A solar light and heat reflecting mirror light-pursuing control method and system

Abstract

The application discloses a solar light-heat reflector light-pursuing control method and system, relates to the technical field of solar light-heat reflector light-pursuing control, and comprises the following steps: acquiring various basic data; generating a theoretical light-pursuing pose of a reflector; collecting a receiver area image to determine an actual light spot position and generate light spot offset data; generating a reflector individual error fingerprint representing a two-axis driving return difference and historical light spot offset characteristics; generating azimuth angle compensation data and elevation angle compensation data based on the light spot offset data and the reflector individual error fingerprint; generating a light-pursuing control instruction based on the theoretical light-pursuing pose and the compensation data and executing a light-pursuing action. The application realizes the closed-loop combination of theoretical light-pursuing and light spot feedback, makes the compensation process consider the long-term offset law related to the driving commutation return difference and the attitude interval through the construction of the reflector individual error fingerprint, improves the compensation pertinence, and reduces the influence of commutation hysteresis and individual differences on light-pursuing accuracy.

Description

Technical Field

[0001] This invention relates to the field of solar thermal reflector tracking control technology, specifically a solar thermal reflector tracking control method and system. Background Technology

[0002] Solar thermal power generation uses mirrors to concentrate solar radiation energy to a receiver, and utilizes the heat absorbed by the receiver to heat the working fluid and perform subsequent energy conversion. As tower solar thermal power generation systems develop towards large-scale mirror fields, high concentration ratios, and high operating temperatures, mirror tracking control technology has gradually evolved from open-loop control based on solar position calculations to closed-loop control combining encoder feedback, image detection, and online correction. Existing tracking systems typically calculate the solar incidence direction based on time, latitude, longitude, and solar altitude angle, then determine the theoretical tracking attitude based on the spatial relationship between the mirror and the receiver, and adjust the mirror attitude through azimuth and pitch axis drive mechanisms to project the reflected light spot onto the target area of ​​the receiver.

[0003] However, existing solar thermal reflector tracking control methods typically calculate the theoretical tracking attitude based on solar position models and ideal geometric relationships, making it difficult to fully account for transmission backlash, direction switching hysteresis, installation deviations, structural deformations, and individual differences caused by long-term operation in the reflector drive mechanism. While some methods detect the actual spot position using receiver images and perform closed-loop correction based on spot offset, they usually only generate compensation based on the current spot offset, failing to consider the drive shaft direction switching status, position feedback, motor current response, and historical spot offset patterns to distinguish between instantaneous disturbances, drive hysteresis, and attitude-related systematic deviations. When the drive shaft reverses direction, the motor has already generated driving force, but the reflector's actual attitude has not yet changed effectively, easily leading to compensation lag, spot oscillation, or repeated corrections. Furthermore, different reflectors exhibit different error characteristics due to manufacturing errors, installation conditions, and transmission wear; using uniform compensation parameters is insufficient to adapt to the error variations of each reflector in different attitude ranges.

[0004] Therefore, existing technologies struggle to construct individualized error representations that simultaneously reflect dual-axis drive backlash and historical spot offset characteristics. They also struggle to achieve synergistic compensation between theoretical tracking pose, current spot offset, and long-term individual errors, thereby affecting the positioning accuracy of the reflected spot, tracking stability, and receiver light uniformity. Summary of the Invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by this invention is that existing solar thermal reflector tracking control methods rely on theoretical tracking models and are difficult to reflect actual spot offsets, have difficulty identifying dual-axis drive hysteresis generated during drive axis reversal, cannot characterize the long-term individual error characteristics of different reflectors in different attitude ranges, and have problems in how to combine theoretical tracking pose, current spot offset, dual-axis drive feedback and historical spot offset patterns to generate individualized compensation data and achieve closed-loop tracking control.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a method for controlling the light tracking of a solar thermal reflector, comprising the following steps: S1: Acquire solar position data, receiver target position data, reflector current attitude data, dual-axis drive feedback data, drive direction change data, and historical spot offset data; S2: Based on the solar position data, the receiver target position data, and the reflector current attitude data, generate the theoretical light-tracking pose of the reflector; S3: Acquire image data of the receiver area, determine the actual spot position corresponding to the reflector based on the image data of the receiver area, and generate spot offset data based on the actual spot position and the receiver target position data; S4: Based on the dual-axis drive feedback data, the drive direction change data, and the historical spot offset data, generate a mirror individual error fingerprint characterizing the dual-axis drive backlash and historical spot offset features of the mirror; S5: Based on the spot offset data and the individual error fingerprint of the reflector, generate azimuth compensation data and pitch compensation data; S6: Based on the theoretical light-tracking pose of the reflector, the azimuth compensation data, and the pitch compensation data, generate a light-tracking control command for the reflector, and control the reflector to perform the light-tracking action according to the light-tracking control command.

[0008] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S1 specifically includes: Acquire current time data and mirror field geographical location data, and calculate solar position data based on the current time data and the mirror field geographical location data; Acquire receiver target location data, which includes the three-dimensional spatial coordinates of the receiver target point in the mirror field coordinate system and the two-dimensional coordinates of the preset target center in the receiver heat absorption area in the receiver surface coordinate system; Within each control cycle, dual-axis drive feedback data is acquired according to a preset sampling interval. The dual-axis drive feedback data includes azimuth axis actual position feedback, pitch axis actual position feedback, azimuth motor current data, and pitch motor current data. Based on the actual position feedback of the azimuth axis, the actual position feedback of the pitch axis, and the preset attitude calibration parameters, the current attitude data of the reflector is generated, which includes the current azimuth angle and the current pitch angle of the reflector. Based on the azimuth axis target position command and pitch axis target position command in the already issued and stored reflector tracking control command, the drive direction change data is obtained. The drive direction change data includes azimuth axis direction switching state, pitch axis direction switching state, drive direction after azimuth axis reversal and drive direction after pitch axis reversal. Acquire historical spot offset data up to the previous control cycle. The historical spot offset data includes effective historical spot offset records and attitude interval identifiers and control cycle identifiers corresponding to each effective historical spot offset record. The effective historical spot offset records include historical horizontal offset components and historical vertical offset components.

[0009] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S2 specifically includes: Construct a solar incidence direction vector based on the solar position data; Based on the three-dimensional spatial coordinates of the receiver target point in the mirror field coordinate system, a target reflection direction vector is constructed in the local coordinate system of the reflector. Based on the solar incident direction vector and the target reflection direction vector, the theoretical normal vector of the reflector is calculated according to the law of reflection. The theoretical normal vector of the reflector is converted into the theoretical azimuth and theoretical elevation angles of the reflector. Based on the current attitude data of the reflector, angular continuity processing is performed on the theoretical azimuth angle of the reflector, and the theoretical azimuth angle and the theoretical pitch angle of the reflector are limited to the allowable movement range of the corresponding drive shaft. The processed theoretical azimuth and theoretical elevation angles of the reflector are used as the theoretical ray-tracking pose of the reflector.

[0010] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S3 specifically includes: Image data of the receiver area is acquired by an image acquisition device deployed on the receiver side; The image data of the receiver area is processed for spot recognition to extract the spot outline corresponding to the current reflector and determine the center pixel coordinates of the spot; Based on the receiver image coordinate calibration relationship, the center pixel coordinates of the light spot are converted into the actual light spot position in the receiver surface coordinate system; The two-dimensional coordinates of the preset target center in the receiver surface coordinate system are used as the target spot position; Based on the actual spot position and the target spot position, spot offset data containing horizontal and vertical offset components is generated.

[0011] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S4 specifically includes: Based on the actual azimuth position feedback before the azimuth target position command of the previous control cycle is executed and the actual azimuth position feedback after the azimuth target position command is executed, azimuth position change data is generated. Based on the pitch axis actual position feedback before the pitch axis target position command of the previous control cycle is executed and the pitch axis actual position feedback after the pitch axis target position command is executed, pitch axis position change data is generated. Azimuth axis current variation data is generated based on the azimuth motor current data within adjacent control cycles, and pitch axis current variation data is generated based on the pitch motor current data within adjacent control cycles. Based on the azimuth axis direction switching state, the azimuth axis driving direction after reversal, the azimuth axis position change data, and the azimuth axis current change data, an azimuth axis driving hysteresis characteristic is generated. Based on the pitch axis direction switching state, the pitch axis drive direction after reversal, the pitch axis position change data, and the pitch axis current change data, pitch axis drive hysteresis characteristics are generated. According to the driving direction after commutation, the azimuth axis drive backlash characteristics and the pitch axis drive backlash characteristics are saved respectively to generate dual-axis drive backlash characteristics.

[0012] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S4 further includes: The current attitude range of the reflector is determined based on the current azimuth angle and the current pitch angle of the reflector. From the historical spot offset data formed up to the previous control cycle, extract the historical horizontal offset component and the historical vertical offset component corresponding to the current attitude range of the reflector. Based on the historical horizontal offset components and the historical vertical offset components, the historical horizontal offset mean, historical vertical offset mean, historical horizontal offset dispersion, historical vertical offset dispersion, and number of valid historical data are determined. The dual-axis drive backlash feature, the historical average horizontal offset, the historical average vertical offset, the historical horizontal offset dispersion, the historical vertical offset dispersion, and the number of valid historical data are correlated. When there is no fingerprint data corresponding to the current attitude range of the reflector in the individual error fingerprint of the reflector, the corresponding fingerprint data is generated based on the correlation result. When there is already fingerprint data corresponding to the current attitude range of the reflector in the individual error fingerprint of the reflector, the corresponding fingerprint data is updated based on the correlation result. The spot offset data generated and marked as valid in the current control cycle is written into the historical spot offset data. The written historical spot offset data is used to update the individual error fingerprint of the reflector in the next control cycle.

[0013] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S5 specifically includes: Extract the horizontal and vertical offset components from the spot offset data; Based on the mapping relationship between the preset spot offset and the dual-axis angle adjustment, the horizontal offset component and the vertical offset component are converted to generate initial azimuth compensation data and initial pitch compensation data. When the azimuth axis adjustment and the pitch axis adjustment have a coupled effect on the horizontal offset component and the vertical offset component, the initial azimuth compensation data and the initial pitch compensation data are generated based on the pre-calibrated two-dimensional joint mapping relationship.

[0014] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S5 further includes: Based on the current attitude range of the reflector, the historical spot offset features of the corresponding attitude range are read from the individual error fingerprint of the reflector. When the azimuth axis direction switching state indicator is in the reversal state, read the azimuth axis angle domain hysteresis response gap and azimuth axis hysteresis state intensity corresponding to the driving direction after the azimuth axis reversal, and generate the azimuth axis hysteresis correction amount; when the azimuth axis direction switching state indicator is in the non-reversal state, set the azimuth axis hysteresis correction amount to 0. When the pitch axis direction switching state indicator is in the reversal state, read the pitch axis angle domain hysteresis response gap and pitch axis hysteresis state intensity corresponding to the driving direction after pitch axis reversal, and generate the pitch axis hysteresis correction amount; when the pitch axis direction switching state indicator is in the non-reversal state, set the pitch axis hysteresis correction amount to 0. Based on the historical spot offset features, azimuth and pitch historical offset corrections are generated. The historical spot offset features include the historical horizontal offset mean, historical vertical offset mean, historical horizontal offset dispersion, historical vertical offset dispersion, and the number of valid historical data. The azimuth axis backlash correction and the azimuth historical offset correction are fused together to generate the azimuth fingerprint correction, and the pitch axis backlash correction and the pitch historical offset correction are fused together to generate the pitch fingerprint correction. The azimuth fingerprint correction amount is used as the incremental correction amount of the initial azimuth compensation data to generate azimuth compensation data. The pitch fingerprint correction amount is used as the incremental correction amount of the initial pitch angle compensation data to generate pitch angle compensation data.

[0015] As a preferred embodiment of the solar thermal reflector tracking control method of the present invention, step S6 specifically includes: The theoretical azimuth and theoretical pitch angles of the reflector are obtained from the theoretical light-tracking pose of the reflector. The theoretical azimuth angle of the reflector is superimposed with the azimuth compensation data to generate the target azimuth angle, and the theoretical elevation angle of the reflector is superimposed with the elevation compensation data to generate the target elevation angle; Angle continuity processing is performed on the target azimuth angle, and the target azimuth angle and the target pitch angle are limited to the allowable movement range of the corresponding drive shaft; Generate a reflector tracking control command based on the processed target azimuth and target elevation angles; The reflector tracking control command is sent to the azimuth axis drive motor and the pitch axis drive motor respectively to control the reflector to perform the tracking action.

[0016] Secondly, embodiments of the present invention provide a solar thermal reflector tracking control system, comprising: Data acquisition module: acquires solar position data, receiver target position data, reflector current attitude data, dual-axis drive feedback data, drive direction change data, and historical spot offset data; Pose generation module: Based on the sun position data, the receiver target position data, and the reflector current attitude data, generate the theoretical light-tracking pose of the reflector; Spot recognition module: Collects image data of the receiver area, determines the actual spot position corresponding to the reflector based on the image data of the receiver area, and generates spot offset data based on the actual spot position and the target position data of the receiver; Fingerprint generation module: Based on the dual-axis drive feedback data, the drive direction change data, and the historical spot offset data, generate a mirror individual error fingerprint that characterizes the dual-axis drive backlash and historical spot offset features of the mirror; Compensation generation module: Based on the spot offset data and the individual error fingerprint of the reflector, it generates azimuth compensation data and pitch compensation data; Command control module: Based on the theoretical light-tracking posture of the reflector, the azimuth compensation data, and the pitch compensation data, it generates a light-tracking control command for the reflector and controls the reflector to perform light-tracking actions according to the light-tracking control command.

