— thereby ensuring uninterrupted, sensorless tracking, optimized power capture, and reliable wind protection without per-row encoders or additional environmental sensors.

[0333] The foregoing description of the embodiments has been presented for purposes of illustration and description only, and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. It will be appreciated by those of ordinary skill that many modifications and variations are possible in light of the above teaching. For example, any feature described in connection with one embodiment may be combined with, substituted for, or omitted in another embodiment, without departing from the scope of the invention. Likewise, any of the thresholds, hysteresis margins, zone classifications, drive-loop architectures, adaptive weighting factors, diagnostic routines, or nonlimiting examples of mechanical hinge arrangements set forth herein may be employed alone or in any sub-combination.

[0334] Moreover, although particular numerical values, angular offsets, sampling intervals, and component arrangements have been recited for clarity, such parameters may be adjusted by a person of ordinary skill to suit site-specific conditions, module geometries, wind regimes, or control preferences. The invention is therefore intended to embrace all such alternatives, modifications, and combinations as fall within the spirit and broad scope of the appended claims.

Claims

CLAIMS1. A solar tracking assembly, comprising: a rotatable support member (102) configured to carry at least one solar panel (500) and rotate about a vertical axis; a horizontal member (105) connecting the at least one solar panel (500) to the rotatable support member (102), wherein the horizontal member (105) permits the at least one solar panel (500) to passively pivot about a horizontal axis in response to wind; a drive mechanism (300, 202, 200) operably coupled to the rotatable support member (102) to effect controlled rotation of the solar tracking assembly about the vertical axis, the drive mechanism (300, 202, 200) controlled by a computing unit; the computing unit configured to execute wind responsive orientation logic, wherein: the computing unit determines current or predicted wind speed and wind direction, and causes the drive mechanism (300, 202, 200) to rotate the rotatable support structure (102) such that the rear side (500B) of the solar panel (500) faces into the direction of incoming wind, enabling the at least one solar panel (500) to tilt upward about the horizontal member (105) in response to wind.

2. The assembly of claim 1, wherein the computing unit configured to receive input from a wind sensor and determine wind direction and wind speed prior to executing the wind responsive orientation logic.

3. The assembly of claim 1, further comprising a load-feedback sensor (204) configured to measure a mechanical load associated with the drive mechanism (300, 202, 204).

4. The assembly of claim 3, wherein the computing unit is further configured to receive an output from the load-feedback sensor (204) and estimate a current azimuthal position of the at least one solar panel (500).

5. The assembly of claim 1, further comprising a current-sensing device for determining electrical current flowing through the drive mechanism (300).

6. The assembly of claim 5, wherein the computer unit is further configured to receive electrical current information from the current-sensing device and utilizing said current information to estimate the current azimuthal position of the at least one solar panel (500).

7. The assembly of claim 4, wherein the computing unit further comprises calibration data mapping mechanical load to rotational orientation of the at least one solar panel (500).

8. The assembly of claim 1, further comprising a power monitoring device configured to measure instantaneous electrical power output of the at least one solar panel (500).

9. The assembly of claim 8, wherein the computing unit controls the drive mechanism (300, 202, 200) to perturb the rotational orientation of the rotatable support member (102) and is further configured to compare output from the power monitoring device to determine whether to further perturb the rotational orientation of the rotatable support member (102).

10. The assembly of claim 1, wherein the computing unit controls the drive mechanism (300, 202, 200) to rotate the rotatable support member (102) to a rotatable orientation, such that the computing unit is further configured to determine the rotatable orientation using wind responsive information including the geometry of the at least one solar panel (500), the horizontal member (105), and a predetermined wind threshold.

11. The assembly of claim 1, wherein under wind loading with the rear surface (500B) oriented into the wind, the at least one solar panel generates an aerodynamic moment about the horizontal hinge axis sufficient to overcome the hinge bias and cause the panel to pivot upward.

12. The assembly of claim 2, wherein the computing unit is configured to determine wind direction by identifying the rotational orientation of the rotatable support member (102) at which drive load is minimized or maximized during controlled panel rotation.

13. The assembly of claim 1, wherein the computing unit is further configured to determine wind speed based on the rate of change in drive load of the drive mechanism (300, 202, 200) with respect to azimuthal displacement of the at least one solar panel (500) from the wind direction.

