A heliostat adaptive concentration regulation method based on soft sphere dynamic potential field evolution

Abstract

This invention discloses an adaptive concentrating control method for heliostats based on the evolution of the dynamic potential field of soft spheres, belonging to the field of solar thermal power generation technology. This invention equates the concentrating spot of each heliostat to a dynamic soft sphere carrying real-time projected energy flow, with the soft sphere's position corresponding to the heliostat's current target point. Three types of potential fields are constructed to jointly constrain the soft sphere's motion: an exponential strong repulsive potential field in the upper and lower protective plate regions, a power-law weak repulsive potential field at the left and right edges, and a mutually repulsive potential field based on the energy flow product between soft spheres. Combining the physical projection domain constraints of each heliostat and the global total energy overload constraints, damped dynamic iterative evolution is performed on all soft spheres. The optimal steady-state distribution position of the soft spheres is then mapped to the real-time target point of the heliostat. Dynamic iterative updates are performed as the sun's position changes, achieving all-day adaptive concentrating control. This invention can achieve uniform energy flow distribution in the receiver, eliminate local overheating hotspots, and adapt to the differentiated heat transfer characteristics of molten salt, significantly improving the operational safety and heat collection efficiency of tower solar thermal power plants.

Description

Technical Field

[0001] This invention belongs to the field of solar thermal power generation technology, and specifically relates to a heliostat adaptive focusing control method based on the evolution of the dynamic potential field of a soft sphere. Background Technology

[0002] In a tower-type concentrated solar power (CSP) system, the heliostat field is the core energy harvesting component. Its function is to reflect and focus scattered sunlight onto the collector at the top of the absorber tower, generating temperatures of hundreds to thousands of degrees Celsius to provide continuous thermal energy for the power generation system. The heliostat uses a tracking and control system to track the sun's azimuth and elevation angle in real time, ensuring that the reflected light spot is accurately projected onto the target area of ​​the collector.

[0003] Current heliostat focusing control methods mostly employ fixed target points or a limited number of target points switching. The basic idea of ​​traditional multi-target point control is: when a certain area of ​​the receiver exceeds the temperature limit, some of the corresponding heliostats are adjusted to another target point. This is achieved by establishing a 0-1 integer linear programming model and using an implicit enumeration algorithm to solve it, thus realizing a balanced energy distribution. The advantages of this method are its simple calculation and flexible control, but it has the following shortcomings: (1) The number of target points is limited, making it difficult to cope with complex heat flow distribution and limiting control accuracy; (2) The control algorithm requires frequent calculations, which places high demands on computing resources, and the control accuracy decreases under extreme weather conditions; (3) It is impossible to distinguish the energy flow load difference between the low temperature panel at the inlet of the receiver and the high temperature panel at the outlet, which can easily lead to long-term overheating of the high temperature panel and redundant load of the low temperature panel. (4) The concentration ratio of tower solar thermal power plants is as high as 900 to 1000 times, and the brightness of the light spot is extremely high. Conventional instruments cannot detect the deviation of a single mirror, and the traditional "deviation measurement - feedback adjustment" closed-loop control mode is difficult to apply. (5) Existing methods do not fully consider the physical projection range limitations of a single heliostat due to its installation location and mechanical limits of rotation angle, which can easily generate unreachable false target points, resulting in the inability to execute control commands; (6) When the global energy is overloaded, the batch random removal of mirrors is often adopted, which can easily cause energy flow voids and distribution imbalance on the plate surface, affecting heat collection efficiency and operational stability.

[0004] Therefore, there is an urgent need for a heliostat focusing control method that can adaptively adjust the light spot arrangement, take into account both safety constraints and heat collection efficiency, and is feasible in engineering. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a heliostat adaptive concentrating control method based on the evolution of the dynamic potential field of a soft sphere, in order to achieve uniform distribution of energy flux density on the surface of the receiver, eliminate local overheating hotspots, adapt to the differentiated heat transfer characteristics of molten salt, ensure safe operation of the equipment, and improve the heat collection efficiency of tower solar thermal power plants.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows: An adaptive focusing control method for heliostats based on the evolution of the dynamic potential field of a soft sphere includes the following steps: Step 1: Divide the molten salt absorber into panels, set the energy threshold and peak energy flow threshold for each panel independently, and establish a two-dimensional coordinate system and boundary parameters for the upper and lower guard plates and left and right edges; Step 2: The focusing spot of each heliostat is equivalent to a dynamic soft sphere carrying a real-time projected energy flow, and the position of the soft sphere corresponds to the current target point of the heliostat; Step 3: Based on the boundary parameters of Step 1 and the soft sphere model of Step 2, construct three types of potential fields: exponential strong repulsion potential field in the upper and lower protective plate areas, power-law weak repulsion potential field in the left and right edges, and mutual repulsion potential field based on the energy flow product between soft spheres. Calculate the repulsion force generated by each potential field on each soft sphere and combine them into the total resultant force on the soft sphere. Step 4: Based on the total resultant force on the soft sphere calculated in Step 3, and combined with the physical projection domain constraints of each heliostat and the global total energy overload constraints, perform damped dynamics iterative evolution on all soft spheres until a steady state of resultant force equilibrium is reached, and solve for the optimal distribution position of each soft sphere. Step 5: Map the optimal coordinates of each soft sphere obtained in Step 4 to the real-time target points of each heliostat, and drive the heliostat to adjust its attitude. Step 6: Collect the sun's position in real time. When the sun's position deviates from the specified threshold, re-trigger steps 3 to 5 to dynamically update the target point and achieve all-day adaptive focusing control.