[0017] The beneficial effects of this invention are as follows: By generating the theoretical light-tracking pose of the reflector and using the receiver area image to determine the actual spot position and spot offset data, this invention can achieve a closed-loop combination of theoretical light tracking and actual spot feedback. Furthermore, by combining dual-axis drive feedback data, drive direction change data, and historical spot offset data, a reflector individual error fingerprint is constructed. This allows the compensation process to simultaneously consider the drive commutation hysteresis and the long-term offset law related to the attitude range, thereby improving the pertinence of azimuth and pitch angle compensation, reducing the impact of commutation hysteresis, installation deviation, and individual differences on light-tracking accuracy, and improving the positioning accuracy of the reflected spot, light-tracking stability, and the effective light-receiving capability of the receiver. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, wherein: Figure 1 This is an overall flowchart of a solar thermal reflector light tracking control method provided in the first embodiment of the present invention; Figure 2 This is a module connection diagram of a solar thermal reflector tracking control system provided in the third embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0020] Example 1, referring to Figure 1As an embodiment of the present invention, a method for controlling the light tracking of a solar thermal reflector is provided.

[0021] S1: Acquire solar position data, receiver target position data, reflector current attitude data, dual-axis drive feedback data, drive direction change data, and historical spot offset data.

[0022] Acquire current time data and mirror field geographical location data, and calculate solar position data based on the current time data and the mirror field geographical location data; Acquire receiver target location data, which includes the three-dimensional spatial coordinates of the receiver target point in the mirror field coordinate system and the two-dimensional coordinates of the preset target center in the receiver heat absorption area in the receiver surface coordinate system; Within each control cycle, dual-axis drive feedback data is acquired according to a preset sampling interval. The dual-axis drive feedback data includes azimuth axis actual position feedback, pitch axis actual position feedback, azimuth motor current data, and pitch motor current data. Based on the actual position feedback of the azimuth axis, the actual position feedback of the pitch axis, and the preset attitude calibration parameters, the current attitude data of the reflector is generated, which includes the current azimuth angle and the current pitch angle of the reflector. Based on the azimuth axis target position command and pitch axis target position command in the already issued and stored reflector tracking control command, the drive direction change data is obtained. The drive direction change data includes azimuth axis direction switching state, pitch axis direction switching state, drive direction after azimuth axis reversal and drive direction after pitch axis reversal. Acquire historical spot offset data up to the previous control cycle. The historical spot offset data includes effective historical spot offset records and attitude interval identifiers and control cycle identifiers corresponding to each effective historical spot offset record. The effective historical spot offset records include historical horizontal offset components and historical vertical offset components.

[0023] In Example 1, step S1 is used to obtain the basic data required for generating the theoretical light-tracking pose of the reflector, identifying the light spot offset, constructing the individual error fingerprint of the reflector, generating compensation data, and issuing light-tracking control commands during the current control cycle.

[0024] It should be noted that the control cycle in this embodiment refers to one complete processing loop from basic data acquisition, generation of the theoretical light-tracking pose of the reflector, identification of light spot offset, construction of individual error fingerprints of the reflector, generation of compensation data, to issuance of light-tracking control commands. In one specific implementation, the control cycle duration can be set to 1 second.

[0025] The dual-axis drive feedback data is collected multiple times within a control cycle at a preset sampling interval, which is shorter than the control cycle duration. For example, the sampling interval can be set to 50 milliseconds, allowing multiple sets of azimuth axis actual position feedback, pitch axis actual position feedback, azimuth motor current data, and pitch motor current data to be obtained within one control cycle. Receiver area image data, spot offset data, theoretical tracking pose of the reflector, azimuth angle compensation data, pitch angle compensation data, and reflector tracking control commands are updated according to the control cycle.

[0026] In the following content Indicates the control cycle number; when it is necessary to describe multiple drive feedback sample values ​​within the same control cycle, use This indicates the sampling point number. With the above settings, the receiver area image data has sufficient acquisition and processing time, and the dual-axis drive feedback data can reflect the subdivision drive status within a single control cycle.

[0027] Furthermore, the current time data and the geographical location data of the mirror field are acquired, and the solar position data is calculated based on the current time data and the geographical location data of the mirror field.

[0028] The current time data includes year, month, day, hour, minute, and second; the telescope's geographical location data includes its longitude, latitude, and altitude. Based on the current time and geographical location data, a pre-defined astronomical position calculation method is used to determine the solar azimuth and solar altitude angles.

[0029] The solar azimuth angle is defined as the angle from true north, rotated clockwise to the direction of the sun's projection onto the horizon, ranging from 0 to 360 degrees. The solar altitude angle is defined as the angle between the direction of the sun and the horizon. In effective sun-tracking mode, the solar altitude angle is greater than 0 degrees. When the solar altitude angle is not greater than 0 degrees, it indicates that the sun is below the horizon, and the mirror sun-tracking operation ceases.

[0030] Furthermore, the receiver target location data is acquired. This receiver target location data includes the three-dimensional spatial coordinates of the receiver target point and the two-dimensional coordinates of a preset target center.

[0031] The three-dimensional spatial coordinates of the receiver target point represent the spatial position of the receiver target point in the mirror field coordinate system, and are used in step S2 to construct the target reflection direction vector pointing from the rotation center of the reflector to the receiver target point.

[0032] The preset target center two-dimensional coordinates represent the position of the target spot center in the receiver surface coordinate system within the receiver's heat absorption area. This is used in step S3 to calculate the horizontal and vertical offset components of the actual spot position relative to the target spot position.

[0033] The three-dimensional spatial coordinates of the receiver target point and the preset two-dimensional coordinates of the target center are determined separately during the system installation and calibration phase and stored in the control system. These two types of coordinates represent the position of the same target's light-receiving area in different calculation stages, and do not use the same set of coordinate values.

[0034] Furthermore, within each control cycle, dual-axis drive feedback data is acquired according to a preset sampling interval. The dual-axis drive feedback data includes azimuth axis actual position feedback, pitch axis actual position feedback, azimuth motor current data, and pitch motor current data.

[0035] The actual position feedback of the azimuth axis and the actual position feedback of the pitch axis are output by the encoder of the corresponding drive axis, which represent the mechanical angular position of the azimuth axis and the pitch axis, respectively; the azimuth motor current data and the pitch motor current data are output by the current detection interface of the corresponding servo driver, which are used to characterize the drive response state of the corresponding drive motor within the control cycle.

[0036] The azimuth motor current data includes the azimuth motor current sampling time and the azimuth motor current feedback value corresponding to the azimuth motor current sampling time; the pitch motor current data includes the pitch motor current sampling time and the pitch motor current feedback value corresponding to the pitch motor current sampling time.

[0037] The current feedback value is the equivalent current value output by the corresponding servo driver based on the motor current detection result, in amperes, and is used to characterize the drive load state of the corresponding drive motor at the corresponding sampling time. Within each control cycle, multiple azimuth motor current feedback values ​​and multiple pitch motor current feedback values ​​are acquired according to a preset sampling interval.

[0038] For ease of explanation, let's denote either drive shaft in the dual-axis configuration as... ,in, Indicates the azimuth axis. Indicates the pitch axis; the current control cycle is denoted as... The number of effective current sampling points acquired within the current control cycle is denoted as... , will the The current feedback value corresponding to each effective current sampling point is denoted as ,in, .

[0039] Current feedback values ​​that are interrupted due to communication failure, data verification anomalies, or exceed the current measurement range of the corresponding servo driver are marked as invalid and removed from the current calculation process of the current control cycle. The drive axis is determined based on the remaining valid current feedback values ​​within the current control cycle. The corresponding control cycle current value is: ; In the formula, Indicates drive shaft In the The representative value of the control cycle current within each control cycle is expressed in amperes. Indicates the first The absolute value of the current feedback value corresponding to each effective current sampling point. By taking the arithmetic average of the absolute values ​​of multiple effective current feedback values ​​within the same control cycle, the impact of abnormal fluctuations in a single current sampling value caused by transient disturbances on the drive response state judgment result can be reduced.

[0040] Based on the current representative value of the control cycle in the current control cycle and the previous control cycle, determine the drive shaft. Corresponding current change data: ; In the formula, Indicates drive shaft The change in current between two adjacent control cycles, expressed in amperes; Indicates drive shaft In the Representative value of control cycle current within each control cycle.

[0041] when hour, This represents the current value during the azimuth motor control cycle. This indicates the azimuth axis current variation data; when hour, This represents the current value during the pitch motor control cycle. This represents the pitch axis current variation data.

[0042] The azimuth axis current change data, together with the azimuth axis direction switching status and azimuth axis position change data, are used to determine whether the azimuth axis is in a hysteresis response state; the pitch axis current change data, together with the pitch axis direction switching status and pitch axis position change data, are used to determine whether the pitch axis is in a hysteresis response state.

[0043] Based on the actual position feedback of the azimuth axis, the actual position feedback of the pitch axis, and preset attitude calibration parameters, the current attitude data of the reflector is obtained. The preset attitude calibration parameters include the drive shaft zero-position deviation, transmission ratio, and installation direction correction parameters.

[0044] Specifically, zero-offset correction, transmission ratio conversion, and installation direction correction are performed on the actual position feedback of the azimuth axis and the actual position feedback of the pitch axis corresponding to the last valid sampling point in the current control cycle, respectively, to obtain the current azimuth angle and the current pitch angle of the reflector.

[0045] The dual-axis actual position feedback represents the mechanical axis position output by the encoder, while the current attitude data of the reflector represents the reflector attitude angle used for optical calculation and attitude range division after calibration. The two represent the relationship between the original data and the derived data.

[0046] Furthermore, the drive direction change data is acquired. This drive direction change data is generated by the control system based on the azimuth axis target position command and pitch axis target position command in the issued and stored reflector tracking control commands. It includes the azimuth axis direction switching state, the pitch axis direction switching state, the drive direction after the azimuth axis reversal, and the drive direction after the pitch axis reversal.

[0047] When the drive direction corresponding to the target position command issued in the previous control cycle is opposite to the previous effective drive direction, the direction switching state of the corresponding drive shaft is determined to be the reversing state, and the drive direction corresponding to the target position command in the previous control cycle is determined to be the reversing drive direction; when the two drive directions are the same, or when the target position command issued in the previous control cycle does not produce an effective angle adjustment, the direction switching state of the corresponding drive shaft is determined to be the non-reversing state.

[0048] The post-reversal drive direction is only valid when the direction switching state of the corresponding drive shaft is in the reversal state; when the direction switching state of the corresponding drive shaft is in the non-reversal state, the corresponding post-reversal drive direction is not invoked to generate the backlash correction amount.

[0049] The control system reads the target position command issued in the previous control cycle and used to form the dual-axis drive feedback data for the current control cycle, and reads the most recent valid target position command before that target position command. It compares each target position command with the actual position of the corresponding drive shaft at the start of execution to determine the signed drive direction of the corresponding target position command.

[0050] When the drive direction corresponding to the target position command issued in the previous control cycle is opposite to the previous effective drive direction, the direction switching state of the corresponding drive shaft is determined to be the reversing state, and the drive direction corresponding to the target position command issued in the previous control cycle is determined to be the reversing drive direction; when the two drive directions are the same, or when the target position command issued in the previous control cycle does not produce an effective angle adjustment, the direction switching state of the corresponding drive shaft is determined to be the non-reversing state.

[0051] When the azimuth axis is driven along the direction of increasing azimuth angle, the driving direction after commutation is the positive direction of the azimuth axis; when driven along the direction of decreasing azimuth angle, the driving direction after commutation is the negative direction of the azimuth axis. When the pitch axis is driven along the direction of increasing pitch angle, the driving direction after commutation is the positive direction of the pitch axis; when driven along the direction of decreasing pitch angle, the driving direction after commutation is the negative direction of the pitch axis.

[0052] Therefore, the data on changes in drive direction can not only indicate whether the corresponding drive shaft has changed direction, but also indicate the drive direction after the change, providing a directional basis for generating the signed angle domain hysteresis response gap in step S4.

[0053] Furthermore, historical spot offset data up to the previous control cycle is acquired. Each valid record in the historical spot offset data includes a horizontal offset component, a vertical offset component, an attitude interval identifier, and a control cycle identifier.

[0054] The spot offset data generated in the current control cycle is not included in the calculation of the historical offset statistics used in the current control cycle. Instead, it is written into the historical spot offset data after the mirror individual error fingerprint is generated in the current control cycle, and then used in the next control cycle.

[0055] Historical spot offset data is stored separately according to attitude intervals. For each attitude interval, when the number of valid historical spot offset records exceeds a preset storage limit, the earliest record is deleted in chronological order of control cycles, and the most recent preset number of valid records are retained. For example, the preset storage limit can be set to 100 to 500 records.

