An intelligent sunshade control method and system for an electric tricycle

Abstract

The application discloses an intelligent sunshade control method and system for an electric tricycle, and relates to the technical field of intelligent sensors.The method comprises the following steps: acquiring the light intensity, the temperature in the vehicle, the driving speed of the vehicle, the residual power of the power battery and the environmental wind force information through a vehicle-mounted sensor, and constructing an environmental state vector; calling a decision model pre-stored in a controller, inputting the environmental state vector into the decision model, calculating the sunshade cooling benefit, the photovoltaic power generation benefit, the regulation energy consumption cost and the wind resistance safety constraint based on a multi-target evaluation function, and outputting a target posture; calculating the limited optimal inclination of the sunshade shed according to the target posture and the real-time solar elevation angle; driving the adjusting mechanism to execute the inclination adjustment, identifying the blocking risk based on the motor current and the resistance torque change in the adjustment process, triggering the reverse rollback and the safety locking, converting the abstract target posture into the limited optimal inclination combined with the real-time solar elevation angle, and ensuring that the theoretical optimal solution is accurately reached and landed on the physical layer.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensor technology, and in particular to an intelligent sunshade control method and system for electric tricycles. Background Technology

[0002] With the increasing popularity of electric tricycles in urban and rural logistics and personal travel, users are demanding higher and higher levels of driving comfort and energy efficiency. In the current technology, electric tricycles are generally equipped with fixed or manually adjustable sunshades. Although such structures have basic sunshade functions, they are difficult to adaptively adjust according to light intensity, ambient temperature and vehicle dynamic conditions, resulting in limited sunshade effect. Moreover, they cannot effectively balance thermal comfort inside the vehicle and photovoltaic power generation efficiency in strong light or high temperature environments. In recent years, some intelligent sunshade systems have begun to introduce photosensitive sensors or temperature sensors to control the opening or angle adjustment of the sunshade through a single-point triggering method, thus initially realizing automated control.

[0003] Most systems rely solely on a single parameter such as light intensity or in-vehicle temperature for control, neglecting key constraints such as vehicle speed, battery charge, and ambient wind speed. This leads to the forced deployment of sunshades under high wind resistance conditions, posing safety hazards, or missing the optimal opportunity for photovoltaic power generation when the battery is low. The control strategies often adopt a "threshold trigger-execution" mode, lacking a multi-objective collaborative optimization mechanism for sunshade, power generation, energy consumption, and safety, making it difficult to achieve globally optimal decisions under different operating conditions. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an intelligent sunshade control method for electric tricycles to address the safety hazards of forcibly deploying sunshades under high wind resistance conditions, or missing the optimal photovoltaic power generation opportunity when the battery is low. The control strategies mostly adopt the "threshold trigger-execution" mode, which lacks a multi-objective collaborative optimization mechanism between sunshade, power generation, energy consumption and safety, making it difficult to achieve the global optimal decision-making problem under different operating conditions.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an intelligent sunshade control method for electric tricycles, which includes acquiring information on light intensity, interior temperature, vehicle speed, remaining power of the power battery, and ambient wind force through on-board sensors, and constructing an environmental state vector; The decision model pre-stored in the controller is invoked, the environmental state vector is input into the decision model, and the shading and cooling benefits, photovoltaic power generation benefits, energy consumption adjustment costs and wind resistance safety constraints are calculated based on the multi-objective evaluation function, and the target attitude is output. Based on the target attitude and the real-time solar altitude angle, the restricted optimal tilt angle of the sunshade is calculated; The drive adjustment mechanism performs tilt adjustment, and during the adjustment process, it identifies the risk of obstruction based on changes in motor current and resistance torque, triggering reverse retraction and safety lock-up; After the adjustment is completed, the actual cooling effect, actual power generation, and actual regulation energy consumption are collected, and the decision model is adaptively updated based on the actual cooling effect, actual power generation, and actual regulation energy consumption.

[0007] As a preferred embodiment of the intelligent sunshade control method for electric tricycles described in this invention, the environmental state vector includes: The data on light intensity, in-vehicle temperature, vehicle speed, remaining battery power, and ambient wind force are filtered, smoothed, time-synchronized, and normalized to eliminate interference and unify the dimensions. The wind force information is decomposed into wind speed and wind direction components. The processed parameters are combined and encoded to form an environmental state vector for subsequent decision-making.