[0007] In the above scheme, the exponential strong repulsive potential field mentioned in step 3 adopts a spot radius compensation mechanism, which shifts the safety constraint boundary at the center of the soft sphere inward by an equivalent radius of the soft sphere. Its potential energy function is: ; in, The repulsion strength coefficients at the upper and lower boundaries. The repulsive force attenuation scale, Let be the equivalent soft sphere radius of the focusing spot of the i-th heliostat. For the geometric boundary of the lower guard plate, For the geometric boundary of the upper guard plate, Let be the ordinate of the center of the soft sphere corresponding to the i-th heliostat.

[0008] In the above scheme, the potential energy function of the power-law weak repulsive potential field mentioned in step 3 is: ; In the formula, The coefficients represent the weak lateral repulsion strength. , The repulsion strength coefficients at the upper and lower boundaries. The lateral relaxation length, The potential field steepness coefficient, For the left boundary, The right boundary Let x be the x-coordinate of the center of the soft sphere corresponding to the i-th heliostat.

[0009] In the above scheme, in the mutually repulsive potential field based on the energy flow product between soft spheres described in step 3, any two heliostats i , j The corresponding mutual repulsive potential energy between the soft spheres is: ; heliostat i The total internal repulsive force experienced by the corresponding soft sphere is: ; In the formula, This is the internal repulsion proportionality coefficient. For heliostats i The corresponding real-time energy flow projected by the soft sphere For heliostats j The corresponding real-time projected energy flow of the soft sphere The spatial distance between the two soft spheres. For heliostats The corresponding soft ball position, For heliostats The corresponding position of the soft ball.

[0010] In the above scheme, the physical projection domain constraint in step 4 is: each soft sphere is independently bound to the movable projection domain of its corresponding heliostat, and when the soft sphere's evolution position exceeds the projection domain, it is locked on the boundary.

[0011] In the above scheme, the global total energy overload constraint in step 4 is: real-time statistics of the total projected energy of the soft ball across the entire court. ,like Then, the low-contribution soft sphere active elimination mechanism is activated, prioritizing the elimination of soft spheres with redundant energy flux density, close to the boundary, and with low contribution to uniformity, until... ,in This represents the maximum permissible energy capacity across the entire receiver area.

[0012] In the above scheme, the governing equation for the iterative evolution of damping dynamics in step 4 is: ; In the formula: For the equivalent mass of a soft ball, For heliostats The corresponding soft ball position, The coefficient of motion damping, The total repulsive force at the upper and lower boundaries. The total repulsive force at the left and right boundaries. The total repulsive force between the internal soft spheres; The system iterates until acceleration and velocity approach zero, and by combining physical projection domain constraints and global total energy overload constraints, it achieves a steady-state optimal distribution.

[0013] In the above scheme, the steady state of the resultant force balance in step 4 must simultaneously satisfy the following: all soft sphere target points are within the physical projection range of the corresponding heliostat, all light spot bodies are within the effective panel of the receiver, the energy flow distribution within a single panel is uniform, the total energy of a single panel is less than its preset threshold, and the total energy of the entire field does not exceed the maximum load-bearing energy of the receiver as a whole.

[0014] Through the above technical solution, the heliostat adaptive focusing control method based on the evolution of the dynamic potential field of a soft sphere provided by the present invention has the following beneficial effects: (i) Achieve differentiated and precise energy control for different zones of the receiver to reduce the risk of thermal damage to the equipment. This invention achieves a differentiated light-concentrating effect of "high inlet load and low outlet load" by independently configuring energy thresholds for each heated panel and spontaneously realizing the potential field evolution, perfectly matching the heat transfer law of molten salt heating along the path. Compared with the traditional method of uniform fixed target point, this invention can reduce the risk of overheating, thermal creep, and thermal fatigue damage to the high-temperature outlet panel from the source, significantly extending the service life of the receiver.