[0056] The attitude interval is jointly determined by the mirror azimuth interval and the mirror pitch interval. For example, the azimuth angle is divided into an azimuth interval every 2 degrees, and the pitch angle is divided into a pitch interval every 2 degrees. The combination of the azimuth interval index and the pitch interval index is used as the attitude interval identifier.

[0057] By using the above attitude partitioning storage method, the historical light spot shift patterns formed by the reflector under different combinations of azimuth and pitch angles can be recorded separately.

[0058] S2: Based on the solar position data, the receiver target position data, and the current attitude data of the reflector, generate the theoretical light-tracking pose of the reflector.

[0059] Construct a solar incidence direction vector based on the solar position data; Based on the three-dimensional spatial coordinates of the receiver target point in the mirror field coordinate system, a target reflection direction vector is constructed in the local coordinate system of the reflector. Based on the solar incident direction vector and the target reflection direction vector, the theoretical normal vector of the reflector is calculated according to the law of reflection. The theoretical normal vector of the reflector is converted into the theoretical azimuth and theoretical elevation angles of the reflector. Based on the current attitude data of the reflector, angular continuity processing is performed on the theoretical azimuth angle of the reflector, and the theoretical azimuth angle and the theoretical pitch angle of the reflector are limited to the allowable movement range of the corresponding drive shaft. The processed theoretical azimuth and theoretical elevation angles of the reflector are used as the theoretical ray-tracking pose of the reflector.

[0060] In Example 1, step S2 is used to generate the theoretical tracking pose of the reflector that should be achieved in the current control cycle based on the solar position data, the receiver target position data, and the reflector current attitude data, so as to provide a geometric optical reference for subsequent compensation calculations.

[0061] The theoretical tracking attitude of the reflector is calculated based on the solar incidence direction and the receiver target direction, without considering the effects of drive hysteresis, installation deviation, support structure deformation, mirror deformation, and historical spot offset. The theoretical tracking attitude of the reflector is not directly used as the final control command, but is superimposed on the azimuth and elevation compensation data in step S6.

[0062] Each mirror in the mirror field performs step S2. Because different mirrors are installed in different positions in the mirror field, their target reflection directions from the mirror rotation center to the receiver target point are different, and therefore the theoretical ray-tracking poses corresponding to different mirrors are also different.

[0063] It should be noted that a mirror field coordinate system should be established before performing step S2. and the local coordinate system of the mirror .

[0064] The mirror field coordinate system takes the preset reference point of the mirror field as the origin, the east direction as the positive x-axis, the north direction as the positive y-axis, and the vertical upward direction as the positive z-axis.

[0065] The local coordinate system of the reflector has its origin at the current center of rotation of the reflector, and its three coordinate axes are parallel to the three coordinate axes of the mirror field coordinate system. The local coordinate system of the reflector is used to represent the spatial orientation relative to the current center of rotation of the reflector and does not rotate with the attitude of the reflector surface.

[0066] Furthermore, a solar incidence direction vector is constructed based on solar position data.

[0067] The solar azimuth angle is denoted as The solar altitude angle is denoted as Let the vector of the direction of solar incidence be denoted as In this embodiment, the solar incidence direction vector refers to the unit direction vector pointing from the center of rotation of the reflector towards the sun, and its direction is opposite to the actual propagation direction of sunlight from the sun to the reflector. ; In the formula, This represents the component of the solar incidence vector in the due east direction. This represents the component of the solar incidence vector in the due north direction. This represents the vertical component of the solar incidence direction vector.

[0068] It should be noted that the solar azimuth angle, solar altitude angle, theoretical azimuth angle of the reflector, and theoretical elevation angle of the reflector are all expressed in degrees as the unit of angle in the instruction manual. When the trigonometric function called by the control program takes radians as input, the solar azimuth angle and solar altitude angle are converted from degrees to radians before the trigonometric function calculation is performed; when the control program uses angle-based trigonometric functions, the corresponding angle values ​​are used directly.

[0069] Furthermore, based on the three-dimensional spatial coordinates of the receiver target point in the receiver target position data, a target reflection direction vector is constructed in the local coordinate system of the reflector.

[0070] The three-dimensional spatial coordinates of the receiver target point are transformed from the mirror field coordinate system to the local coordinate system of the reflector to obtain the local coordinates of the receiver target point relative to the center of rotation of the reflector, denoted as: ; in, , , These represent the coordinates of the receiver target point relative to the center of rotation of the reflector along the three coordinate axes of the reflector's local coordinate system.

[0071] Construct the target reflection direction vector based on the local coordinates: ; In the formula, This represents the unit direction vector pointing from the center of rotation of the reflector to the target point of the receiver; This indicates the spatial distance between the receiver target point and the center of rotation of the reflector.

[0072] Furthermore, based on the solar incident direction vector and target reflection direction vector The theoretical normal vector of the mirror is calculated based on the law of reflection.

[0073] When the above direction definition is used, the theoretical normal vector of the reflector is determined by the angle bisector of the solar incidence vector and the target reflection vector: ; In the formula, This represents the theoretical normal vector of the mirror; Represents vector The Euclidean norm.

[0074] Furthermore, the theoretical normal vector of the reflector is converted into the theoretical azimuth and theoretical elevation angles of the reflector. The theoretical normal vector of the reflector... The three coordinate components are denoted as , , ,in Let be the component of the normal vector in the eastward direction (x-axis). This represents the component of the normal vector in the due north direction (y-axis). This is the component of the normal vector in the vertical direction (z-axis).

[0075] Theoretical azimuth angle of the reflecting mirror Defined as the angle between the projection direction of the theoretical normal vector onto the horizontal plane and the true north direction, expressed as: ; In the formula, For the arctangent function in the four quadrants, the component of the normal vector in the due east direction is... The first parameter is the component of the normal vector in the due north direction. The second parameter is used to calculate the angle between the projection onto the horizontal plane and the true north direction. The function automatically determines the quadrant based on the signs of the two parameters, and the returned angle range is typically [value missing]. arrive (Corresponding to -180 degrees to 180 degrees). When the return value is negative, add 360 degrees to the return value to make the theoretical azimuth angle of the reflector. The value range is converted to 0 degrees to 360 degrees, consistent with the convention for representing azimuth axis angles.

[0076] Theoretical pitch angle of a reflecting mirror Defined as the angle of elevation of the theoretical normal vector relative to the horizontal plane, its expression is: ; In the formula, This represents the length of the projection of the theoretical normal vector onto the horizontal plane. ( The angle between the normal vector and the horizontal plane is calculated using the projection length. Since the theoretical normal vector of the reflecting mirror points outwards from the mirror surface and is located in the upper hemisphere, Therefore The value ranges from 0 degrees to 90 degrees. This conversion method has good numerical stability when the normal vector is close to vertical, and can accurately give the pitch angle.

[0077] Furthermore, based on the current attitude data of the reflector, angular continuity processing is performed on the theoretical azimuth angle of the reflector.

[0078] When the theoretical azimuth angle of the reflector and the current azimuth angle of the reflector cross the boundary between 0 degrees and 360 degrees, the direction with the smaller absolute value of the rotation angle is selected as the azimuth axis rotation direction based on the periodic equivalence relationship between the two.

[0079] When the azimuth axis is allowed to rotate continuously, the theoretical azimuth angle of the processed reflector is limited to a range greater than or equal to 0 degrees and less than 360 degrees; when the azimuth axis is mechanically limited, the theoretical azimuth angle of the reflector is limited to the range of motion allowed by the azimuth axis.

[0080] The theoretical pitch angle of the reflector is limited to the allowable range of motion of the pitch axis. The allowable range of motion of the azimuth axis and the allowable range of motion of the pitch axis are predetermined according to the mechanical structure of the dual-axis drive mechanism of the reflector.

[0081] Therefore, the theoretical azimuth angle of the mirror obtained through coordinate transformation, reflection law calculation, and angular continuity processing is... And the theoretical pitch angle of the reflecting mirror As the theoretical pose for light tracking using a reflector, that is: ; In the formula, The theoretical tracking pose of the reflector is represented by a two-dimensional angle result composed of the theoretical azimuth and theoretical pitch angles of the reflector. This theoretical tracking pose does not yet consider the influence of individual manufacturing errors, installation deviations, drive hysteresis, and mirror deformation on the actual reflected light path. Its function is to provide an angular reference before compensation in the subsequent step S5, and to be superimposed with the compensation data in step S6 to form the final target tracking angle.

[0082] It should also be noted that the theoretical tracking pose of the reflector generated in step S2 is recalculated in each control cycle as the sun's position changes. Although the theoretical tracking pose of the reflector does not consider individual error factors such as manufacturing errors, installation deviations, drive hysteresis, and mirror deformation, the target reflection direction vector corresponding to each reflector differs due to the spatial position difference between the reflector and the receiver because of the different installation positions of the different reflectors in the mirror field. Therefore, each reflector in the mirror field independently performs step S2 to obtain the theoretical tracking pose corresponding to its installation position.

[0083] S3: Acquire image data of the receiver area, determine the actual spot position corresponding to the reflector based on the image data of the receiver area, and generate spot offset data based on the actual spot position and the receiver target position data.

[0084] Image data of the receiver area is acquired by an image acquisition device deployed on the receiver side; The image data of the receiver area is processed for spot recognition to extract the spot outline corresponding to the current reflector and determine the center pixel coordinates of the spot; Based on the receiver image coordinate calibration relationship, the center pixel coordinates of the light spot are converted into the actual light spot position in the receiver surface coordinate system; The two-dimensional coordinates of the preset target center in the receiver surface coordinate system are used as the target spot position; Based on the actual spot position and the target spot position, spot offset data containing horizontal and vertical offset components is generated.

[0085] In Example 1, step S3 is used to collect image data of the receiver area, identify the actual position of the light spot projected onto the surface of the receiver by the current reflector, and generate light spot offset data based on the actual light spot position and the preset two-dimensional coordinates of the target center.

[0086] The theoretical tracking pose of the reflector generated in step S2 is a theoretical result obtained under ideal geometric optics conditions. Step S3 obtains the actual position of the reflected light spot on the receiver's heat absorption area. By comparing the actual spot position with the target spot position, the actual tracking error of the reflector can be converted into quantifiable spot offset data.

[0087] Furthermore, image data of the receiver area is acquired. This image data is obtained by an image acquisition device deployed on the receiver side. The image acquisition device can be a visible light camera, a near-infrared camera, or a thermal imaging camera, and is installed in a fixed position that can cover the heat-absorbing area of ​​the receiver.

[0088] The field of view of the image acquisition device covers the receiver's heat absorption area and the preset allowable offset range, so that even if the center of the light spot deviates from the receiver's heat absorption area, it can still be captured by the image acquisition device without exceeding the preset safety range.

[0089] In this embodiment, the image acquisition device acquires at least one frame of receiver area image data in each control cycle, so that the spot offset data generated in the current control cycle can participate in the subsequent processing of the same control cycle.

[0090] When the single-frame exposure time and image processing time of the image acquisition device are shorter than the control cycle, multiple frames of images can be continuously acquired within one control cycle, and the image with sharpness and exposure that meets the preset requirements can be selected from the multiple frames as the receiver area image data for the current control cycle.

[0091] The spatial resolution of the image acquisition device is matched with the effective threshold for spot offset. The effective threshold for spot offset includes the effective threshold for horizontal offset. and vertical offset effective threshold The effective threshold for horizontal offset and the effective threshold for vertical offset are determined based on the physical pixel size in the horizontal and vertical directions of the receiver surface coordinate system, the spot center positioning error, and the receiver's allowable tracking error, respectively.

[0092] After completing the receiver image coordinate calibration, the physical dimensions of the receiver surface corresponding to a single pixel in the horizontal and vertical directions of the image are determined near the preset target center on the receiver surface. The physical dimension corresponding to a single pixel in the horizontal direction is denoted as... The physical size corresponding to a single pixel in the vertical direction of the image is denoted as The and The unit is millimeters per pixel.

[0093] Furthermore, under the condition that the mirror orientation and the light spot projection state remain unchanged, continuous data acquisition is performed. Frame receiver region image data, where, Not less than 30. Determine the horizontal and vertical coordinates of the spot center in the image data of each frame's receiver area.

[0094] The first The horizontal coordinates of the spot center corresponding to the image data of the frame receiver region are as follows: Mark the vertical coordinates of the corresponding light spot center as ,in, .

[0095] The average value of the horizontal coordinate of the light spot center is expressed as: ; The average value of the vertical coordinate of the light spot center is expressed as: ; The horizontal spot center positioning error is determined based on the dispersion of the horizontal coordinates of the spot centers across multiple frames. : ; The vertical spot center positioning error is determined based on the dispersion of the vertical coordinates of the spot centers in multiple frames. : ; In the formula, This indicates the error in the center positioning of the horizontal light spot. The vertical spot center positioning error is represented by millimeters. The horizontal spot center positioning error and the vertical spot center positioning error are used to characterize the spot center positioning fluctuations during image acquisition and spot recognition when the mirror attitude and spot projection state remain unchanged.