[0008] As a preferred embodiment of the intelligent sunshade control method for electric tricycles described in this invention, the decision model includes: An initial decision model is established by training based on historical environmental state data and corresponding shading adjustment results; features are extracted from the input environmental state vector, and a set of candidate shading postures is obtained by matching through the decision model. A comprehensive benefit evaluation is performed on the candidate poses, and the optimal result is selected to determine the target pose.

[0009] As a preferred embodiment of the intelligent sunshade control method for electric tricycles described in this invention, the multi-objective evaluation function includes: Based on the current environmental state vector and historical change sequence, the temperature response delay, which characterizes the thermal inertia hysteresis characteristics in the vehicle, and the power generation response delay, which characterizes the instantaneous characteristics of photovoltaic photoelectric conversion, are extracted. Based on the temperature response delay and the power generation response delay, the time phase difference, which reflects the degree of resource competition between thermal regulation and photoelectric conversion in the time dimension, is determined. When the time phase difference exceeds a preset threshold, a time-layered control strategy including a power generation priority zone and a cooling priority zone is constructed. In the power generation priority zone, the first target attitude is output with the constraint of maximizing instantaneous power generation. In the cooling priority zone, the second target attitude is output with the constraint of maximizing cumulative cooling, thereby forming a multi-objective coordinated control strategy based on time layering. Based on the aforementioned time phase difference, feedforward compensation is performed on the evaluation time of the cooling target's benefits to offset the impact of heat transfer lag on control accuracy, so that the cooling effect and power generation effect reach Pareto optimality under a unified time reference. Based on the time-layered control strategy and time compensation mechanism, the target attitude and switching timing parameters corresponding to each control time interval are determined, thereby outputting a phased control command that includes time interval division, target attitude sequence and switching time.

[0010] As a preferred embodiment of the intelligent sunshade control method for electric tricycles described in this invention, the limited optimal tilt angle includes: The solar incidence direction is calculated based on the real-time solar altitude angle and the vehicle's current attitude, and the initial theoretical tilt angle is determined in conjunction with the phased control commands. Based on the time-layered control strategy, the power generation priority interval and the cooling priority interval are corresponding to different tilt angle control stages; Within the priority power generation zone, the tilt angle of the sunshade is adjusted according to the angle between the surface normal of the photovoltaic module and the direction of solar incidence, so that the angle approaches zero in order to obtain the maximum incident irradiance, thereby determining the first tilt angle; Within the cooling priority zone, the tilt angle of the sunshade is adjusted according to the correspondence between the coverage area of ​​the sunshade projected on the vehicle top and the rate of change of the temperature inside the vehicle, so as to maximize the shaded area inside the vehicle and thus determine the second tilt angle. The switching time between the first tilt angle and the second tilt angle is determined by the time phase difference, and the tilt angle adjustment is performed in stages according to the switching time; Based on the time phase difference and the response time of the adjustment mechanism, the first tilt angle and the second tilt angle are time-corrected so that the environmental state corresponding to the time of adjustment completion matches the target control stage. During the tilt adjustment process, the equivalent wind load on the surface of the sunshade is calculated based on the vehicle speed and ambient wind force, and the tilt angle is limited to the safe tilt angle range corresponding to the wind load. When the wind load exceeds the preset threshold, the tilt adjustment corresponding to the safety constraint is executed first.

[0011] As a preferred embodiment of the intelligent sunshade control method for electric tricycles described in this invention, the obstruction risk identification includes: Collect and adjust motor current signals, mechanism position feedback and resistance torque estimation data to comprehensively reflect the operating status of the adjustment mechanism, perform joint analysis to identify abnormal resistance characteristics, and form a blocking judgment result; When a risk of obstruction is detected, the control and adjustment mechanism executes reverse drive and locks the position to achieve safe retraction and protection.