[0015] (ii) Spontaneously achieve uniform energy flow distribution on the plate surface, eliminating local hot spots and large temperature gradients. This invention constructs a mutually repulsive potential field for energy-fluid coupling soft spheres, causing high-energy-fluidity focusing units to automatically disperse and low-energy-fluidity units to fill in the gaps, relying on dynamic evolution to spontaneously equalize the energy flow distribution on the plate surface. Compared to the problem of fixed spot positions and the formation of stubborn hot spots in traditional fixed-target-point schemes, this invention can achieve a smooth transition of the plate surface temperature field without manual intervention, effectively reducing thermal stress damage and improving the operational stability of the equipment.

[0016] (III) Dual-layer differentiated boundary constraints, balancing zero spillover safety and global uniform energy flexibility This invention employs a strong repulsive potential with radius compensation at the top and bottom, and a flexible constraint with a weak field at the left and right. This completely solves the problem of light spillage from the top and bottom protective plates caused by the light spot radius, while retaining the ability to make minor adjustments to the left and right edges. Compared to traditional single-boundary constraint methods, this invention maximizes the utilization of the effective heat-receiving area of ​​the heat absorber and improves heat collection efficiency while ensuring equipment safety.

[0017] (iv) Adhere to the physical projection limit of the heliostat to ensure that the control command engineering can be implemented. This invention binds an independent projection domain constraint to each heliostat, strictly limiting the evolution position of the soft sphere to a physically reachable range. Compared to traditional adaptive algorithms that are prone to generating false target points and causing control failures, this invention ensures that all control commands are executed precisely, avoiding ineffective control and tracking disorder, and improving the reliability of system control.

[0018] (v) Optimize the global overload mirror removal logic to balance safety and heat collection efficiency. This invention employs a potential field-correlated low-contribution soft sphere precise removal strategy when total energy exceeds the limit, prioritizing the removal of concentrating units with boundary redundancy and low uniformity contribution. Compared to the traditional batch random lens removal method, which easily causes energy flow voids and distribution imbalances, this invention rapidly reduces the total energy back to a safe threshold while maximizing the preservation of energy flow uniformity on the plate surface, significantly reducing efficiency losses and temperature fluctuations caused by overload control.

[0019] (vi) Full-condition adaptive dynamic control to adapt to real-time changes in solar trajectory. This invention updates the potential field state and the optimal arrangement of soft spheres in real time according to the sun's position, and dynamically iterates the target point of the heliostat. Compared with the traditional fixed target point scheme, which cannot adapt to the sun's position in real time and leads to the degradation of the light-gathering effect, this invention achieves intelligent adaptive light-gathering control under all-day, all-weather conditions, with significantly better adaptability and robustness. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0021] Figure 1 This is a schematic diagram of a heliostat adaptive focusing control method based on the evolution of the dynamic potential field of a soft sphere, as disclosed in an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram comparing the distribution of target points before and after the dynamic evolution of the soft ball according to the present invention; where (a) is the distribution of target points in the initial state, and (b) is the distribution of target points after the evolution reaches equilibrium.

[0023] Figure 3 is a schematic diagram comparing the energy flux density distribution of the heat-absorbing panel before and after the kinetic evolution of the present invention; where (a) is the energy flux density distribution in the initial state and (b) is the energy flux density distribution after the evolution equilibrium. Detailed Implementation

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0025] I. Implementation Scenarios and System Configuration This embodiment uses a typical tower-type concentrated solar power (CSP) plant as an example. Located in a desert region at 40°N, 95°E, the plant has a designed power output of 50MW and employs molten salt receivers and exposed tubular absorber panels. The heliostat field consists of 24,000 heliostats, each with an area of ​​30m². 2 With a reflectivity of 94%, it is installed in an annular area 300m to 800m away from the heat absorber tower. The heat absorber's heating panel has a geometric dimension of 9.9m (height) × 16.5m (width), and the panel surface is coated with a selective absorption coating. The rated operating temperature range is 290℃ (inlet) to 565℃ (outlet).

[0026] The control system architecture is as follows: An industrial control computer (CPU clock speed 3.5GHz, memory 32GB, running Linux real-time operating system) serves as the central controller, responsible for executing the soft sphere dynamics evolution algorithm; it communicates with 24,000 heliostat controllers in the field via EtherCAT industrial Ethernet, with a communication cycle of 1 second. Each heliostat is equipped with a dual-axis servo motor and an absolute encoder, with an angle control accuracy of ±0.02°. 120 thermocouples (K-type, accuracy ±1.5℃) are evenly arranged on the receiver panel to monitor temperature distribution. The data is acquired by the PLC and uploaded to the central controller, but is only used for safety monitoring and does not participate in closed-loop feedback control—because conventional closed-loop control is not applicable at high concentration ratios, which is precisely the technical deficiency that this invention aims to address.