[0096] Horizontal offset effective threshold Represented as: ; Vertical offset effective threshold Represented as: ; In the formula, This represents the physical offset between two pixels that can be stably distinguished in the horizontal direction during the image acquisition and spot recognition process. This represents the physical offset between two pixels that can be stably distinguished in the vertical direction during the image acquisition and spot recognition process. This indicates the horizontal positioning error limit determined based on the fluctuation of the center positioning of multiple light spots; This indicates the vertical positioning error limit determined based on the fluctuation of the center positioning of multiple light spots.

[0097] The maximum allowable tracking error of the receiver in the horizontal direction is denoted as . The maximum allowable tracking error of the receiver in the vertical direction is denoted as... The aforementioned and The size of the target light-receiving area of ​​the receiver, the allowable deviation range of the light spot, and the receiver's safe light-receiving requirements are preset.

[0098] After determining the effective thresholds for horizontal and vertical offsets, verify them against the corresponding allowable tracking errors, and ensure that: ; ;when When this occurs, it indicates that the current image acquisition device's spatial resolution in the horizontal direction or the stability of the spot center positioning cannot meet the requirements of light tracking control; when When this occurs, it indicates that the current image acquisition device's spatial resolution in the vertical direction or the stability of the spot center positioning cannot meet the requirements of light tracking control.

[0099] When the effective threshold for horizontal offset or vertical offset does not meet the corresponding allowable tracking error requirements, the image resolution of the image acquisition device is increased, the field of view of the image acquisition is reduced, the installation position of the image acquisition device is adjusted, or the receiver image coordinate calibration is re-executed, so that the newly determined effective threshold for horizontal offset and effective threshold for vertical offset do not exceed the corresponding allowable tracking error.

[0100] In one specific implementation, the physical size of pixels in the horizontal direction of the image for Millimeters per pixel, the physical size of a pixel in the vertical direction of an image. for Millimeters per pixel. Continuous acquisition while maintaining constant mirror orientation and light spot projection state. Frame receiver area image data to determine the horizontal spot center positioning error for Millimeters, vertical spot center positioning error for Millimeters. Horizontal allowable tracking error. and vertical allowable tracking error All set to Millimeters.

[0101] Based on the above parameters, the effective threshold for horizontal offset is: ; The effective threshold for vertical offset is: ; Both the effective threshold for horizontal offset and the effective threshold for vertical offset are less than the corresponding allowable tracking error. Therefore, the spatial resolution and spot center positioning stability of the current image acquisition device can meet the requirements of tracking control.

[0102] In other implementations, the physical size of the pixels in the horizontal direction Physical size of pixels in the vertical direction It can range from 1.5 mm per pixel to 4.0 mm per pixel; horizontal spot center positioning error Vertical spot center positioning error The allowable depth ranges from 0.5 mm to 2.0 mm; horizontal tracking error is permissible. and vertical allowable tracking error The effective thresholds for horizontal and vertical offsets, determined according to the above calculations, can be from 15 mm to 30 mm.

[0103] Furthermore, the validity of the horizontal and vertical offset components is determined separately.

[0104] When horizontal offset component The absolute value is less than the effective threshold of horizontal offset. At that time, set the horizontal offset component of the current control cycle to 0: ; When horizontal offset component The absolute value is greater than or equal to the effective threshold of horizontal offset. At the same time, the horizontal offset component of the current control cycle is retained and included in the generation of subsequent compensation data: ; When vertical offset component The absolute value is less than the effective threshold of vertical offset. At that time, set the vertical offset component of the current control cycle to 0: ; When vertical offset component The absolute value is greater than or equal to the effective threshold of vertical offset. At the same time, the vertical offset component of the current control cycle is retained and included in the generation of subsequent compensation data: ; By setting effective thresholds for horizontal and vertical offsets respectively, we can avoid misidentifying image acquisition noise and spot center positioning fluctuations as effective tracking deviations, while retaining actual spot offsets exceeding the stable recognition capability for use in the generation of subsequent azimuth and pitch compensation data.

[0105] It should also be noted that when multiple mirrors in the mirror field simultaneously reflect sunlight to the same receiver, time-division identification or time-series association methods can be used to determine the light spot area corresponding to the current mirror.

[0106] Time-division recognition refers to the process of performing a small angular disturbance on the mirror to be identified during the calibration or correction phase, and identifying the spot area in the receiver area image that moves synchronously with the angular disturbance as the spot area corresponding to the mirror to be identified.

[0107] Temporal correlation refers to controlling the reflector to be identified to produce minute angle changes according to a preset forward and reverse angle change sequence within a preset identification period, and continuously acquiring multiple frames of receiver area image data. By comparing the brightness changes in consecutive image frames, image regions that match the brightness change sequence with the angle change sequence of the reflector to be identified are extracted, and these image regions are determined as the light spot regions corresponding to the reflector to be identified.

[0108] The amplitude of the minute angle change is less than the maximum compensation angle allowed within a single control cycle, and meets the receiver's safe light reception requirements.

[0109] When there is no significant overlap between the spot areas corresponding to different reflectors, candidate recognition areas can be set according to the theoretical light-tracking pose of the reflectors and the position of the target spot, and the spot contour can be extracted within the candidate recognition areas.

[0110] Furthermore, spot recognition processing is performed on the image data of the receiver area.

[0111] First, the image data of the receiver area is processed by grayscale conversion and brightness normalization, and candidate bright areas are extracted according to a preset spot brightness threshold. The preset spot brightness threshold can be calibrated based on the bit depth of the image acquisition device, the surface reflection characteristics of the receiver, the camera exposure parameters, and the solar irradiance.

[0112] Subsequently, connected component analysis was performed on the candidate bright areas to remove interference areas with areas smaller than the preset minimum spot area, aspect ratios exceeding the preset range, or area fill ratios lower than the preset range, in order to eliminate the influence of image noise, receiver edge reflections, and local bright spots on the spot recognition results.

[0113] For candidate spot regions that meet the preset conditions, their outer contours are extracted as spot contours, and the center pixel coordinates of the spot are determined.

[0114] In one specific implementation, the center pixel coordinates of the light spot can be determined using a brightness-weighted centering method. The pixel coordinates within the light spot outline are then... The pixel coordinates of each pixel are: The grayscale value or brightness value of this pixel is recorded as The total number of pixels within the light spot outline is denoted as Then the center pixel coordinates of the light spot , is represented as: ; ; In the formula, This represents the pixel coordinates of the light spot center in the horizontal direction of the image. This represents the pixel coordinates of the center of the light spot in the vertical direction of the image.

[0115] When using the brightness-weighted centroid method, the higher brightness areas contribute more to the center position of the spot, which can reduce the impact of brightness attenuation at the spot edge and local noise on the center positioning result. When the spot brightness distribution is relatively uniform, the geometric centroid method can also be used to determine the center pixel coordinates of the spot.

[0116] Furthermore, based on the receiver image coordinate calibration relationship, the center pixel coordinates of the light spot are converted into the actual light spot position in the receiver surface coordinate system.

[0117] The receiver image coordinate calibration relationship is established during the system installation or debugging phase. When the receiver surface is approximately planar, multiple calibration points with known physical coordinates can be selected in the receiver's heat absorption area, and a homography transformation relationship can be established based on the pixel coordinates and physical coordinates of the calibration points.

[0118] The receiver image coordinate calibration matrix is ​​denoted as... The actual spot position is recorded as The transformation relation is then expressed as: ; In the formula, This represents the scale factor, used to align homogeneous coordinates. This represents the horizontal coordinate of the actual light spot position in the receiver surface coordinate system. This represents the vertical coordinate of the actual light spot position in the receiver surface coordinate system.

[0119] When the receiver's heat-absorbing surface is curved, the receiver's heat-absorbing surface can be divided into multiple local regions, and a local calibration relationship between the image pixel coordinates and the receiver surface coordinates can be established for each local region.

[0120] Furthermore, the preset two-dimensional coordinates of the target center in the receiver target location data are used as the target spot position, and the target spot position is denoted as... .

[0121] Both the actual spot position and the target spot position lie in the same receiver surface coordinate system. Horizontal and vertical offset components are generated based on the actual and target spot positions: ; ; In the formula, This represents the horizontal offset component of the current control cycle; This represents the vertical offset component of the current control cycle.

[0122] When the horizontal offset component is positive, it indicates that the actual light spot deviates from the target light spot position along the preset positive direction in the horizontal direction of the receiver surface; when the horizontal offset component is negative, it indicates that the actual light spot deviates from the target light spot position in the opposite direction. The sign of the vertical offset component is determined according to the vertical axis direction of the receiver surface coordinate system.

[0123] When a valid spot contour that meets the brightness, area, shape, and position conditions is identified within the current control cycle, the spot offset data generated in the current control cycle is marked as valid data.

[0124] Based on the aforementioned effective thresholds for horizontal and vertical offsets, the horizontal and vertical offset components are processed respectively. When the absolute value of the horizontal offset component is less than the effective threshold, the horizontal offset component is set to 0; when the absolute value of the horizontal offset component is greater than or equal to the effective threshold, the horizontal offset component is retained. When the absolute value of the vertical offset component is less than the effective threshold, the vertical offset component is set to 0; when the absolute value of the vertical offset component is greater than or equal to the effective threshold, the vertical offset component is retained.

[0125] When both the processed horizontal and vertical offset components are 0, angle compensation caused by the current spot offset is not triggered; when the processed horizontal or vertical offset components are not 0, the corresponding offset components are involved in the generation of subsequent compensation data.

[0126] If no valid spot contour that simultaneously meets the brightness, area, shape, and position conditions is identified within the current control cycle, the spot offset data for the current control cycle will be marked as invalid data.

[0127] Invalid spot offset data is not written into the historical spot offset data, nor is it used as the current spot offset input in the generation of the initial angle compensation data. When the current spot offset data is invalid, the spot offset data from the previous control cycle is not directly used.

[0128] The current control cycle can generate mirror tracking control commands based solely on the mirror's theoretical tracking pose. When the historical offset statistics corresponding to the current attitude range of the mirror reach the preset reliability requirements, conservative compensation data can also be generated based on the historical offset characteristics and drive backlash characteristics in the mirror's individual error fingerprint.

[0129] The effective spot offset data is bound to the attitude interval identifier and control cycle identifier of the current control cycle. The attitude interval identifier is determined based on the current azimuth angle and current pitch angle of the reflector, and is used to store historical spot offset data according to the attitude interval.

[0130] S4: Based on the dual-axis drive feedback data, the drive direction change data, and the historical spot offset data, generate a mirror individual error fingerprint that characterizes the dual-axis drive backlash and historical spot offset features of the mirror.

[0131] Based on the actual azimuth position feedback before the azimuth target position command of the previous control cycle is executed and the actual azimuth position feedback after the azimuth target position command is executed, azimuth position change data is generated. Based on the pitch axis actual position feedback before the pitch axis target position command of the previous control cycle is executed and the pitch axis actual position feedback after the pitch axis target position command is executed, pitch axis position change data is generated. Azimuth axis current variation data is generated based on the azimuth motor current data within adjacent control cycles, and pitch axis current variation data is generated based on the pitch motor current data within adjacent control cycles. Based on the azimuth axis direction switching state, the azimuth axis driving direction after reversal, the azimuth axis position change data, and the azimuth axis current change data, an azimuth axis driving hysteresis characteristic is generated. Based on the pitch axis direction switching state, the pitch axis drive direction after reversal, the pitch axis position change data, and the pitch axis current change data, pitch axis drive hysteresis characteristics are generated. According to the driving direction after commutation, the azimuth axis drive backlash characteristics and the pitch axis drive backlash characteristics are saved respectively to generate dual-axis drive backlash characteristics.

[0132] The current attitude range of the reflector is determined based on the current azimuth angle and the current pitch angle of the reflector. From the historical spot offset data formed up to the previous control cycle, extract the historical horizontal offset component and the historical vertical offset component corresponding to the current attitude range of the reflector. Based on the historical horizontal offset components and the historical vertical offset components, the historical horizontal offset mean, historical vertical offset mean, historical horizontal offset dispersion, historical vertical offset dispersion, and number of valid historical data are determined. The dual-axis drive backlash feature, the historical average horizontal offset, the historical average vertical offset, the historical horizontal offset dispersion, the historical vertical offset dispersion, and the number of valid historical data are correlated. When there is no fingerprint data corresponding to the current attitude range of the reflector in the individual error fingerprint of the reflector, the corresponding fingerprint data is generated based on the correlation result. When there is already fingerprint data corresponding to the current attitude range of the reflector in the individual error fingerprint of the reflector, the corresponding fingerprint data is updated based on the correlation result. The spot offset data generated and marked as valid in the current control cycle is written into the historical spot offset data. The written historical spot offset data is used to update the individual error fingerprint of the reflector in the next control cycle.

[0133] Step S4 includes the dual-axis drive backlash feature generation stage and the attitude partitioning error fingerprint construction stage.

[0134] Phase 1: Generation of dual-axis driven hysteresis features.

[0135] It should be noted that drive backlash refers to the phenomenon that when the dual-axis drive mechanism of the reflector switches directions, the drive motor has already generated a driving action, but the load end has not synchronously formed an effective position response due to transmission clearance, lead screw idle, elastic deformation of the transmission chain, and mechanical lag.