14. The assembly of claim 1, wherein the computing unit is further configured to disable solar tracking and maintain a wind-stable orientation when the wind speed exceeds a survivability threshold.

15. The assembly of claim 1, further comprising a power-monitoring device includes including a maximum power point tracking (MPPT) subsystem that communicates with the computing unit to assess solar panel output.

16. The assembly of claim 1, wherein the load-feedback sensor comprises a motor current sensor integrated into the drive mechanism (300).

17. The assembly of claim 1, further comprising a torque sensor integrated into the drive mechanism (300).

18. The assembly of claim 7, wherein the computing unit continuously updates the calibration data based on historical comparisons between measured power output and mechanical load estimates.

19. The assembly of claim 1, wherein the drive mechanism comprises a motor (300) coupled to the rotatable support member (102) via a tensioned cable loop (202) and pulley arrangement (200).

20. The assembly of claim 1, wherein the computing unit prioritizes placing the at least one solar panel (500) in a wind responsive orientation over a power optimization orientation when wind speed exceeds a predefined switching threshold.

21. The assembly of claim 1, further comprising a damping mechanism associated with the horizontal hinge member (105) to reduce oscillations during high wind conditions.

22. The assembly of claim 19, wherein the tensioned cable loop (202) comprises opposing first and second cable branches, and further comprises a load-feedback sensor which comprises respective tension sensors configured to measure cable tension in each branch, the computing unit being further configured to detect misalignment or crosswind loading based on a difference between the measured tensions.

23. The assembly of claim 1, wherein the computing unit further includes calibration data defining a mapping between a mechanical load parameter and a corresponding rotational orientation of the at least one solar panel (500).

24. The assembly of claim 1, wherein the computing unit is further configured to suspend control of the drive mechanism (300, 200, 202) in response to detection of at least one of a wind-load index exceeding a safety threshold, a wildlife-safe orientation, or a hail, snow-loading signal, and to resume controlling the drive mechanism (300, 200, 202) when the condition has cleared.

25. The solar-tracking assembly of claim 19, wherein the drive mechanism further comprises:(a) a motor-current sensor operatively coupled to the drive mechanism (300), and(b) a mechanical load-feedback sensor (204) configured to measure cable tension in the tensioned cable loop (202), wherein the computing unit is further configured to(i) compare contemporaneous readings from the load-feedback sensor (204) and the motorcurrent sensor,(ii) detect that one sensor is unreliable when its reading deviates from the other sensor by more than a predefined consistency threshold for a predefined time window, and(iii) thereafter substitute the remaining reliable sensor for estimating panel azimuth and for executing wind responsive orientation logic until the unreliable sensor is restored to within the threshold.

26. The solar-tracking assembly of claim 7, wherein the computing unit is further configured to(a) determine a rolling average energy-yield metric over a first predefined time window,(b) compare the rolling average energy-yield metric with a historical baseline corrected for irradiance, and(c) perform a calibration sweep when the deviation between the rolling average energy-yield metric and the historical baseline exceeds a predetermined drift threshold.

27. The solar-tracking assembly of claim 19, further comprising a break-away clutch incorporated into the tensioned cable loop (202), the break-away clutch being configured to disengage the drive motor (300) from the rotatable support member (102) when cable tension exceeds a preset overload threshold, thereby protecting the drive train and the solar panel structure from extreme wind or mechanical jam conditions.

28. The assembly of claim 19, further comprising a wind sensor, a motor-current sensor operatively coupled to the drive mechanism (300), and a mechanical load-feedback sensor (204) configuredto measure cable tension in the tensioned cable loop (202), wherein the computing unit is additionally configured to:(a) compare a wind-direction value reported by the wind sensor with a mechanically implied wind direction derived from at least one of motor current output by the motor-current sensor and cable tension data output by the load-feedback sensor (204);(b) identify a persistent bias when the reported wind direction differs from the mechanically implied wind direction by more than a configurable angular tolerance for a predefined duration; and(c) upon identifying the persistent bias, automatically apply a correction factor to the winddirection value and / or transmit an alert indicating that the wind sensor requires inspection or recalibration.