[0027] II. Specific implementation steps, such as Figure 1 As shown, the process includes the following: Step 1: Absorber zone modeling and threshold parameter initialization (1) Panel area division Based on the flow path and temperature rise characteristics of the molten salt working fluid in the absorber tube bundle, the absorber is generally designed as a polygonal prism, with each prism's side defined as a panel. The north-facing panel is the molten salt inlet region, where the tube wall temperature is low and the thermal stress margin is large, allowing for a higher energy flux density. The south-facing panel is the molten salt outlet region, where the tube wall temperature is close to the material's limit (heat-resistant steel can withstand a maximum of approximately 600°C), and the energy flux carrying capacity is the lowest. The remaining panels, from north to south, fall between these two extremes.

[0028] (2) Differentiation threshold setting Based on the thermodynamic simulation data and engineering safety margins provided by the receiver manufacturer (usually 80% of the limit value), two key thresholds are independently preset for each panel: Low-temperature inlet zone: Maximum permissible total energy threshold for a single panel is 2.5 MW, and the local peak energy flow threshold is 800 kW / m². 2 ; Intermediate temperature transition zone: Maximum allowable total energy threshold for a single panel is 2.2 MW, and the local peak energy flow threshold is 700 kW / m². 2 ; High-temperature exit zone: Maximum permissible total energy threshold for a single panel is 1.8 MW, and the local peak energy flow threshold is 550 kW / m². 2 ; Simultaneously preset the maximum allowable energy capacity of the entire heat absorber area. (90% of the sum of the three panels, with a 10% safety margin). These thresholds will serve as hard constraints in subsequent dynamic evolution.

[0029] (3) Establishing a coordinate system Establish a two-dimensional plane coordinate system for the absorber: the origin is located at the geometric center of the panel, the X-axis is horizontal to the right (east is positive), and the Y-axis is vertically upward (zenith is positive). Coordinates of the hard boundaries of the upper and lower protective plates: lower protective plate geometric boundary... upper protective plate geometric boundary (That is, the effective area of ​​the panel is 9.9m plus the gap between the top and bottom guard plates of 0.5m). Left and right flexible edge boundaries: Left boundary right boundary This corresponds to half of the panel width of 16.5m. The global constraint parameters are initialized and stored in the control system database.

[0030] Step 2: Equivalent Modeling of Heliostat Spot Softening and Spherization (1) Light spot geometry and energy calculation For each heliostat ( The central controller is based on the current position of the sun (altitude angle). Azimuth Heliostat installation coordinates And assuming the current target point (initially set as the center of the receiver), the projection of the heliostat's reflected light spot onto the receiver plane is calculated using ray tracing. Due to the slight curvature, mirror slope error, and tracking error of the heliostat surface, the light spot actually exhibits a two-dimensional Gaussian distribution or a slanted elliptical Gaussian distribution, rather than an ideal point. This invention treats each light spot as an equivalent dynamic "soft sphere," each soft sphere carrying the real-time projected energy flow power information of the corresponding heliostat, and the position of the soft sphere corresponds to the current focusing target point position of the heliostat. All soft spheres possess the characteristic of dynamically evolving positions, abandoning the traditional fixed coordinate target point constraint, and providing a model basis for adaptive control. Its core characteristic parameters include: equivalent radius Defined as the average distance from the center of the light spot to the boundary of the region containing 85% of the total energy, in meters. In this embodiment, different heliostats... The distance between the heliostat and the heat-absorbing tower varies from 0.3m to 0.6m, depending on the distance between the heliostat and the tower. The spot size of the heliostat is smaller when it is closer to the tower (about 0.3m) and larger when it is farther away (up to 0.6m).

[0031] Equivalent total power According to real-time direct solar normal irradiance (DNI, unit W / m²) 2 The reflectance is calculated from the heliostat's reflective area and reflectivity in this embodiment. The range is between 1.5kW and 4.2kW ​​(DNI=800W / m). 2 (Under typical operating conditions).

[0032] (2) Initialization of soft ball The initial position of each soft ball on the absorber plane is set to be randomly scattered within the panel area (satisfying uniform distribution), but the following basic constraints must be met: the ordinate of the center of the soft ball It must not exceed the hard boundaries of the upper and lower guard plates (i.e. Otherwise, it will be forcibly pulled back. The initial set of all soft balls' initial positions constitutes the initial target point distribution, such as... Figure 2 As shown in (a), the initial distribution is usually chaotic, with a lot of overlap and local stacking, which urgently needs to be optimized by subsequent potential field evolution.

[0033] Step 3: Construct a triple-coupled potential field constraint system This invention constructs three types of potential fields to jointly constrain the motion of a soft sphere, enabling refined and differentiated control of light concentration. The potential energy functions of the three types of potential fields and their corresponding repulsive forces are calculated as follows.

[0034] (1) Strong repulsive potential field of upper and lower protective plates Physical Background: There are non-heated shielding areas above and below the heat absorber. If the light spot shines onto these shielding areas, it not only wastes energy but also causes the shielding to overheat, deform, or even burn due to the high heat flux. Traditional methods use the geometric boundaries of the panel as constraints, but ignore the fact that the light spot itself has a radius. This means that even if the center of the soft sphere does not cross the boundary, the edge of the light spot may still overflow onto the shielding. This invention corrects this by providing radius compensation.