[0136] Drive hysteresis typically manifests as follows: after the direction of the corresponding drive shaft changes, the motor current changes effectively, but the actual position feedback changes little within the preset response time.

[0137] For ease of explanation, let's denote either drive shaft in the dual-axis configuration as... ,in Indicates the azimuth axis. Indicates the pitch axis.

[0138] The actual position of the drive shaft before the start of the target position command of the previous control cycle is recorded as . The actual position of the drive shaft obtained in the current control cycle after the execution of the target position command is recorded as . The actual change in position of the drive shaft Represented as: ; For the azimuth axis, when adjacent actual position feedbacks cross the 0-degree and 360-degree boundaries, the change in the actual azimuth axis position is determined according to the signed shortest angular distance; for the pitch axis, the change in the actual pitch axis position is determined according to the direct difference between the two pitch angles.

[0139] When collecting multiple motor current feedback values ​​within a control cycle, invalid current feedback values ​​are marked and removed as described in step S1. The absolute values ​​of the remaining valid current feedback values ​​are then arithmetically averaged to obtain the representative current value of the corresponding drive shaft during the control cycle. The absolute value of the difference between the representative current value of the current control cycle and the representative current value of the previous control cycle is used as the current change data of the corresponding drive shaft.

[0140] Specifically, representative values ​​of the azimuth motor control cycle current are generated based on the azimuth motor current data, and azimuth axis current variation data are further generated; representative values ​​of the pitch motor control cycle current are generated based on the pitch motor current data, and pitch axis current variation data are further generated.

[0141] Furthermore, the system determines whether the drive shaft is in a hysteresis response state based on the direction switching status, actual position change data, and current change data.

[0142] When the drive shaft changes direction, the current change data reaches the corresponding effective current change threshold, and the absolute value of the actual position change is less than the minimum effective position change in a single control cycle under normal drive conditions, the corresponding drive shaft is determined to be in hysteresis response state; otherwise, the corresponding drive shaft is determined to be not in hysteresis response state.

[0143] The effective current change threshold is determined through commutation calibration tests based on the motor's rated current, no-load current, and stable operating current. The minimum effective position change is determined based on the minimum position change that the drive mechanism can stably form under normal hysteresis response conditions, the same control cycle, and the same sampling conditions.

[0144] For a drive shaft in a hysteresis response state, the intensity of the hysteresis state is determined based on the degree of current change and the degree of insufficient actual position response.

[0145] Hysteresis intensity is represented by a value between 0 and 1, and is determined by the hysteresis state mapping relationship established through the commutation calibration test of the drive mechanism. The more significant the current change and the smaller the absolute value of the actual position change, the greater the hysteresis intensity; the smaller the hysteresis intensity, the closer the absolute value of the actual position change is to the normal minimum effective position change.

[0146] drive shaft The angular domain hysteresis response gap is expressed as: ; In the formula, Indicates drive shaft The hysteresis response state; it is 1 when in hysteresis response state and 0 when not in hysteresis response state; This indicates the sign of the driving direction after commutation; the value is 1 for the direction of increasing angle and -1 for the direction of decreasing angle. Indicates drive shaft Minimum effective position change within a single control cycle under normal driving conditions Indicates drive shaft The hysteresis response gap in the angular domain.

[0147] The unit of the angular domain hysteresis response notch is the same as the unit of the actual position feedback of the drive shaft. When the actual position feedback is expressed in degrees, the unit of the angular domain hysteresis response notch is degrees.

[0148] It should be noted that the hysteresis response gap in the angle domain represents the insufficiency of the actual position response after direction switching relative to the normal minimum effective position response. It is an estimate of the hysteresis response used for compensation control and is not directly equivalent to the actual mechanical backlash angle of the drive mechanism.

[0149] when When, the azimuth axis hysteresis response gap and azimuth axis hysteresis state intensity are obtained; when At that time, the pitch axis angle domain hysteresis response gap and pitch axis hysteresis state intensity are obtained. The above data together constitute the dual-axis drive hysteresis characteristics of the current control cycle.

[0150] The dual-axis drive hysteresis characteristics are stored separately according to the drive direction after commutation. When the drive direction after commutation is the angle increasing direction, the corresponding angle domain hysteresis response gap and hysteresis state intensity are written into the hysteresis characteristic field of the angle increasing direction; when the drive direction after commutation is the angle decreasing direction, the corresponding data are written into the hysteresis characteristic field of the angle decreasing direction.

[0151] By classifying and storing data by direction, we can avoid the cancellation of hysteresis response gaps with opposite signs under different commutation directions during subsequent fingerprint updates.

[0152] Phase 2: Construction and updating of attitude partitioning error fingerprint.

[0153] Furthermore, the current attitude range of the reflector is determined based on the current azimuth angle and the current pitch angle of the reflector.

[0154] The azimuth and elevation ranges are divided into intervals according to preset step sizes, and the combination of the azimuth interval index to which the current azimuth of the reflector belongs and the elevation interval index to which the current elevation of the reflector belongs is used as the current attitude interval identifier of the reflector.

[0155] For example, the azimuth angle can be divided into azimuth intervals of 2 degrees each, and the pitch angle into pitch intervals of 2 degrees each. The smaller the step size for dividing the attitude intervals, the more detailed the description of attitude changes by the individual error fingerprint of the reflector; the larger the step size for dividing the attitude intervals, the more historical data can be accumulated within a single attitude interval.

[0156] Based on the historical spot offset data formed up to the previous control cycle, the effective historical horizontal offset component and the effective historical vertical offset component corresponding to the current attitude range of the reflector are extracted.

[0157] The effective historical horizontal offset components and effective historical vertical offset components are statistically processed separately to obtain the historical average horizontal offset and the historical average vertical offset.

[0158] The historical average horizontal offset is used to characterize the long-term horizontal offset trend of the mirror within the current attitude range, while the historical average vertical offset is used to characterize the long-term vertical offset trend of the mirror within the current attitude range.

[0159] Meanwhile, based on the degree of dispersion of the effective historical spot offset relative to the corresponding historical offset mean, the historical horizontal offset dispersion and historical vertical offset dispersion are determined.

[0160] The smaller the historical offset dispersion, the more stable the historical offset pattern within the corresponding attitude range; the larger the historical offset dispersion, the more significant the impact of short-term wind load, image noise, and other random disturbances on the historical offset data.

[0161] Record the number of valid historical data points within the current attitude range of the reflector. The more valid historical data points and the smaller the historical offset dispersion, the higher the reliability of the historical offset statistics; when the number of valid historical data points is insufficient or the historical offset dispersion is large, the reliability of the historical offset statistics decreases.

[0162] When there are no valid historical spot offset records in the current attitude range of the reflector, the historical offset statistics result is marked as unavailable, and the historical offset confidence level is set to the lowest level. The state of no historical data is not interpreted as the historical offset being 0.

[0163] Furthermore, the dual-axis drive backlash characteristics, historical average horizontal offset, historical average vertical offset, historical horizontal offset dispersion, historical vertical offset dispersion, and the number of valid historical data corresponding to the current attitude range of the reflector are correlated to form fingerprint data corresponding to the current attitude range of the reflector.

[0164] The fingerprint data corresponding to each pose range includes: The hysteresis response gap and hysteresis state intensity in the angular domain corresponding to the direction of increasing azimuth axis angle; The angular domain hysteresis response gap and hysteresis state intensity corresponding to the direction of decrease in azimuth axis angle; The angle domain hysteresis response gap and hysteresis state intensity corresponding to the direction of increasing pitch axis angle; The angle domain hysteresis response gap and hysteresis state intensity corresponding to the direction of pitch axis angle decrease; Historical horizontal offset mean, historical vertical offset mean, historical horizontal offset dispersion, historical vertical offset dispersion, and number of valid historical data.

[0165] When a new effective drive hysteresis feature is generated in the current control cycle, the original drive hysteresis feature under the same attitude range and the same drive direction is smoothly updated according to the preset update ratio.

[0166] The preset update ratio is used to adjust the degree of participation between the original hysteresis features and the current hysteresis features. It can be preset according to the stability requirements of the individual error fingerprint of the mirror and the response requirements for recent error changes. When it is necessary to improve the response capability for recent changes, the participation degree of the current hysteresis features is increased; when it is necessary to improve fingerprint stability, the participation degree of the original hysteresis features is increased.

[0167] When the current control cycle does not identify an effective hysteresis response in the corresponding direction, the original hysteresis characteristics of the current attitude range and the corresponding driving direction of the reflector remain unchanged.

[0168] The fingerprint data corresponding to each attitude range of the reflector together constitute the individual error fingerprint of the reflector.

[0169] It should be noted that the spot offset data generated in the current control cycle is not included in the historical offset statistics in the individual error fingerprint of the reflector called in the current control cycle.

[0170] Before writing the spot offset data of the current control cycle into the historical spot offset data, step S4 first generates the individual error fingerprint of the reflector required for the current control cycle based on the historical spot offset data formed up to the previous control cycle, for reading by step S5 of the current control cycle.

[0171] After the individual error fingerprint of the reflector in the current control cycle is generated, the current spot offset data generated and marked as valid in step S3 is bound to the reflector's current attitude interval identifier and the current control cycle identifier and then written into the historical spot offset data.

[0172] The written historical spot offset data is only used for historical offset statistics and mirror individual error fingerprint updates in the next control cycle, and is not returned for compensation data calculation in the current control cycle.

[0173] By setting the timing parameters as described above, we can prevent the current spot offset data from being reused as both the current spot compensation input and the current historical offset input.

[0174] If the spot offset data generated in the current control cycle is marked as invalid data, it will not be written into the historical spot offset data, nor will the historical offset statistics for the corresponding attitude range be updated. When a valid drive hysteresis feature is identified in the current control cycle, only the drive hysteresis feature for the corresponding attitude range and the corresponding drive direction needs to be updated.

[0175] Through the above two stages, step S4 completes the dual-axis drive backlash feature recognition and the construction of individual error fingerprints of the reflector based on the attitude range, so that the subsequent step S5 can simultaneously use the current spot offset data and the long-term error features of the reflector in the current attitude range to generate compensation data.

[0176] S5: Based on the spot offset data and the individual error fingerprint of the reflector, generate azimuth compensation data and pitch compensation data.

[0177] Extract the horizontal and vertical offset components from the spot offset data; Based on the mapping relationship between the preset spot offset and the dual-axis angle adjustment, the horizontal offset component and the vertical offset component are converted to generate initial azimuth compensation data and initial pitch compensation data. When the azimuth axis adjustment and the pitch axis adjustment have a coupled effect on the horizontal offset component and the vertical offset component, the initial azimuth compensation data and the initial pitch compensation data are generated based on the pre-calibrated two-dimensional joint mapping relationship.

[0178] Based on the current attitude range of the reflector, the historical spot offset features of the corresponding attitude range are read from the individual error fingerprint of the reflector. When the azimuth axis direction switching state indicator is in the reversal state, read the azimuth axis angle domain hysteresis response gap and azimuth axis hysteresis state intensity corresponding to the driving direction after the azimuth axis reversal, and generate the azimuth axis hysteresis correction amount; when the azimuth axis direction switching state indicator is in the non-reversal state, set the azimuth axis hysteresis correction amount to 0. When the pitch axis direction switching state indicator is in the reversal state, read the pitch axis angle domain hysteresis response gap and pitch axis hysteresis state intensity corresponding to the driving direction after pitch axis reversal, and generate the pitch axis hysteresis correction amount; when the pitch axis direction switching state indicator is in the non-reversal state, set the pitch axis hysteresis correction amount to 0. Based on the historical spot offset features, azimuth and pitch historical offset corrections are generated. The historical spot offset features include the historical horizontal offset mean, historical vertical offset mean, historical horizontal offset dispersion, historical vertical offset dispersion, and the number of valid historical data. The azimuth axis backlash correction and the azimuth historical offset correction are fused together to generate the azimuth fingerprint correction, and the pitch axis backlash correction and the pitch historical offset correction are fused together to generate the pitch fingerprint correction. The azimuth fingerprint correction amount is used as the incremental correction amount of the initial azimuth compensation data to generate azimuth compensation data. The pitch fingerprint correction amount is used as the incremental correction amount of the initial pitch angle compensation data to generate pitch angle compensation data.

[0179] In Example 1, step S5 is used to generate azimuth compensation data and pitch compensation data based on the spot offset data of the current control cycle and the individual error fingerprint of the reflector generated in step S4.

[0180] Step S5 includes an initial compensation generation stage and a fingerprint correction compensation stage. The initial compensation generation stage is used to convert the current spot offset data into initial angle compensation data; the fingerprint correction compensation stage is used to incrementally correct the initial angle compensation data based on the driving backlash characteristics and historical spot offset characteristics corresponding to the current attitude range of the reflector.

[0181] Phase 1: Initial compensation generation.

[0182] The purpose of initial compensation generation is to convert the current spot offset data obtained in step S3 into initial azimuth compensation data and initial pitch compensation data with angular dimensions and a clear compensation direction.

[0183] Furthermore, the horizontal offset component is extracted from the spot offset data. and vertical offset components .