[0012] As a preferred embodiment of the intelligent sunshade control method for electric tricycles described in this invention, the adaptive update includes: The changes in vehicle interior temperature, photovoltaic power generation output, and regulation energy consumption after sunshade adjustment were statistically analyzed to obtain actual operating data, which were then compared with the corresponding predicted benefits to calculate the benefit deviation. The parameters of the decision-making model are adjusted based on the profit deviation, thereby improving the accuracy and adaptability of subsequent control strategies.

[0013] Secondly, the present invention provides an intelligent sunshade control system for electric tricycles, including an environmental perception module, a strategy decision-making module, an attitude calculation module, a safety execution module, and a model optimization module; The environmental perception module is used to acquire information on light intensity, interior temperature, vehicle speed, remaining battery power, and ambient wind speed through onboard sensors, and to construct an environmental state vector. The strategy decision module is used to call the decision model pre-stored in the controller, input the environmental state vector into the decision model, calculate the shading and cooling benefits, photovoltaic power generation benefits, energy consumption adjustment costs and wind resistance safety constraints based on the multi-objective evaluation function, and output the target attitude. The attitude calculation module is used to calculate the limited optimal tilt angle of the sunshade based on the target attitude and the real-time solar altitude angle. The safety execution module is used to drive the adjustment mechanism to perform tilt adjustment, and during the adjustment process, it identifies the risk of obstruction based on the changes in motor current and resistance torque, triggering reverse retraction and safety locking. The model optimization module is used to collect the actual cooling effect, actual power generation, and actual regulation energy consumption after the adjustment is completed, and to adaptively update the decision model based on the actual cooling effect, actual power generation, and actual regulation energy consumption.

[0014] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the intelligent sunshade control method for electric tricycles as described in the first aspect of the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the intelligent sunshade control method for electric tricycles as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: by calling a pre-stored decision model and based on a multi-objective evaluation function, the conflicting constraints of shading and cooling, photovoltaic power generation, energy consumption regulation, and wind resistance safety are transformed into a calculable quantitative game, realizing the global optimal search among comfort, economy, and safety, effectively avoiding the control defect of sacrificing one aspect for another. By combining the real-time solar altitude angle, the abstract target attitude is transformed into a constrained optimal tilt angle, ensuring the accurate attainment of the theoretical optimal solution at the physical level and its engineering implementation. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. 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.

[0018] Figure 1 A flowchart of an intelligent sunshade control method for electric tricycles; Figure 2 This is a schematic diagram of an intelligent sunshade control system for an electric tricycle. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Reference Figures 1-2 As one embodiment of the present invention, this embodiment provides an intelligent sunshade control method for an electric tricycle, comprising the following steps: S1. Obtain information on light intensity, interior temperature, vehicle speed, remaining battery power, and ambient wind speed through onboard sensors to construct an environmental state vector.

[0023] Furthermore, the data on light intensity, in-vehicle temperature, vehicle speed, remaining battery power, and ambient wind force are filtered, smoothed, time-synchronized, and normalized to eliminate interference and unify the dimensions. Among them, the wind force information is decomposed into wind speed and wind direction components.

[0024] The processed parameters are combined and encoded to form an environmental state vector for subsequent decision-making.

[0025] It should be noted that in the process of multi-source environmental perception, the signals output by different sensors are inconsistent in the time dimension and amplitude dimension. Therefore, the original data needs to be uniformly processed before constructing the environmental state vector.

[0026] Considering that sensor outputs are susceptible to random disturbances, directly using instantaneous values ​​in state calculations would mislead the decision-making system with high-frequency noise, leading to frequent and disordered actions. By extracting a historical window forward from the current moment and applying an equal-weighted average to all sampled values ​​within the window, random noise in both positive and negative directions is suppressed due to mutual cancellation, while stable trend changes are amplified due to continuous directional accumulation. This constructs a smooth quantity representing the main trend of the environment within the time window, expressed as: ; in, The number of historical sampling points. Indicates in Raw sensor readings at time of day. The smoothed trend value is obtained by adjusting the trend value. Calculating the average of historical data is equivalent to introducing a finite impulse response low-pass filter, which suppresses high-frequency noise while preserving low-frequency changes, thereby highlighting the trend of environmental changes.