[0035] Safety constraint boundary correction: Let the equivalent radius of the soft sphere be... The safe range of motion for the center of the soft ball should be:

[0036] That is, it shrinks inward by the radius of a soft sphere. In this embodiment, for The soft sphere has a lower boundary safety constraint of The upper boundary safety constraint is It is perfectly aligned with the effective area of ​​the panel.

[0037] Potential function design: An exponential potential energy model is used, characterized by a sharp increase in potential energy near the boundary, generating a strong repulsive force to prevent the soft ball from entering the penalty area. The formula is as follows: ; in, The boundary repulsion strength coefficient is taken in this embodiment. ; To measure the repulsive force attenuation, take When the center of the soft sphere penetrates the restricted area by 0.05m, its potential energy reaches approximately 5000J, generating a huge repulsive force. This force counteracts the excess light caused by the radius of the soft sphere itself, completely preventing the edge of the light spot from illuminating the upper and lower protective plate areas from a physical mechanism, thus achieving zero longitudinal light leakage constraint. The corresponding boundary repulsive force is a negative potential energy gradient; the closer to the safety constraint boundary, the stronger the repulsive force, forcing the entire soft sphere to be completely constrained within the effective heating panel area.

[0038] Repulsive force calculation: The repulsive force is the negative gradient of the potential energy. For the lower boundary, when hour: ; in, This represents the repulsive force at the lower boundary; For the upper boundary, when hour: ; in, This represents the repulsive force at the upper boundary; The repulsive force increases exponentially with the depth of penetration, ensuring that the center of the soft ball can hardly cross the safety boundary, thus completely preventing the edge of the light spot from illuminating the upper and lower protective plates from a physical mechanism.

[0039] (2) Weak repulsive potential field at the left and right edges Physical Background: The receiver lacks protective plates on both sides, but excessive deviation of the light spot from the panel edge can lead to energy loss and localized overheating. However, a moderate, small deviation (e.g., no more than 0.2m) is permissible in practical concentrating and provides more flexibility in adjusting energy flow uniformity. Therefore, the left and right boundaries employ weak constraints rather than strong repulsion.

[0040] Potential function design: A power-law potential energy is adopted, which increases gradually with increasing distance. ;

[0041] Parameter settings in this embodiment: Transverse weak repulsion strength coefficient Significantly smaller than the upper and lower boundaries ; Lateral relaxation length That is, the potential energy is approximately 200 J when the depth exceeds 0.2 m; the potential field steepness coefficient It lies between linear (α=1) and quadratic (α=2), ensuring that the repulsive force increases gradually.

[0042] Repulsive force calculation: For the left boundary, when hour: ; in, This represents the repulsive force at the left boundary; For the right boundary, when hour: ; in, This represents the repulsive force at the right boundary; The repulsive force is small (<100N) when the distance exceeds 0.1m, allowing the soft ball to slightly exceed the limit; when the distance exceeds 0.5m, the repulsive force increases significantly (>1000N), effectively limiting excessive deviation.

[0043] (3) Mutually repulsive potential field inside a soft sphere with energy flow coupling Physical Background: The main reason for the uneven energy flow distribution on the absorber panel is that high-energy flow spots are concentrated in local areas. This invention designs a novel repulsive potential, such that the repulsive strength between the soft spheres is proportional to the product of their energy flows, thereby achieving a self-organizing behavior of "high-energy flow strongly repelling each other and automatically dispersing; low-energy flow weakly repelling each other and filling gaps".

[0044] Potential function design: for any two soft spheres and The mutual repulsive potential energy between them is: ; in, The distance between the centers of the two soft balls. For heliostats The corresponding soft ball position, For heliostats The corresponding position of the soft ball; This is the internal repulsion proportionality coefficient; in this embodiment, it is taken as... This potential energy form is similar to the Coulomb repulsion between charged particles, but the charge is replaced by an energy flux product, which automatically increases the repulsion distance between high-energy flux spots.

[0045] Repulsive force calculation: soft ball Subjected to soft ball The repulsive force is the negative gradient of potential energy: ; Then soft ball The total internal repulsive force exerted by all other soft spheres is: ; The direction of this force is consistent with the line connecting the centers of the two soft spheres, and its magnitude is directly proportional to the product of energy flux and inversely proportional to the square of the distance. In the simulation, attention should be paid to when… When the distance is too small (less than 0.05m), it will generate a large repulsive force, causing the values ​​to be unstable. Therefore, a softening distance is introduced. In actual calculations, the following method is used: Alternative .

[0046] Step 4: Solving for the optimal focusing position using soft sphere dynamics evolution. This step treats all soft spheres as point masses moving in a potential field, and iterates until the system reaches equilibrium by solving the damped dynamic equations and combining them with rigid engineering constraints.