29. A solar-tracker plant, comprising:(a) a plurality of solar-tracker assemblies (100) according to any preceding claim, each assembly having a centre of rotation defined by a rotatable support member (102); and(b) a ground layout in which the centres of rotation are placed at the corner points of a repeating grid made up entirely of equal-sized regular hexagons, the grid covering the site without gaps or overlaps, whereby(i) every solar-tracker assembly (100) is the same distance from each of its six nearest neighbours;(ii) the circular 360-degree rotation envelopes of adjacent solar-panel structures (500) touch but do not overlap; and(iii) the layout naturally supplies a set of row-direction families that are spaced 60 degrees apart in azimuth.

30. The plant of claim 29, wherein a computing unit assigns each solar-tracker assembly (100) to one of the row-direction families and is configured to switch all assemblies simultaneously from one family to another in response to at least one of:(a) detection of inter-row shading;(b) a predetermined low-sun-altitude clock event; or(c) activation of a wildlife-safe, hail, snow-mitigation, or wind-stow mode.

31. The plant of claim 29 or 30, wherein at least one tensioned drive cable (202) that transmits rotational motion between solar-tracker assemblies (100) is routed along a straight edge of the hexagonal grid so that successive cable spans are substantially equal in length.

32. The plant of any one of claims 29-31, wherein a site-design algorithm stored in the computing unit receives a land-boundary outline and automatically fills the interior of the outline with the regular-hexagon grid, trimming partial hexagons at the boundary to maximise ground-coverage ratio while preserving non-overlapping rotation envelopes.

33. The plant of any one of claims 29-32, wherein, during construction, the position of each new solar-tracker assembly (100) is established by measuring a single repeated centre-to-centre distance from two already-installed assemblies, thereby preventing cumulative angular error across the array.

34. The plant of any one of claims 29-33, wherein the computing unit automatically identifies perimeter solar-tracker assemblies (100) from the neighbour geometry of the hexagonal grid and applies lower wind-stow thresholds to those perimeter assemblies in accordance with the environmental-priority routine of Figure 10.

35. A solar-tracker plant, comprising:(a) a plurality of solar-tracker assemblies (100) according to any preceding claim, each assembly having a centre of rotation defined by a rotatable support member (102); and(b) a ground layout in which the centres of rotation are placed at the corner points of a repeating grid made up entirely of equal-sized equilateral triangles, the grid covering the site without gaps or overlaps, whereby(i) every solar-tracker assembly (100) is the same distance from each of its three nearest neighbours;(ii) the circular 360-degree rotation envelopes of adjacent solar-panel structures (500) touch but do not overlap; and(iii) the layout naturally supplies a set of row-direction families that are spaced 60 degrees apart in azimuth.

36. The plant of claim 35, wherein a computing unit assigns each solar-tracker assembly (100) to one of the row-direction families and is configured to switch all assemblies simultaneously from one family to another in response to at least one of:(a) detection of inter-row shading;(b) a predetermined low-sun-altitude clock event; or(c) activation of a wildlife-safe, hail, snow-mitigation, or wind-stow mode.

37. The plant of claim 35 or 36, wherein at least one tensioned drive cable (202) that transmits rotational motion between solar-tracker assemblies (100) is routed along a straight edge of the triangular grid so that successive cable spans are substantially equal in length.

38. The plant of any one of claims 35-37, wherein a site-design algorithm stored in the computing unit receives a land-boundary outline and automatically fills the interior of the outline with the equilateral-triangle grid, trimming partial triangles at the boundary to maximise groundcoverage ratio while preserving non-overlapping rotation envelopes.

39. The plant of any one of claims 35-38, wherein, during construction, the position of each new solar-tracker assembly (100) is established by measuring a single repeated centre-to-centre distance from two already-installed assemblies, thereby preventing cumulative angular error across the array.

40. The plant of any one of claims 35-39, wherein the computing unit automatically identifies perimeter solar-tracker assemblies (100) from the neighbour geometry of the triangular grid and applies lower wind-stow thresholds to those perimeter assemblies in accordance with the environmental-priority routine of Figure 10.

41. A method of operating a solar-tracking assembly, the method comprising:(a) providing a rotatable support member (102) carrying at least one solar panel (500) and a horizontal member (105) that permits the panel to passively pivot about a horizontal axis under wind load;(b) commanding, via a computing unit, a drive mechanism (300, 202, 200) coupled to the support member (102) to effect rotation about a vertical axis; and(c) determining, with the computing unit, current or predicted wind speed and direction and, in response, rotating the support member (102) so that the rear side (500B) of the solar panel(500) faces into the wind, thereby enabling the panel to tilt upward about the horizontal member (105).