[0027] Considering the offset in sampling times of different sensors, slight time differences in the actual data from each channel can lead to causal misalignment when combining data in the state vector. Without processing, changes in illumination and temperature will be incorrectly paired, causing erroneous shading inferences. This is addressed by uniformly mapping the observed values ​​of all variables to the control command execution time. This ensures that each state variable is precisely latched onto the same time base, thereby eliminating the time base bias introduced by asynchronous sampling, and constructing the expression: ; in, Indicates the first The instruction execution time of each control cycle To align the sensor values ​​at the time of instruction execution, all state variables are expressed on the same time section, thereby avoiding state mismatch caused by sampling time difference.

[0028] Considering the differences in numerical scales among various physical quantities—for example, light intensity often reaches thousands of lux, while vehicle speed is only tens of kilometers per hour—if directly involved in calculations, parameters with vastly different values ​​will dominate the entire cost function in inner product or distance metrics, masking the small but crucial changes in parameters with smaller dimensions. To eliminate this scale-dominated effect, range normalization is performed on each variable separately, linearly projecting the original measurements to a unified interval. This ensures that variables with different dimensions have equivalent influence in subsequent angle decisions or energy calculations, leading to the following expression: ; in, and These are the minimum and maximum values ​​of the variable observed within the statistical period, respectively. The normalized dimensionless value amplifies the relative changes within the original variable while suppressing differences in absolute magnitude, so that changes in illumination and vehicle speed have the same weight in subsequent decisions.

[0029] To comprehensively characterize the multiphysics field under driving and charging conditions, illumination, in-vehicle temperature, vehicle speed, photovoltaic power, wind speed, and wind direction are sequentially combined to form an environmental state vector. Wind force is decomposed into two independent components: wind speed and wind direction. This is because the overturning moment of the wind load on the sun visor is related to the square of the wind speed and the aerodynamic angle. If the wind speed scalar is combined, it would be impossible to distinguish between favorable and dangerous wind directions. Therefore, decomposition makes the wind direction information independent, providing directional input for subsequent wind safety constraints. Finally, the environmental state vector is constructed. ; in, To normalize the light intensity, To normalize the in-vehicle temperature, To normalize vehicle speed, To normalize photovoltaic power generation, To normalize wind speed, This is the wind direction angle.

[0030] Instead of simply stacking sensor data into a vector, we first use time window averaging to transform random disturbances into reliable trend quantities, then unify asynchronous sampling to the same decision moment through time alignment, eliminate dimension dominance through scale compression, and directionally split wind parameters in vector construction. This ensures that the generated state vector not only filters out high-frequency interference but also maintains the equivalent expressive power of each variable in the time, magnitude, and direction dimensions. Quantization is designed specifically based on clear physical relationships—the randomness of noise, the asynchronous nature of sampling, the differences in dimensions, and the directionality of wind. This allows subsequent multi-objective control to operate on a well-thought-out unified information framework, rather than based on empirical adjustments on raw, coarse data.

[0031] S2. Call the decision model pre-stored in the controller, input the environmental state vector into the decision model, calculate the shading and cooling benefits, photovoltaic power generation benefits, energy consumption adjustment costs and wind resistance safety constraints based on the multi-objective evaluation function, and output the target attitude.

[0032] Furthermore, an initial decision model is established by training based on historical environmental state data and corresponding shading adjustment results. Features are extracted from the input environmental state vector, and a set of candidate shading postures is obtained through matching using the decision model.

[0033] A comprehensive benefit evaluation is performed on the candidate poses, and the optimal result is selected to determine the target pose.

[0034] Based on the current environmental state vector and historical change sequence, the temperature response delay, which characterizes the thermal inertia hysteresis characteristics inside the vehicle, and the power generation response delay, which characterizes the instantaneous characteristics of photovoltaic power conversion, are extracted. Based on the temperature response delay and the power generation response delay, the time phase difference, which reflects the degree of resource competition between thermal regulation and photovoltaic conversion in the time dimension, is determined.

[0035] When the time phase difference exceeds a preset threshold, a time-layered control strategy is constructed, which includes a power generation priority zone and a cooling priority zone. In the power generation priority zone, the first target attitude is output with the constraint of maximizing instantaneous power generation. In the cooling priority zone, the second target attitude is output with the constraint of maximizing cumulative cooling, thereby forming a multi-objective coordinated control strategy based on time layering.