[0047] (1) Spatial constraints of the heliostat projection field (limitation of the range of motion of the soft sphere) In practical engineering, the elevation and azimuth angles of each heliostat are limited by mechanical limit switches, and its reflected light can only cover a finite sub-region on the receiver. For example, the south heliostat (located on the south side of the receiver tower) cannot project a light spot onto the north side of the receiver (x>0 region) because the reflected light needs to be deflected significantly northward, exceeding the elevation angle limit. Therefore, this invention independently calibrates the physically projectable area of ​​each heliostat.

[0048] Calibration method: During the installation and commissioning phase of the heliostat, the range of its elevation angles is measured through an angle scanning experiment. and azimuth range By combining the solar position reference, the reachable coordinate range on the receiver plane is obtained through geometric optics inverse calculation. In this embodiment, typical values ​​are as follows: South heliostat (azimuth angle approximately 180°): ; ; North heliostat (azimuth angle approximately 0°): ; ; Eastern heliostat (azimuth angle approximately 90°): However, the range of y is relatively narrow, approximately 3~7m.

[0049] During the dynamic evolution process, the soft sphere coordinates are checked after each iteration. Is it within the corresponding projection domain? If Then force set to ;like , set as The same applies to the y-coordinate. This process ensures that all evolved target points are engineering-reachable, avoiding the "false target point" problem common in traditional algorithms.

[0050] (2) Global total energy overload constraint and soft ball removal mechanism Overload detection: Real-time calculation of total projected energy of the soft ball across the entire court. , where N is the number of heliostats currently participating in focusing (initially N=3000). If (In this embodiment, it is 6.0MW), all soft spheres participate normally in the potential field evolution. If If this occurs, it is determined to be an energy overload, and the focusing power must be reduced. This represents the maximum permissible energy capacity across the entire receiver area.

[0051] Removal Strategy: Unlike traditional coarse-grained random or region-based phase removal, this invention designs a precise removal mechanism based on potential field contribution. The specific steps are as follows: Calculate the "removal priority index" for each soft ball. : ; in, The distance (in meters) is the nearest Euclidean distance from the center of the soft sphere to the boundary of the absorber panel. To prevent small amounts from being divided by zero; The local energy flow uniformity index of the soft sphere is defined as the energy flow variance between the soft sphere and its 8 nearest neighbor soft spheres, normalized to [0,1]; For weighting coefficients, this formula means that a soft sphere with low power, located near the boundary, and in a region of dense energy flux has a smaller weighting coefficient. Those that are prioritized for removal are as follows: Step 1, press all the soft balls Sort by size from smallest to largest.

[0052] Step 2: Elimination The smallest soft sphere (i.e., stopping the heliostat's focusing, invalidating its target point, or defocusing it), while updating... Recalculate .

[0053] Step 3: Repeat steps 1-3 until... .

[0054] This rejection mechanism can rapidly reduce total energy while preserving the uniformity of energy flow distribution on the plate surface to the greatest extent. The rejected heliostats enter a standby state and can be reactivated through the reverse process when subsequent solar irradiance decreases or other heliostats are deactivated due to malfunction.

[0055] (3) Evolutionary equations and solutions Each soft ball is considered to have a mass of . The equation of motion for a particle moving in a damped medium is as follows: ; in: (The equivalent mass of the soft ball is only used to adjust inertia; its actual value has no effect on steady state.) (Damping coefficient controls the convergence speed; too large a coefficient leads to slow convergence, while too small a coefficient causes oscillations.) This represents the total repulsive force at the upper and lower boundaries (Y-direction component only). The total repulsive force at the left and right boundaries (X-direction component only). The total repulsive force between the internal soft spheres (in both the X and Y directions).

[0056] Numerical solution method: A second-order Euler-Cromer integration method is adopted, which has good energy stability. Let the time step be... ,speed The iteration format is then: ; ; in, For the soft sphere corresponding to the i-th heliostat Step speed, For the soft sphere corresponding to the i-th heliostat Step speed, The soft sphere corresponding to the i-th heliostat The position of the step, The soft sphere corresponding to the i-th heliostat The position of the step; For time step; is the motion damping coefficient.

[0057] Immediately after each update, apply heliostat projection domain constraints and boundary enforcement constraints (for the upper and lower boundaries, if the boundary is exceeded, reverse the velocity and reset the boundary).

[0058] Convergence criterion: Calculate the total kinetic energy of the system ,when And the maximum value of the resultant force of all soft balls When the system reaches a steady state, it is determined that the system has reached a steady state. In this embodiment, the typical number of iterations is 2000 to 5000, which corresponds to 20 to 50 seconds of physical time, which is acceptable in actual engineering (because the mechanical adjustment time of the heliostat is also on the order of tens of seconds).

[0059] (4) Steady-state constraint verification After the evolution is complete, the system automatically checks whether all five of the following constraints are satisfied: Physically Realizable Constraints: All Soft Sphere Coordinates All are within the dedicated physical projection range of the corresponding heliostat.