[0184] Based on the mapping relationship between preset spot offset and dual-axis angle adjustment, the horizontal offset component and the vertical offset component are converted into initial azimuth compensation data and initial pitch compensation data.

[0185] The mapping relationship is obtained through calibration during the system installation or debugging phase. During the calibration process, the azimuth and pitch axes are controlled to generate preset small angle adjustments, and the corresponding horizontal and vertical spot displacements on the receiver surface are collected. The mapping parameters are determined based on the correspondence between the angle adjustment amount and the two-dimensional spot displacement amount.

[0186] When the mirror installation relationship causes azimuth axis adjustment to primarily cause horizontal spot movement and pitch axis adjustment to primarily cause vertical spot movement, a separate mapping method can be used: ; ; In the formula, This represents the initial azimuth compensation data. This represents the initial pitch angle compensation data; This represents the mapping coefficient between the horizontal spot offset and the azimuth compensation. This represents the mapping coefficient between the vertical spot offset and the pitch angle compensation.

[0187] When the spot offset is expressed in millimeters and the angle compensation is expressed in degrees... and The unit is degrees per millimeter; when the spot offset is expressed in meters, the corresponding unit is degrees per meter.

[0188] The negative sign in the formula indicates that the direction of the light spot movement generated by the compensation is opposite to the direction of the current light spot offset, so the secondary judgment of the compensation direction is no longer performed separately.

[0189] When the azimuth axis adjustment affects both the horizontal and vertical spot positions, or the pitch axis adjustment affects both the horizontal and vertical spot positions, the two-dimensional joint mapping relationship obtained from calibration is used to jointly process the horizontal and vertical offset components, and simultaneously generate initial azimuth compensation data and initial pitch compensation data.

[0190] The two-dimensional joint mapping relationship is used to characterize the coupled effect of azimuth axis adjustment on vertical spot displacement and the coupled effect of pitch axis adjustment on horizontal spot displacement. The separate mapping method is a simplified implementation when the two-axis coupling effect can be ignored.

[0191] Phase Two: Fingerprint Correction and Compensation.

[0192] The purpose of fingerprint correction compensation is to use the driving backlash features and historical spot offset features in the individual error fingerprint of the reflector that correspond to the current attitude range of the reflector to correct the initial angle compensation data, so that the final compensation data simultaneously considers the current spot offset and the long-term error law of the reflector in the current attitude range.

[0193] Furthermore, based on the current attitude range of the reflector, the historical spot offset features for the corresponding attitude range are read from the individual error fingerprint of the reflector. The historical spot offset features include the historical mean horizontal offset, the historical mean vertical offset, the historical horizontal offset dispersion, the historical vertical offset dispersion, and the number of valid historical data.

[0194] When the azimuth axis direction switching status indicator is in the reversal state, the azimuth axis angle domain hysteresis response gap and azimuth axis hysteresis status intensity corresponding to the driving direction after the azimuth axis reversal are read from the individual error fingerprint of the reflector; when the azimuth axis direction switching status indicator is in the non-reversal state, the azimuth axis direction related hysteresis features are not read, and the azimuth axis hysteresis correction amount is set to 0.

[0195] When the pitch axis direction switching status indicator is in the reversal state, the pitch axis angle domain hysteresis response gap and pitch axis hysteresis state intensity corresponding to the driving direction after pitch axis reversal are read from the individual error fingerprint of the reflector; when the pitch axis direction switching status indicator is in the non-reversal state, the pitch axis direction related hysteresis features are not read, and the pitch axis hysteresis correction amount is set to 0.

[0196] The azimuth-axis hysteresis response gap serves as the fundamental azimuth-axis hysteresis correction quantity in the azimuth domain, and the pitch-axis hysteresis response gap serves as the fundamental pitch-axis hysteresis correction quantity in the azimuth domain. The hysteresis state strength is used to determine the degree to which the corresponding hysteresis response gap participates in compensation, without changing the physical dimensions of the hysteresis response gap in the azimuth domain.

[0197] When the corresponding drive shaft does not change direction, the hysteresis correction amount of the drive shaft is set to 0; when the corresponding drive shaft changes direction, the participation ratio of the hysteresis response gap in the angle domain is determined according to the hysteresis state intensity.

[0198] A smaller participation ratio is used when the hysteresis intensity is low, and a larger participation ratio is used when the hysteresis intensity is high. The specific participation ratio is determined through the drive mechanism commutation calibration test.

[0199] Therefore, the azimuth axis hysteresis correction is generated based on the azimuth axis angle domain hysteresis response gap and the corresponding participation ratio, and the pitch axis hysteresis correction is generated based on the pitch axis angle domain hysteresis response gap and the corresponding participation ratio.

[0200] Furthermore, using the same mapping relationship as in the initial compensation generation stage, the historical average horizontal offset and historical average vertical offset are converted into azimuth historical offset corrections and pitch historical offset corrections.

[0201] The historical offset correction is used to characterize the long-term systematic offset trend of the mirror within the current attitude range, and its correction direction is opposite to the corresponding historical spot offset direction.

[0202] The historical offset correction is used as an incremental correction for the initial angle compensation data, and is not used as an independent and complete compensation with the same amplitude as the current initial angle compensation data.

[0203] The degree of participation of historical offset correction is determined based on the amount of valid historical data and the dispersion of historical offset.

[0204] When the amount of valid historical data is small or the historical offset dispersion is large, reduce the participation of historical offset correction; when the amount of valid historical data reaches the preset requirement and the historical offset dispersion is small, appropriately increase the participation of historical offset correction.

[0205] When the current spot offset data is valid, the initial angle compensation data generated by the current spot offset remains dominant, and the participation weight of the historical offset correction amount does not exceed a preset historical correction upper limit. For example, the preset historical correction upper limit can be set to 0.3.

[0206] By limiting the involvement of historical offset corrections, we can prevent long-term systematic offsets already included in the current spot offset from being fully compensated again by historical offset corrections.

[0207] Furthermore, the azimuth axis hysteresis correction and the azimuth historical offset correction are fused to obtain the azimuth fingerprint correction; the pitch axis hysteresis correction and the pitch historical offset correction are fused to obtain the pitch fingerprint correction.

[0208] Incremental corrections are made to the initial azimuth compensation data based on the azimuth fingerprint correction to generate azimuth compensation data before amplitude limiting; incremental corrections are also made to the initial pitch compensation data based on the pitch fingerprint correction to generate pitch compensation data before amplitude limiting.

[0209] In one specific implementation, it can be represented as: ; ; In the formula, This indicates the orientation fingerprint correction amount. Indicates the pitch fingerprint correction amount; This represents the azimuth compensation data before amplitude limiting. This indicates the pitch angle compensation data before limiting.

[0210] When the compensation data before the limit exceeds the maximum allowable compensation angle for a single control cycle of the corresponding drive shaft, it is limited to the maximum allowable compensation angle in the corresponding direction.

[0211] Furthermore, amplitude limiting processing is performed on the azimuth compensation data and the pitch compensation data before amplitude limiting.

[0212] When the absolute value of the azimuth compensation data before the amplitude limit is greater than the maximum azimuth compensation amount allowed in a single control cycle, the positive and negative directions of the azimuth compensation data before the amplitude limit are kept unchanged, and its absolute value is limited to the maximum azimuth compensation amount; when the absolute value of the azimuth compensation data before the amplitude limit is not greater than the maximum azimuth compensation amount, the azimuth compensation data before the amplitude limit is directly used as the azimuth compensation data.

[0213] When the absolute value of the pitch angle compensation data before limiting is greater than the maximum pitch angle compensation allowed in a single control cycle, the positive and negative directions of the pitch angle compensation data before limiting remain unchanged, and its absolute value is limited to the maximum pitch angle compensation. When the absolute value of the pitch angle compensation data before limiting is not greater than the maximum pitch angle compensation, the pitch angle compensation data before limiting is directly used as the pitch angle compensation data.

[0214] Both the maximum azimuth compensation and the maximum pitch compensation are positive values, and are preset based on the response capability of the reflector drive mechanism, the control cycle duration, and the safe light-receiving range of the receiver. For example, the maximum azimuth compensation and the maximum pitch compensation can be set to 0.05 degrees to 0.2 degrees.

[0215] When the current spot offset data is marked as invalid data, initial angle compensation data based on the current spot offset is not generated.

[0216] If the historical offset statistics corresponding to the current attitude range of the reflector meet the preset reliability requirements, conservative compensation data can be generated based solely on the historical offset correction amount and the backlash correction amount corresponding to the direction switching state. If the historical offset statistics do not meet the preset reliability requirements, the azimuth compensation data and pitch compensation data of the current control cycle will be set to 0, so that the reflector will only perform the tracking action according to the theoretical tracking pose.

[0217] Through the two stages described above, the azimuth compensation data and pitch compensation data generated in step S5 can not only enable the current actual light spot to return to the target light spot position, but also incrementally correct the dual-axis drive hysteresis and long-term systematic offset within the current attitude range of the reflector.

[0218] S6: Based on the theoretical light-tracking pose of the reflector, the azimuth compensation data, and the pitch compensation data, generate a light-tracking control command for the reflector, and control the reflector to perform the light-tracking action according to the light-tracking control command.

[0219] The theoretical azimuth and theoretical pitch angles of the reflector are obtained from the theoretical light-tracking pose of the reflector. The theoretical azimuth angle of the reflector is superimposed with the azimuth compensation data to generate the target azimuth angle, and the theoretical elevation angle of the reflector is superimposed with the elevation compensation data to generate the target elevation angle; Angle continuity processing is performed on the target azimuth angle, and the target azimuth angle and the target pitch angle are limited to the allowable movement range of the corresponding drive shaft; Generate a reflector tracking control command based on the processed target azimuth and target elevation angles; The reflector tracking control command is sent to the azimuth axis drive motor and the pitch axis drive motor respectively to control the reflector to perform the tracking action.

[0220] Step S6 superimposes the theoretical tracking pose of the reflector with the compensation data to form the target azimuth and target pitch angles, and converts the target angles into reflector tracking control commands that can be executed by the azimuth axis drive motor and the pitch axis drive motor.

[0221] Furthermore, the theoretical azimuth angle of the mirror is obtained from the theoretical light-tracking pose of the mirror. And the theoretical pitch angle of the reflecting mirror And obtain the azimuth compensation data generated in step S5. Pitch angle compensation data .

[0222] The azimuth and pitch compensation data already include current spot offset compensation, drive hysteresis correction, and historical offset correction, so step S6 will not repeat the calculation of spot offset correction and fingerprint correction.

[0223] The target azimuth is generated by superimposing the theoretical azimuth angle of the reflector with the azimuth compensation data; the target elevation angle is generated by superimposing the theoretical elevation angle of the reflector with the elevation compensation data. ; ; In the formula, Indicates the target azimuth angle for the current control cycle. This indicates the target pitch angle for the current control cycle.

[0224] The target azimuth and target pitch angles include both the basic attitude required for theoretical beam tracking and the compensation amount used to correct the current beam offset and individual mirror errors.

[0225] Furthermore, the target azimuth and elevation angles are standardized and the range of motion is verified.

[0226] When the azimuth axis is allowed to rotate continuously, the target azimuth angle is periodically normalized to keep it within the range of 0 to 360 degrees. For example, if the target azimuth angle is 361 degrees, it is normalized to 1 degree; if the target azimuth angle is -2 degrees, it is normalized to 358 degrees.

[0227] When there are physical mechanical limits on the azimuth axis, the target azimuth angle, after periodic standardization, will be further limited to the actual mechanically permissible range of motion of the azimuth axis.

[0228] The target pitch angle does not have periodic equivalence; therefore, the target pitch angle is directly limited to the actual mechanically permissible range of the pitch axis. When the target pitch angle exceeds this range, it is corrected to the permissible boundary value closest to the original target pitch angle.

[0229] When the target azimuth angle has an equivalent angle representation that crosses the boundaries of 0 degrees and 360 degrees, the target azimuth angle representation corresponding to the shortest rotation path is selected based on the current azimuth angle of the reflector.

[0230] For example, when the current azimuth angle of the reflector is 359 degrees and the standardized target azimuth angle is 2 degrees, the azimuth axis rotation path is determined by rotating 3 degrees in the forward direction, not by rotating 357 degrees in the reverse direction.

[0231] Furthermore, a reflector tracking control command is generated based on the processed target azimuth and target elevation angles.

[0232] The reflector tracking control commands include the target position on the azimuth axis, the target position on the pitch axis, the reflector identifier, the control cycle identifier, and the execution time identifier.

[0233] The target position on the azimuth axis is determined by the processed target azimuth angle, and the target position on the pitch axis is determined by the processed target pitch angle; the reflector identifier is used to identify the reflector corresponding to the control command; the control cycle identifier is used to distinguish the control commands generated in different control cycles; the execution time identifier is used to limit the effective execution time of the control command.

[0234] The controller encapsulates the above data into reflector tracking control commands according to the communication protocol of the azimuth axis drive motor and the pitch axis drive motor.

[0235] When the mirror tracking control command received by the drive controller has exceeded the preset effective time window, the control command is discarded to avoid the lagging control command changing the current mirror attitude.

[0236] Furthermore, the reflector tracking control commands are sent to the azimuth axis drive motor and the pitch axis drive motor, respectively.