42. The method of claim 41, further comprising receiving wind-speed and wind-direction data from a wind sensor prior to rotating the support member (102).

43. The method of claim 41, further comprising measuring a mechanical load associated with the drive mechanism (300, 202, 200) using a load-feedback sensor (204).

44. The method of claim 43, further comprising estimating a current azimuthal position of the solar panel (500) from the measured mechanical load.

45. The method of claim 41, further comprising measuring electrical current flowing through the drive mechanism (300) with a current-sensing device.

46. The method of claim 45, further comprising estimating the current azimuthal position of the solar panel (500) from the measured electrical current.

47. The method of claim 44, further comprising storing calibration data that map mechanical load to rotational orientation of the solar panel (500) and using those data when estimating azimuth.

48. The method of claim 41, further comprising measuring instantaneous electrical power output of the solar panel (500) with a power-monitoring device.

49. The method of claim 48, further comprising:(a) perturbing the azimuthal orientation of the support member (102) by a small increment;(b) comparing successive power measurements; and(c) continuing, reversing, or holding the perturbation based on whether power increased, decreased, or remained unchanged.

50. The method of claim 41, further comprising determining a safe wind-stow orientation from the panel geometry, hinge properties, and a predetermined wind threshold, and rotating the support member (102) to that orientation.

51. The method of claim 41, wherein rotating the support member comprises generating an aerodynamic moment about the horizontal axis sufficient to overcome hinge bias and cause the panel to pivot upward under wind load.

52. The method of claim 42, further comprising verifying wind direction by identifying the azimuth at which measured drive load is minimized or maximized during controlled rotation.

53. The method of claim 41, further comprising determining wind speed from the rate of change of drive load with respect to incremental azimuthal displacement of the solar panel (500).

54. The method of claim 41, further comprising disabling normal sun-tracking and maintaining the panel in a wind-stable orientation when wind speed exceeds a survivability threshold.

55. The method of claim 41, further comprising communicating with an MPPT subsystem to assess solar-panel output as part of power-monitoring.

56. The method of claim 41, wherein measuring mechanical load comprises sensing motor current in the drive mechanism (300).

57. The method of claim 41, further comprising sensing drive-shaft torque with a torque sensor integrated into the drive mechanism (300).

58. The method of claim 47, further comprising continuously updating the calibration data based on historical comparisons between measured power output and mechanical-load estimates.

59. The method of claim 41, wherein the drive mechanism comprises a motor (300) coupled to the support member (102) via a tensioned cable loop (202) and pulley arrangement (200).

60. The method of claim 41, further comprising giving priority to wind-responsive orientation over power-optimization when wind speed exceeds a predefined switching threshold.

61. The method of claim 41, further comprising damping oscillations of the panel about the horizontal axis with a damping mechanism.

62. The method of claim 59, further comprising measuring tension in opposing branches of the cable loop (202), comparing the measured tensions, and detecting misalignment or cross-wind loading from any difference.

63. The method of claim 41, further comprising storing, in non-volatile memory, a mapping between a mechanical-load parameter and a corresponding panel orientation, and using that mapping during operation.

64. The method of claim 41, further comprising suspending control of the drive mechanism (300, 202, 200) upon detection of at least one of: (a) a wind-load index exceeding a safety threshold;(b) a wildlife-activity signal; or (c) a hail- or snow-loading signal, and resuming control when the condition clears.

65. The method of claim 59, further comprising:(a) obtaining contemporaneous motor-current and cable-tension readings;(b) detecting that one sensor is unreliable when its reading deviates from the other by more than a consistency threshold for a predefined time window; and(c) thereafter using only the remaining reliable sensor for azimuth estimation and wind- responsive control until reliability is restored.

66. The method of claim 47, further comprising:(a) determining a rolling average energy-yield metric over a predefined time window;(b) comparing that metric to an irradiance-corrected historical baseline; and(c) performing a calibration sweep when the deviation exceeds a drift threshold.

67. The method of claim 59, further comprising disengaging the motor (300) from the support member (102) with a break-away clutch when cable tension exceeds an overload threshold to protect the drive train.