[0060] Boundary safety constraints: All soft spheres Strictly meet ;all Exceeding The amount should not exceed 0.3m (small deviations are allowed on the left and right edges, but must be controlled within the safety threshold).

[0061] Uniform energy flux constraint: Within each panel region, calculate the coefficient of variation (standard deviation / mean) of the energy flux density distribution, requiring... (i.e., relatively uniform).

[0062] Zone threshold constraints: The cumulative total power of each panel is less than the corresponding preset threshold (2.5MW / 2.2MW / 1.8MW), and the local peak energy flux density is less than the corresponding threshold (800 / 700 / 550 kW / m²). 2 ).

[0063] Global overload constraints: .

[0064] If any constraint is not satisfied, the system will take supplementary measures according to the specific situation: for example, if the partition threshold is exceeded, the mutual exclusion strength coefficient of the soft spheres corresponding to that panel will be adjusted. Re-evolution; global overload further eliminates low-contribution soft balls.

[0065] Step 5: Map the steady-state position of the soft sphere to the heliostat attitude. The optimal steady-state coordinates of each soft sphere obtained after the convergence of the dynamic evolution. As a corresponding heliostat The real-time focusing target point needs to be converted into the heliostat's elevation angle. and azimuth Control commands.

[0066] Inverse geometric optics calculation: Given the coordinates of the heat absorption tower heliostat center coordinates Target point coordinates (Assuming the receiver plane is located at height Z=0), unit vector in the solar direction. (pointing towards the sun), then the unit vector of the reflected ray direction. Let be the vector pointing from the heliostat to the target point. According to the law of reflection, the normal vector of the mirror surface... Should be divided equally and The included angle: ; The elevation and azimuth angles of the heliostat can be calculated from the normal vector. The controller sends the target angle to the heliostat's servo drive via Ethernet to perform closed-loop adjustment. Due to mechanical inertia, it typically takes 5-15 seconds from the issuance of the command to the arrival of the angle. During this period, the system maintains the target point from the previous round, ensuring continuous operation.

[0067] Step 6: Dynamically update in real time according to the sun's position Triggering condition: Throughout the entire operation of the power plant, the central controller collects the solar altitude angle at a frequency of 1Hz. and azimuth (This can be obtained through astronomical algorithms or solar sensors). If the change in the current solar position compared to the solar position during the last evolution satisfies... or This triggers a new round of soft sphere dynamics evolution.

[0068] Hot-start strategy: The new evolution does not start from zero, but takes the position of the soft ball in the previous steady state as the initial state, and only updates the power of each soft ball. and radius (Because changes in the sun's position cause changes in the shape and energy of the light spot projection). Since the initial state is already close to the optimal solution under the new conditions, convergence can usually be achieved in just 200 to 500 iterations, significantly reducing the computational load.

[0069] The system operates continuously throughout the day: starting at sunrise, it automatically performs potential field evolution and target point updates every 5-10 minutes until sunset. Even if cloud cover causes significant fluctuations in DNI, the system can respond in real time—because... The internal repulsive forces are adjusted accordingly as the system updates in real time, causing the soft balls to rearrange automatically.

[0070] III. Simulation Verification and Result Analysis To verify the effectiveness of the present invention, the applicant built a simulation system based on the MATLAB / Simulink platform, using the parameters described in the above embodiments.

[0071] Simulation conditions: Number of heliostats (corresponding to a single panel): 3000; Solar altitude angle 45°, azimuth angle 180° (noon condition); Initial target points are randomly distributed; The receiver panel is meshed into 60×40=2400 micro-elements, and the energy flux density is calculated for each micro-element. Light spot model: Two-dimensional oblique elliptical Gaussian distribution, standard deviation σ x =0.5~1.2m, σ y =0.4~1.0m, tilt angle ±15°.

[0072] Figure 2 (a) shows the initial state where the target points are randomly distributed, with a large number of soft spheres piled up in the center of the panel; Figure 2 As shown in Figure (b), after the evolution of this invention, the soft balls are evenly distributed in the effective area of ​​the panel, and the spacing is automatically adjusted according to the power level—the distance between high-power soft balls is large, while the low-power soft balls are dense but uniform. Figure 3 In Figure (a), the initial energy flux density distribution is shown, with multiple values ​​>900 kW / m². 2 Hot topics; Figure 3 (b) shows the energy flux density distribution after evolution, with hotspots completely eliminated and the peak value reduced to 520 kW / m². 2 The entire panel displays a smooth transition.

[0073] Dynamic tracking performance test: Simulating the sun moving from an altitude angle of 45° to 60° (a change of approximately 1 hour), the method of this invention updates the target point every 5 minutes. Throughout the process, the peak energy flow of the high-temperature outlet panel remains below 800 kW / m². 2 The panel CV is less than 0.22, which meets the requirements for safe operation.