[0237] The azimuth axis drive motor drives the reflector to adjust to the target azimuth angle according to the target position on the azimuth axis, and the pitch axis drive motor drives the reflector to adjust to the target pitch angle according to the target position on the pitch axis.

[0238] The azimuth axis drive motor and the pitch axis drive motor execute corresponding control commands according to the preset position closed-loop control mode inside the driver, thereby controlling the reflector to complete the light-tracking action of the current control cycle.

[0239] After the reflector performs the light-tracking action, the actual position feedback of the azimuth axis, the actual position feedback of the pitch axis, the azimuth motor current data, and the pitch motor current data are reacquired in step S1 of the next control cycle; the image data of the receiver area is reacquired in step S3 of the next control cycle.

[0240] This forms a closed-loop tracking control process, which includes basic data acquisition, generation of theoretical tracking pose of the reflector, identification of actual spot position, construction of individual error fingerprint of the reflector, generation of azimuth and pitch compensation data, issuance of tracking control commands to the reflector, and feedback updates in the next control cycle.

[0241] Through the above steps, this embodiment can implement individualized light-tracking control for each reflector based on the light-tracking results of geometric optics theory, combined with the actual light spot offset at the receiver end, the hysteresis of the dual-axis drive of the reflector, and the historical light spot offset patterns related to the attitude range. This allows the actual reflected light spot of the reflector to gradually return to the preset target center, improving the accuracy, stability, and adaptability of the light-tracking control of the solar thermal reflector.

[0242] Example 2 is the second embodiment of the present invention, which differs from the previous embodiment in that: If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art or the current technical solution, can be embodied in the form of a software product. This current computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0243] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0244] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0245] Example 3, referring to Figure 2 As an embodiment of the present invention, a solar thermal reflector tracking control system is provided, which includes a data acquisition module, a pose generation module, a spot recognition module, a fingerprint generation module, a compensation generation module, and an instruction control module.

[0246] Data acquisition module: acquires solar position data, receiver target position data, reflector current attitude data, dual-axis drive feedback data, drive direction change data, and historical spot offset data; Pose generation module: Based on the sun position data, the receiver target position data, and the reflector current attitude data, generate the theoretical light-tracking pose of the reflector; Spot recognition module: Collects image data of the receiver area, determines the actual spot position corresponding to the reflector based on the image data of the receiver area, and generates spot offset data based on the actual spot position and the target position data of the receiver; Fingerprint generation module: Based on the dual-axis drive feedback data, the drive direction change data, and the historical spot offset data, generate a mirror individual error fingerprint that characterizes the dual-axis drive backlash and historical spot offset features of the mirror; Compensation generation module: Based on the spot offset data and the individual error fingerprint of the reflector, it generates azimuth compensation data and pitch compensation data; Command control module: Based on the theoretical light-tracking posture of the reflector, the azimuth compensation data, and the pitch compensation data, it generates a light-tracking control command for the reflector and controls the reflector to perform light-tracking actions according to the light-tracking control command.

[0247] Example 4 is an embodiment of the present invention, which provides a method for controlling the light tracking of a solar thermal reflector. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation / comparative experiments.

[0248] To verify the light-tracking control effect of the method of the present invention in a real-world operating scenario, a three-hour continuous test was conducted at a tower-type solar thermal test power station. The test power station is located at 37.4 degrees North latitude and 121.8 degrees East longitude, at an altitude of 56 meters. The receiver is installed at the top of the tower, 65 meters above the ground. The receiver's heat-absorbing area is rectangular, with an effective light-receiving width of 2.0 meters and a height of 3.0 meters. The preset target center's two-dimensional coordinates in the receiver surface coordinate system are 0.0 mm horizontally and 0.0 mm vertically. The receiver target point's three-dimensional spatial coordinates in the mirror field coordinate system are 12.3 meters east, 58.7 meters north, and 65.0 meters vertically upwards.

[0249] Six glass-based reflectors, numbered M01 to M06, were selected as test objects in the mirror field. Each reflector has an effective reflective area of ​​4.2 square meters. Both the azimuth and pitch axes are driven by servo motors and worm gear reducers. Reflectors M01 and M02 were operated using a conventional astronomical algorithm with open-loop tracking and single spot correction every ten minutes. During operation, no individual error fingerprints were built for each reflector, and no drive hysteresis feature identification was performed. During spot correction, compensation data was generated based solely on the spot offset detected in the current control cycle and directly superimposed onto the theoretical tracking pose, serving as a comparative example. Reflectors M03 to M06 were operated using the method of this invention, serving as the embodiment group. All reflectors were equipped with the same azimuth and pitch encoders. The azimuth encoder resolution was 0.0003 degrees, the pitch encoder resolution was 0.0002 degrees, and the motor current sampling accuracy was 0.01 amperes.

[0250] Before testing, complete the following preparations: Calibrate the clock using the satellite timing module deployed in the mirror field control system, and write the mirror field geographic location parameters into the control system configuration storage area. Install an industrial camera on the receiver side, with the camera's field of view covering the receiver's heat-absorbing area and a preset allowable offset range of 0.4 meters beyond the boundary. The camera resolution is 2048 pixels by 1536 pixels, and after calibration, a single pixel corresponds to a physical size of approximately 1.3 mm on the receiver surface. Select 12 calibration markers with known physical coordinates on the receiver's heat-absorbing area surface, acquire calibration images containing all calibration markers, and establish a homography mapping matrix between the image pixel coordinates and the receiver surface physical coordinates as the receiver image coordinate calibration relationship.

[0251] The control cycle duration is set to 1 second, and the preset sampling interval for dual-axis drive feedback data is set to 50 milliseconds. Within each control cycle, 20 sets of azimuth axis actual position feedback, pitch axis actual position feedback, azimuth motor current data, and pitch motor current data are acquired. Small-amplitude angle scan calibrations are performed on the azimuth and pitch axes respectively, with a scan angle range of -0.1 degrees to +0.1 degrees and a step size of 0.005 degrees. The horizontal and vertical displacements of the receiver surface spot center are recorded at each angle step size. The mapping coefficients between the preset spot offsets of each reflector in multiple attitude ranges and the dual-axis angle adjustments are determined through linear fitting.

[0252] In the individual error fingerprint of the reflectors in the example group, the hysteresis features and historical offset statistics for each attitude interval were initialized to an unavailable state before the test began. The attitude interval division step size was set to 2 degrees for azimuth and 2 degrees for pitch, and the preset storage limit for historical spot offset data was set to 300 records. The spot correction of the comparative reflector adopted the same spot recognition method as the example group, including the brightness weighted centroid method to determine the center pixel coordinates of the spot and the same effective offset threshold.

[0253] The test period was from 8:30 AM to 11:30 AM local time. During this period, the solar azimuth angle changed from 109.3 degrees to 152.7 degrees, and the solar altitude angle increased from 23.8 degrees to 46.5 degrees. The weather was clear, and the wind speed remained between 2.1 meters and 4.3 meters per second. Every hour, a specific reflector was artificially blocked for three minutes. The blocking material was an opaque light-blocking plate that completely covered the reflector surface to verify the light-tracking control capability under invalid light spot data conditions. The artificial blocking periods were 9:15 AM to 9:18 AM, 10:15 AM to 10:18 AM, and 11:15 AM to 11:18 AM, with a different reflector blocked each time.

[0254] After the test, extract the spot offset data, drive hysteresis characteristic data, historical offset statistics data and compensation data of the six mirrors in all control cycles during the three-hour test period. The average spot offset is the arithmetic mean of the spot offset vector magnitudes of all effective control cycles. The standard deviation of the spot offset is the sample standard deviation of the spot offset vector magnitudes of the effective control cycles. The mean spot offset of the first cycle after a direction switch is the arithmetic mean of the spot offset vector magnitudes of the first control cycle after each direction switch within the test period. The reversal recovery time is the average number of control cycles required for the spot offset to recover to within 1.2 times the average spot offset of the three control cycles before the switch. The effective spot recognition rate is the ratio of the number of control cycles with effective spot recognition to the total number of control cycles. The time required for the historical offset confidence to reach 0.5 is the cumulative time from the start of the test to the first time the historical offset confidence in the corresponding attitude interval reaches 0.5. The spot offset during the invalid spot period is the arithmetic mean of the spot offset vector magnitudes of the control cycles in which the spot data is marked as invalid. The number of times the compensation data exceeds the limit is the cumulative number of control cycles in the test period in which the compensation data exceeds the maximum allowable compensation amount per cycle and is subject to amplitude limiting. The maximum compensation amount per cycle is the global maximum value among the absolute values ​​of the azimuth compensation amount and the pitch compensation amount within the test period. The experimental data were compiled into a summary table according to the above statistical method, as shown in Table 1.

[0255] Table 1: Reference Table for Experimental Data

[0256] As shown in the table above, the average spot offset over three hours for reflectors M03 to M06 in the example group ranges from 1.7 mm to 2.4 mm, while the average spot offsets for reflectors M01 and M02 in the comparative examples are 7.8 mm and 8.4 mm, respectively. The average spot offset of the example group is approximately 0.21 to 0.29 of that in the comparative examples, indicating that the method of the present invention, by compensating for the current spot offset and the individual error fingerprint of the reflectors, can keep the reflected spot continuously closer to the target center. Using a receiver heat absorption area of ​​1.0 m half-width and 1.5 m half-height as a reference, the spot offset of the example group is only 0.17 to 0.24% of the half-width of the heat absorption area, which is within the acceptable high-precision range for the receiver surface.

[0257] The standard deviation of the light spot offset reflects the stability of light tracking. The standard deviations of comparative examples M01 and M02 are 3.9 mm and 4.7 mm, respectively, while the standard deviation of the example group ranges from 1.5 mm to 2.2 mm. The standard deviation of the comparative examples is approximately 1.8 to 3.1 times that of the example group, indicating that the light spot position of the comparative mirrors fluctuates significantly during the test. The example group, by utilizing historical offset dispersion to control the participation of historical offset correction, automatically reduces the weight of historical correction when the reliability of historical data is low. This makes the compensation process less prone to over-adjustment due to short-term disturbances, thus improving the stability of light tracking.

[0258] The average spot offset in the first control cycle after direction switching reflects the degree of hysteresis impact during the commutation of the drive mechanism. Comparative examples M01 and M02 showed average spot offsets as high as 16.3 mm and 18.7 mm in the first cycle after commutation, respectively, while the example group showed values ​​of only 2.8 mm to 4.6 mm. The example group identified drive hysteresis characteristics based on direction switching status, position change data, and current change data in step S4, and extracted the corresponding angular domain hysteresis response gap according to the drive direction after commutation in step S5 to generate a hysteresis correction amount. This enabled the reflector to obtain targeted hysteresis compensation at the moment of commutation, reducing the spot offset in the first cycle after commutation by approximately 71% to 83% compared to the comparative examples, effectively avoiding the impact of large spot offsets in the early stages of commutation on the heat flow uniformity of the receiver.

[0259] The commutation recovery time is measured in control cycles. The comparative example requires 8 to 9 control cycles to restore the spot offset after commutation to the average level before commutation, while the example group only requires 2 to 3 control cycles. Assuming a control cycle length of 1 second, the commutation recovery time for the example group is approximately 2 to 3 seconds, and for the comparative example, it is approximately 8 to 9 seconds. This shorter recovery time allows the reflector to return to normal tracking mode more quickly during frequent small-angle adjustments, reducing the accumulation of localized heat flux fluctuations in the receiver caused by commutation hysteresis.

[0260] Regarding the effective spot recognition rate, the example group ranged from 0.936 to 0.943, while the comparative group ranged from 0.921 to 0.927, showing little difference and all falling between 0.92 and 0.95. This result demonstrates that the method of the present invention does not rely on higher-specification image acquisition equipment and can successfully complete spot recognition under conventional cameras and the same environmental conditions. The slightly higher effective spot recognition rate in the example group is mainly due to the temporal correlation recognition method, which reduces the probability of misidentification when multiple mirror spots overlap through perturbation coding and correlation matching of brightness changes between image frames, rather than relying on improvements in image acquisition hardware performance.

[0261] The time required for the historical offset confidence level to reach 0.5 reflects the runtime required for the individual error fingerprint of a reflector to go from initialization to having a certain reference value. M04 and M06 require only 38 minutes and 35 minutes respectively, while M03 and M05 require 43 minutes and 51 minutes respectively. The difference in time required for each reflector is due to the different number of attitude intervals traversed by different reflectors during the test and the different amount of valid historical data accumulated within each attitude interval. For example, M06 traversed multiple attitude intervals when the solar azimuth angle changed rapidly in the initial period, and historical data exceeding the minimum sample size requirement was accumulated quickly within each interval, thus achieving the confidence level in a shorter time. Comparative studies do not construct individual error fingerprints for reflectors, therefore this indicator is not applicable. The above data shows that the method of this invention can generate a meaningful individual error fingerprint within about one hour after the reflector is put into operation, and can quickly adapt to differences in installation deviations and transmission wear of different reflectors, without requiring a pre-calibration process that can last for several days or weeks.