[0036] Based on the time phase difference, feedforward compensation is performed on the evaluation time of the cooling target's benefits to offset the impact of heat transfer lag on control accuracy, so that the cooling effect and power generation effect reach Pareto optimality under a unified time reference.

[0037] Based on the time-layered control strategy and time compensation mechanism, the target attitude and switching timing parameters corresponding to each control time interval are determined, thereby outputting phased control commands that include time interval division, target attitude sequence and switching time.

[0038] It should be noted that the control strategy does not simply weigh the benefits of shading and cooling, photovoltaic power generation, energy consumption adjustment costs, and wind resistance safety constraints. Instead, it reorganizes the solution structure of the control problem based on the asynchronous characteristics of different physical processes in dynamic response. The original control method of multi-objective trade-off optimization at the same moment is reconstructed into a phased single-objective control method based on time phase division. This allows the power generation target and the shading and cooling target to be optimized in the time intervals where the response is most sensitive and the control benefits are most easily amplified. This avoids forcibly placing different physical processes into the same evaluation moment for comparison when the time scales are inconsistent.

[0039] The photovoltaic power generation process is mainly affected by the current incident light state. The output power responds quickly to changes in the direction and intensity of light. Therefore, within the power generation priority interval determined by the time-driven function, the control system takes the optimal power generation posture as the main factor, so that the normal of the photovoltaic module matches the direction of solar incident light as much as possible. Since the power generation can be fed back in a timely manner with the posture change within the stage, the benefits brought by the posture adjustment can be reflected in a short time. Therefore, the time interval belongs to the sensitive response interval of the power generation target.

[0040] The cooling process inside the vehicle is affected by the vehicle's heat capacity, air heat exchange, heat absorption by interior materials, and heat diffusion. Temperature changes do not occur immediately with the sunshade posture, but gradually become apparent after the shading continues for a period of time. Within the cooling priority range determined by the time phase difference, the control system prioritizes the optimal sunshade posture. By increasing the sunshade projection coverage area, the solar radiation entering the vehicle is continuously reduced. During the period when the temperature response begins to be released, the contribution of the sunshade action to the subsequent temperature drop is more easily accumulated and manifested. Therefore, the time range is the sensitive response range of the sunshade cooling target.

[0041] The power generation target and the shading and cooling target do not compromise with each other at the same time, but are optimized separately in the time intervals in which they are more likely to produce an effective response, so that the control results are more in line with the real dynamic characteristics of the photoelectric conversion and heat transfer process.

[0042] In multi-objective control, power generation and cooling have fundamentally different time response characteristics: photovoltaic power generation has almost no inertia and changes synchronously with sudden changes in light intensity. However, the temperature inside the vehicle, due to the involvement of the vehicle's heat capacity, conduction, and convection heat transfer, exhibits a significant lag in its changes. If the two responses are directly optimized within the same time coordinate system, it will inevitably lead to conflicting actions. Therefore, time phase analysis is introduced to divide the dominant time periods for different objectives.

[0043] For the temperature change process, considering the transmission delay between input and output caused by thermal inertia, to capture the feedback moment of the most drastic temperature change within this delay, the time series of the temperature change rate after shading action is continuously observed to find the maximum point of the absolute value of the derivative. This allows the hysteresis phase caused by thermal inertia to be quantified as a characteristic time delay, thereby amplifying the dominant delay characteristic of the temperature response and constructing a representative moment of the temperature response, expressed as: ; in, This is the time offset relative to the current moment. In order to be in The temperature inside the car at any given time. Let be the rate of temperature change at time t, and search for the value of the rate with the largest absolute value. value, The moment of most dramatic temperature response after shading action was identified, which reflects the hysteresis scale of thermal inertia.

[0044] The time offset with the largest absolute value of the temperature change rate is selected as the temperature response delay because the time corresponds to the stage in which the sunshade action affects the temperature inside the vehicle. Near this time, the temperature inside the vehicle responds most strongly to changes in the sunshade status. Continuing to take sunshade cooling as the primary objective allows the shading effect to be more fully converted into actual cooling effect. The temperature response delay is not only used to describe the time difference of thermal inertia but also to determine the time basis for the sunshade cooling target to enter the sensitive response range.