[0074] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A heliostat adaptive focusing control method based on the evolution of the dynamic potential field of a soft sphere, characterized in that, Includes the following steps: Step 1: Divide the molten salt absorber into panels, set the energy threshold and peak energy flow threshold for each panel independently, and establish a two-dimensional coordinate system and boundary parameters for the upper and lower guard plates and left and right edges; Step 2: The focusing spot of each heliostat is equivalent to a dynamic soft sphere carrying a real-time projected energy flow, and the position of the soft sphere corresponds to the current target point of the heliostat; Step 3: Based on the boundary parameters of Step 1 and the soft sphere model of Step 2, construct three types of potential fields: exponential strong repulsion potential field in the upper and lower protective plate areas, power-law weak repulsion potential field in the left and right edges, and mutual repulsion potential field based on the energy flow product between soft spheres. Calculate the repulsion force generated by each potential field on each soft sphere and combine them into the total resultant force on the soft sphere. Step 4: Based on the total resultant force on the soft sphere calculated in Step 3, and combined with the physical projection domain constraints of each heliostat and the global total energy overload constraints, perform damped dynamics iterative evolution on all soft spheres until a steady state of resultant force equilibrium is reached, and solve for the optimal distribution position of each soft sphere. Step 5: Map the optimal coordinates of each soft sphere obtained in Step 4 to the real-time target points of each heliostat, and drive the heliostat to adjust its attitude. Step 6: Collect the sun's position in real time. When the sun's position deviates from the specified threshold, re-trigger steps 3 to 5 to dynamically update the target point and achieve all-day adaptive focusing control.

2. The method according to claim 1, characterized in that, The exponential strong repulsive potential field described in step 3 employs a spot radius compensation mechanism, shifting the safety constraint boundary at the center of the soft sphere inward by an equivalent radius of the soft sphere. Its potential energy function is: ； in, The repulsion strength coefficients at the upper and lower boundaries. The repulsive force attenuation scale, Let be the equivalent soft sphere radius of the focusing spot of the i-th heliostat. For the geometric boundary of the lower guard plate, For the geometric boundary of the upper guard plate, Let be the ordinate of the center of the soft sphere corresponding to the i-th heliostat.

3. The method according to claim 1, characterized in that, The potential energy function of the power-law weak repulsive potential field mentioned in step 3 is: ； In the formula, The coefficients represent the weak lateral repulsion strength. , The repulsion strength coefficients at the upper and lower boundaries. The lateral relaxation length, The potential field steepness coefficient, For the left boundary, The right boundary Let x be the x-coordinate of the center of the soft sphere corresponding to the i-th heliostat.

4. The method of claim 1, wherein, In step 3, the mutual repulsive potential energy between any two heliostats i and j corresponding to the soft spheres in the mutual repulsive potential field based on the energy flow product between the soft spheres is: ； The total internal repulsive force on the soft sphere corresponding to heliostat i is: ； In the formula, This is the internal repulsion proportionality coefficient. For the real-time projected energy flow of the soft sphere corresponding to heliostat i, For the real-time projected energy flow of the soft sphere corresponding to heliostat j, The spatial distance between the two soft spheres. For heliostats The corresponding soft ball position, For heliostats The corresponding position of the soft ball.

5. The method according to claim 1, characterized in that, The physical projection domain constraint mentioned in step 4 is as follows: each soft sphere is independently bound to the movable projection domain of its corresponding heliostat, and when the soft sphere evolves beyond the projection domain, it is locked on the boundary.

6. The method of claim 1, wherein, The global total energy overload constraint mentioned in step 4 is: real-time statistics of the total projected energy of the soft ball across the entire court. ,like Then, the low-contribution soft sphere active elimination mechanism is activated, prioritizing the elimination of soft spheres with redundant energy flux density, close to the boundary, and with low contribution to uniformity, until... ,in This represents the maximum permissible energy capacity across the entire receiver area.

7. The method of claim 1, wherein, The governing equations for the iterative evolution of damping dynamics in step 4 are: ； In the formula: For the equivalent mass of a soft ball, For heliostats The corresponding soft ball position, The coefficient of motion damping, The total repulsive force at the upper and lower boundaries. The total repulsive force at the left and right boundaries. The total repulsive force between the internal soft spheres; The system iterates until acceleration and velocity approach zero, and by combining physical projection domain constraints and global total energy overload constraints, it achieves a steady-state optimal distribution.

8. The method of claim 1, wherein, The steady state of the resultant force balance described in step 4 must simultaneously satisfy the following: all soft sphere target points are within the physical projection range of the corresponding heliostat, all light spot bodies are within the effective panel of the receiver, the energy flow distribution within a single panel is uniform, the total energy of a single panel is less than its preset threshold, and the total energy of the entire field does not exceed the maximum load-bearing energy of the receiver as a whole.

Images