[0262] The spot offset during invalid spot periods reflects the reflector's tracking performance when spot recognition fails due to cloud cover, human-caused obstruction, or image overexposure. Comparative examples M01 and M02 showed spot offsets of 15.4 mm and 16.1 mm during invalid spot periods, respectively, while example groups M03 to M06 showed offsets between 4.3 mm and 6.2 mm. When spot recognition fails, the example groups do not directly use the spot offset data from the previous cycle. Instead, they generate conservative compensation data based on historical offset features and driving hysteresis features in the reflector's individual error fingerprint, ensuring the spot remains near the target position. The spot offset during invalid spot periods in the example groups is approximately 2.5 to 2.6 times the corresponding average spot offset, while the ratio for the comparative examples is approximately 2.0 times. However, the absolute offset in the example groups is significantly lower than that in the comparative examples, indicating that the method of this invention can effectively maintain tracking accuracy even during short-term interruptions in spot data.

[0263] The number of times compensation data exceeded limits was counted, representing the number of times compensation data generated during the test period exceeded the maximum allowable compensation amount per cycle and was therefore subject to throttling. Comparative examples M01 and M02 exceeded limits 47 and 52 times respectively. Based on 10,800 control cycles over three hours, the exceedance rate was approximately 0.44% to 0.48%. Example groups M03 to M06 had between 0 and 2 exceedances, with an exceedance rate not exceeding 0.02%. By using historical offset corrections as incremental corrections instead of independent complete compensation amounts, and limiting the weight of historical offset corrections to no more than a preset historical correction upper limit, the example groups effectively reduced the occurrence of excessively large accumulated compensation amounts triggering throttling, ensuring that the compensation data remained within a reasonable range for the vast majority of control cycles.

[0264] The maximum compensation amount per single cycle reached 0.22 degrees and 0.24 degrees in the comparative examples, and between 0.15 degrees and 0.19 degrees in the example group. The smaller maximum compensation amount per single cycle in the example group indicates that the method of the present invention, by combining the incremental correction method of historical offset patterns, can achieve continuous optimization of the spot position without relying on a large single compensation, and has less impact on the drive mechanism during long-term operation.

[0265] In summary, the reflectors in the example group outperform the comparative reflectors in terms of average spot offset, tracking stability, commutation response speed, ability to maintain the spot during periods of invalidity, and the rationality of compensation data. The method of this invention constructs an individual error fingerprint of the reflector and saves drive hysteresis features and historical offset features according to the attitude range and the driving direction after commutation, respectively. When generating compensation data, it takes into account both the real-time feedback of the current spot offset and the long-term error patterns of the individual reflectors, achieving coordinated compensation between the theoretical tracking pose, the current spot offset, and the long-term individual errors. This improves the positioning accuracy of the reflected spot, the tracking stability, and the uniformity of light received by the receiver.

[0266] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for controlling the light tracking of a solar thermal reflector, characterized in that, include: S1: Acquire solar position data, receiver target position data, reflector current attitude data, dual-axis drive feedback data, drive direction change data, and historical spot offset data; S2: Based on the solar position data, the receiver target position data, and the reflector current attitude data, generate the theoretical light-tracking pose of the reflector; S3: Acquire image data of the receiver area, determine the actual spot position corresponding to the reflector based on the image data of the receiver area, and generate spot offset data based on the actual spot position and the receiver target position data; S4: Based on the dual-axis drive feedback data, the drive direction change data, and the historical spot offset data, generate a mirror individual error fingerprint characterizing the dual-axis drive backlash and historical spot offset features of the mirror; S5: Based on the spot offset data and the individual error fingerprint of the reflector, generate azimuth compensation data and pitch compensation data; S6: Based on the theoretical light-tracking pose of the reflector, the azimuth compensation data, and the pitch compensation data, generate a light-tracking control command for the reflector, and control the reflector to perform the light-tracking action according to the light-tracking control command.

2. The solar thermal reflector tracking control method as described in claim 1, characterized in that, Step S1 specifically includes: Acquire current time data and mirror field geographical location data, and calculate solar position data based on the current time data and the mirror field geographical location data; Acquire receiver target location data, which includes the three-dimensional spatial coordinates of the receiver target point in the mirror field coordinate system and the two-dimensional coordinates of the preset target center in the receiver heat absorption area in the receiver surface coordinate system; Within each control cycle, dual-axis drive feedback data is acquired according to a preset sampling interval. The dual-axis drive feedback data includes azimuth axis actual position feedback, pitch axis actual position feedback, azimuth motor current data, and pitch motor current data. Based on the actual position feedback of the azimuth axis, the actual position feedback of the pitch axis, and the preset attitude calibration parameters, the current attitude data of the reflector is generated, which includes the current azimuth angle and the current pitch angle of the reflector. Based on the azimuth axis target position command and pitch axis target position command in the already issued and stored reflector tracking control command, the drive direction change data is obtained. The drive direction change data includes azimuth axis direction switching state, pitch axis direction switching state, drive direction after azimuth axis reversal and drive direction after pitch axis reversal. Acquire historical spot offset data up to the previous control cycle. The historical spot offset data includes effective historical spot offset records and attitude interval identifiers and control cycle identifiers corresponding to each effective historical spot offset record. The effective historical spot offset records include historical horizontal offset components and historical vertical offset components.

3. The solar thermal reflector tracking control method as described in claim 2, characterized in that, Step S2 specifically includes: Construct a solar incidence direction vector based on the solar position data; Based on the three-dimensional spatial coordinates of the receiver target point in the mirror field coordinate system, a target reflection direction vector is constructed in the local coordinate system of the reflector. Based on the solar incident direction vector and the target reflection direction vector, the theoretical normal vector of the reflector is calculated according to the law of reflection. The theoretical normal vector of the reflector is converted into the theoretical azimuth and theoretical elevation angles of the reflector. Based on the current attitude data of the reflector, angular continuity processing is performed on the theoretical azimuth angle of the reflector, and the theoretical azimuth angle and the theoretical pitch angle of the reflector are limited to the allowable movement range of the corresponding drive shaft. The processed theoretical azimuth and theoretical elevation angles of the reflector are used as the theoretical ray-tracking pose of the reflector.

4. The solar thermal reflector tracking control method as described in claim 3, characterized in that, Step S3 specifically includes: Image data of the receiver area is acquired by an image acquisition device deployed on the receiver side; The image data of the receiver area is processed for spot recognition to extract the spot outline corresponding to the current reflector and determine the center pixel coordinates of the spot; Based on the receiver image coordinate calibration relationship, the center pixel coordinates of the light spot are converted into the actual light spot position in the receiver surface coordinate system; The two-dimensional coordinates of the preset target center in the receiver surface coordinate system are used as the target spot position; Based on the actual spot position and the target spot position, spot offset data containing horizontal and vertical offset components is generated.

5. The solar thermal reflector tracking control method as described in claim 4, characterized in that, Step S4 specifically includes: Based on the actual azimuth position feedback before the azimuth target position command of the previous control cycle is executed and the actual azimuth position feedback after the azimuth target position command is executed, azimuth position change data is generated. Based on the pitch axis actual position feedback before the pitch axis target position command of the previous control cycle is executed and the pitch axis actual position feedback after the pitch axis target position command is executed, pitch axis position change data is generated. Azimuth axis current variation data is generated based on the azimuth motor current data within adjacent control cycles, and pitch axis current variation data is generated based on the pitch motor current data within adjacent control cycles. Based on the azimuth axis direction switching state, the azimuth axis driving direction after reversal, the azimuth axis position change data, and the azimuth axis current change data, an azimuth axis driving hysteresis characteristic is generated. Based on the pitch axis direction switching state, the pitch axis drive direction after reversal, the pitch axis position change data, and the pitch axis current change data, pitch axis drive hysteresis characteristics are generated. According to the driving direction after commutation, the azimuth axis drive backlash characteristics and the pitch axis drive backlash characteristics are saved respectively to generate dual-axis drive backlash characteristics.

6. The solar thermal reflector tracking control method as described in claim 5, characterized in that, Step S4 also includes: The current attitude range of the reflector is determined based on the current azimuth angle and the current pitch angle of the reflector. From the historical spot offset data formed up to the previous control cycle, extract the historical horizontal offset component and the historical vertical offset component corresponding to the current attitude range of the reflector. Based on the historical horizontal offset components and the historical vertical offset components, the historical horizontal offset mean, historical vertical offset mean, historical horizontal offset dispersion, historical vertical offset dispersion, and number of valid historical data are determined. The dual-axis drive backlash feature, the historical average horizontal offset, the historical average vertical offset, the historical horizontal offset dispersion, the historical vertical offset dispersion, and the number of valid historical data are correlated. When there is no fingerprint data corresponding to the current attitude range of the reflector in the individual error fingerprint of the reflector, the corresponding fingerprint data is generated based on the correlation result. When there is already fingerprint data corresponding to the current attitude range of the reflector in the individual error fingerprint of the reflector, the corresponding fingerprint data is updated based on the correlation result. The spot offset data generated and marked as valid in the current control cycle is written into the historical spot offset data. The written historical spot offset data is used to update the individual error fingerprint of the reflector in the next control cycle.

7. The solar thermal reflector tracking control method as described in claim 6, characterized in that, Step S5 specifically includes: Extract the horizontal and vertical offset components from the spot offset data; Based on the mapping relationship between the preset spot offset and the dual-axis angle adjustment, the horizontal offset component and the vertical offset component are converted to generate initial azimuth compensation data and initial pitch compensation data. When the azimuth axis adjustment and the pitch axis adjustment have a coupled effect on the horizontal offset component and the vertical offset component, the initial azimuth compensation data and the initial pitch compensation data are generated based on the pre-calibrated two-dimensional joint mapping relationship.

8. The solar thermal reflector tracking control method as described in claim 7, characterized in that, Step S5 also includes: Based on the current attitude range of the reflector, the historical spot offset features of the corresponding attitude range are read from the individual error fingerprint of the reflector. When the azimuth axis direction switching state indicator is in the reversal state, read the azimuth axis angle domain hysteresis response gap and azimuth axis hysteresis state intensity corresponding to the driving direction after the azimuth axis reversal, and generate the azimuth axis hysteresis correction amount; when the azimuth axis direction switching state indicator is in the non-reversal state, set the azimuth axis hysteresis correction amount to 0. When the pitch axis direction switching state indicator is in the reversal state, read the pitch axis angle domain hysteresis response gap and pitch axis hysteresis state intensity corresponding to the driving direction after pitch axis reversal, and generate the pitch axis hysteresis correction amount; when the pitch axis direction switching state indicator is in the non-reversal state, set the pitch axis hysteresis correction amount to 0. Based on the historical spot offset features, azimuth and pitch historical offset corrections are generated. The historical spot offset features include the historical horizontal offset mean, historical vertical offset mean, historical horizontal offset dispersion, historical vertical offset dispersion, and the number of valid historical data. The azimuth axis backlash correction and the azimuth historical offset correction are fused together to generate the azimuth fingerprint correction, and the pitch axis backlash correction and the pitch historical offset correction are fused together to generate the pitch fingerprint correction. The azimuth fingerprint correction amount is used as the incremental correction amount of the initial azimuth compensation data to generate azimuth compensation data. The pitch fingerprint correction amount is used as the incremental correction amount of the initial pitch angle compensation data to generate pitch angle compensation data.

9. The solar thermal reflector tracking control method as described in claim 8, characterized in that, Step S6 specifically includes: The theoretical azimuth and theoretical pitch angles of the reflector are obtained from the theoretical light-tracking pose of the reflector. The theoretical azimuth angle of the reflector is superimposed with the azimuth compensation data to generate the target azimuth angle, and the theoretical elevation angle of the reflector is superimposed with the elevation compensation data to generate the target elevation angle; Angle continuity processing is performed on the target azimuth angle, and the target azimuth angle and the target pitch angle are limited to the allowable movement range of the corresponding drive shaft; Generate a reflector tracking control command based on the processed target azimuth and target elevation angles; The reflector tracking control command is sent to the azimuth axis drive motor and the pitch axis drive motor respectively to control the reflector to perform the tracking action.

10. A solar thermal reflector tracking control system, used to implement the solar thermal reflector tracking control method as described in any one of claims 1 to 9, characterized in that, include: Data acquisition module: acquires solar position data, receiver target position data, reflector current attitude data, dual-axis drive feedback data, drive direction change data, and historical spot offset data; Pose generation module: Based on the sun position data, the receiver target position data, and the reflector current attitude data, generate the theoretical light-tracking pose of the reflector; Spot recognition module: Collects image data of the receiver area, determines the actual spot position corresponding to the reflector based on the image data of the receiver area, and generates spot offset data based on the actual spot position and the target position data of the receiver; Fingerprint generation module: Based on the dual-axis drive feedback data, the drive direction change data, and the historical spot offset data, generate a mirror individual error fingerprint that characterizes the dual-axis drive backlash and historical spot offset features of the mirror; Compensation generation module: Based on the spot offset data and the individual error fingerprint of the reflector, it generates azimuth compensation data and pitch compensation data; Command control module: Based on the theoretical light-tracking posture of the reflector, the azimuth compensation data, and the pitch compensation data, it generates a light-tracking control command for the reflector and controls the reflector to perform light-tracking actions according to the light-tracking control command.

Images