68. The method of claim 59, further comprising:(a) comparing a wind-direction value from a wind sensor with a mechanically implied wind direction derived from at least one of motor-current and cable-tension data;(b) identifying a persistent bias when the two differ by more than a configurable angular tolerance for a predefined duration; and(c) upon identifying the bias, automatically applying a correction factor to the wind-direction value and / or issuing an alert for sensor inspection.

69. A computer-implemented method of operating a solar tracking assembly, the method comprising: executing, by a computing unit, control logic to cause a rotatable support member to carry at least one solar panel and rotate about a vertical axis; wherein a horizontal member connects the at least one solar panel to the rotatable support member and permits the at least one solar panel to passively pivot about a horizontal axis in response to wind; the method further comprising: determining, with the computing unit, current or predicted wind speed and wind direction; and rotating, with the computing unit, the rotatable support member such that a rear side of the solar panel faces into the direction of incoming wind, thereby enabling the at least one solar panel to tilt upward about the horizontal member in response to wind.

70. The computer-implemented method of claim 69, further comprising: receiving, with the computing unit, wind direction and wind speed data from a wind sensor; and determining, with the computing unit, the wind direction and wind speed prior to rotating the rotatable support member.

71. The computer-implemented method of claim 69, further comprising: measuring, with a load-feedback sensor, a mechanical load associated with a drive mechanism operably coupled to the rotatable support member; and providing the measured mechanical load to the computing unit.

72. The computer-implemented method of claim 71, further comprising: estimating, with the computing unit, a current azimuthal position of the at least one solar panel based on the measured mechanical load.

73. The computer-implemented method of claim 69, further comprising: measuring, with a current-sensing device, electrical current flowing through the drive mechanism; and providing the measured electrical current to the computing unit.

74. The computer-implemented method of claim 73, further comprising: estimating, with the computing unit, the current azimuthal position of the at least one solar panel based on the measured electrical current.

75. The computer-implemented method of claim 72, further comprising: accessing, with the computing unit, calibration data defining a mapping between mechanical load and rotational orientation of the at least one solar panel; and applying the calibration data to the estimation of azimuthal position.

76. The computer-implemented method of claim 69, further comprising: measuring, with a power monitoring device, instantaneous electrical power output of the at least one solar panel; and providing the power measurement to the computing unit.

77. The computer-implemented method of claim 76, further comprising: perturbing, with the computing unit, the rotational orientation of the rotatable support member; comparing, with the computing unit, successive power measurements; and determining, with the computing unit, whether to further perturb the rotational orientation based on the comparison.

78. The computer-implemented method of claim 69, further comprising: determining, with the computing unit, a rotatable orientation of the rotatable support member using wind responsive information, the information including at least geometry of the solar panel, geometry of the horizontal member, and a predetermined wind threshold;and rotating, with the computing unit, the rotatable support member to the determined rotatable orientation.

79. The computer-implemented method of claim 69, further comprising: generating, under wind loading, an aerodynamic moment about the horizontal axis of the solar panel sufficient to overcome a hinge bias and cause the panel to pivot upward.

80. The computer-implemented method of claim 70, further comprising: identifying, with the computing unit, a rotational orientation of the rotatable support member at which drive load is minimized or maximized during controlled panel rotation; and determining, with the computing unit, wind direction based on the identified orientation.

81. The computer-implemented method of claim 69, further comprising: determining, with the computing unit, wind speed based on a rate of change in drive load of the drive mechanism with respect to azimuthal displacement of the at least one solar panel from the wind direction.

82. The computer-implemented method of claim 69, further comprising: disabling, with the computing unit, solar tracking; and maintaining, with the computing unit, a wind-stable orientation when wind speed exceeds a survivability threshold.

83. The computer-implemented method of claim 69, further comprising: communicating, with the computing unit, with a maximum power point tracking subsystem to assess solar panel output.

84. The computer-implemented method of claim 69, further comprising: measuring, with a motor current sensor integrated into the drive mechanism, mechanical load associated with the drive mechanism; and providing the measured mechanical load to the computing unit.

85. The computer-implemented method of claim 69, further comprising: measuring, with a torque sensor integrated into the drive mechanism, drive shaft torque; and providing the measured torque to the computing unit.

86. The computer-implemented method of claim 75, further comprising: continuously updating, with the computing unit, the calibration data based on historical comparisons between measured power output and mechanical load estimates.

Images