[0045] For the power generation process, photovoltaic output and changes in illumination are almost instantaneously synchronized, and the response lacks obvious thermal capacity-like inertia. To measure the instantaneous response using a unified time delay parameter framework, we also search for extreme points in the rate of change of power generation, using these extreme points as time markers for the instantaneous response. This amplifies the zero-hysteresis characteristic of the power generation process, constructing a representative moment of the power generation response, expressed as: ; in, for Photovoltaic power generation at any given time The rate of change of power, Typically close to zero, this indicates the immediacy of the power generation response.

[0046] The time offset with the largest absolute value of the rate of change of power generation is selected as the power generation response delay because the moment corresponds to the stage when the photovoltaic output is most sensitive to changes in attitude or illumination. Since the dynamic response of the photovoltaic power generation process is relatively fast, the power generation response delay usually corresponds to the time position close to the current control moment. Therefore, prioritizing the execution of the optimal power generation attitude within the time interval can enable the photovoltaic module to capture the power generation gain brought about by changes in solar incidence conditions in a timely manner.

[0047] The difference between two representative moments is not a simple subtraction, but rather a comparison of the thermal inertia delay and the instantaneous photoelectric response on the same time scale. This makes the separation between the two on the time axis explicit, thereby amplifying the degree of misalignment between the dynamic responses of the two types of targets. This provides a clear boundary for subsequent segmented control, constructing the time misalignment quantity, expressed as:

[0048] in, This represents the lag time between the temperature response and the power generation response. The larger the value, the more severe the time disconnect between the two dynamics, and the more time-sharing dominance is needed.

[0049] The time phase difference is used to represent the degree of misalignment between the temperature response and the power generation response on the time axis. When the time phase difference is large, it means that the benefits of photovoltaic power generation can be realized earlier, while the benefits of shading and cooling need to go through a longer thermal response process to be realized. The control period is divided into a power generation priority interval and a cooling priority interval, so that the immediate sensitivity of photovoltaic output is given priority in the former stage, and the cooling sensitivity during the release of thermal response is given priority in the latter stage, so that the two objectives are controlled in the time period when the benefits are most easily amplified.

[0050] By using time misalignment, the multi-objective optimization problem, which originally required trade-offs at every moment, is transformed into a phased single-objective decision-making problem. This is achieved by combining continuous time with misalignment. By comparison, when the system operates within the misalignment range, the power generation response is considered to be dominant; otherwise, the temperature hysteresis is considered to be fully manifested. Therefore, a piecewise function is used to divide the time axis into a power generation priority interval and a cooling priority interval, introducing a time-driven function, expressed as: ; Among them when When the control system enters the power generation priority stage, which corresponds to the time interval in which the power generation response appears earlier than the temperature response, the photovoltaic output can quickly reflect the changes in the sunshade's posture. By using the optimal power generation angle as the primary objective, more definite power generation benefits can be obtained in a short period of time.

[0051] when When the system enters the sunshade priority phase, which corresponds to the time interval of gradual temperature response, the contribution of continuous shading to the decrease in vehicle interior temperature begins to be concentrated. Therefore, using the optimal sunshade angle as the primary target can make full use of the sunshade effect during the heat response release phase.

[0052] Through phased control, the power generation target and the shading and cooling target are optimized independently in time intervals that are more sensitive to dynamic response, rather than being compromised at the same time.

[0053] To enable the upper-level controller to not only know the current stage but also predict the switching time to plan the actuator actions in advance, the stage indication function and the switching time are combined to form a control output with time structure characteristics, ultimately constructing: ; in, The timing of the phase transition clearly indicates the point at which the time structure of the control strategy switches.

[0054] By analyzing the dynamic differential characteristics of the two types of controlled physical processes, representative response moments are extracted for each. The difference between these two moments is then used as the sole basis for dividing the control stages. The determination of the boundary is entirely determined by the system's current thermal inertia and light response speed, exhibiting real-time adaptability. When the ambient temperature rises rapidly or the photovoltaic response shows a delay, [the system will adapt accordingly]. and It will move spontaneously, making Dynamic adjustments and a corresponding re-division of the control phase transform multi-objective conflicts from "weight compromise" to "time separation," enabling the orderly succession of different objectives in the time domain, rather than their mutual compromise at the same instant.

[0055] S3. Calculate the restricted optimal tilt angle of the sunshade based on the target attitude and the real-time solar altitude angle.

[0056] Furthermore, the solar incidence direction is calculated based on the real-time solar altitude angle and the vehicle's current attitude, and the initial theoretical tilt angle is determined in conjunction with the phased control commands.

[0057] Based on the time-layered control strategy, the power generation priority zone and the cooling priority zone are corresponding to different tilt angle control stages.

[0058] Within the priority power generation zone, the tilt angle of the sunshade is adjusted according to the angle between the surface normal of the photovoltaic module and the direction of solar incidence, so that the angle approaches zero in order to obtain the maximum incident irradiance, thereby determining the first tilt angle.

[0059] Within the cooling priority zone, the tilt angle of the sunshade is adjusted based on the correspondence between the coverage area of ​​the sunshade projected on the vehicle roof and the rate of change of the vehicle interior temperature, so as to maximize the shaded area inside the vehicle and thus determine the second tilt angle.

[0060] The switching time between the first and second dip angles is determined by the time phase difference, and the dip angle adjustment is performed in stages according to the switching time.

[0061] Based on the time phase difference and the response time of the adjustment mechanism, the first tilt angle and the second tilt angle are corrected in time sequence so that the environmental state corresponding to the time of adjustment completion matches the target control stage.

[0062] During the tilt adjustment process, the equivalent wind load on the surface of the sunshade is calculated based on the vehicle speed and ambient wind force, and the tilt angle is limited to the safe tilt angle range corresponding to the wind load. When the wind load exceeds the preset threshold, the tilt adjustment corresponding to the safety constraint is executed first.

[0063] It should be noted that in the process of controlling the angle of the sunshade, the power generation efficiency and the sunshade cooling effect correspond to different physical mechanisms. If a single angle is used to take into account both at the same time, both types of performance will be compressed. Angle models are established for maximizing power generation and maximizing sunshade respectively, and then coupled through time structure.

[0064] For the power generation process, the radiant energy received by the photovoltaic module depends on the angle between the solar incident direction and the normal direction of the module surface. Based on the classic cosine projection relationship, and considering the sunshade as an adjustable plane, the normal direction... It's an angle. The function, in order to maximize power generation, uses the solar direction vector By performing an inner product operation with the adjustable normal and maximizing the inner product, we can amplify the physical effect that maximizes energy capture when sunlight is incident perpendicularly. This allows us to find the angle that brings the instantaneous photovoltaic output power closer to the theoretical peak, suppressing the cosine loss caused by non-perpendicular incidence, and constructing the expression: ; in, Let be the unit vector pointing towards the sun. Let be the unit vector normal to the sun visor surface, and be the angle. The function, It is equal to the cosine of the angle of incidence. This refers to the angle of the shading panel that maximizes its internal area, which is also the optimal posture for photovoltaic power generation.

[0065] The angle is used in the power generation priority stage. Since the photovoltaic output responds quickly to attitude changes during the stage, the change in the angle between the solar incident direction and the normal of the photovoltaic module can be directly reflected as the change in effective incident irradiance. Therefore, by maximizing the inner product of the two, the photovoltaic module can obtain the maximum incident energy in the power generation sensitive range. Within the time interval, the power generation benefit is most sensitive to the tilt angle change. Therefore, selecting the optimal photovoltaic angle can maximize the control benefit of the stage.

[0066] During the sun shading process, the temperature rise inside the vehicle mainly stems from solar radiation passing through the windows and being absorbed by the interior surfaces. To minimize the amount of radiation entering the vehicle, the sun shading problem is transformed into a geometric projection problem. The goal is to find the orientation that maximizes the projected area of ​​the sun visor on the plane perpendicular to the sunlight, thereby maximizing the interception and reflection of radiation and suppressing heat entering the vehicle. This establishes the optimal sun shading angle. ; in, The area of ​​the sunshade is the projected area of ​​the sunshade on the plane perpendicular to the sunlight, and the area varies with the angle. change, This is the angle with the largest projected area, which is also the optimal posture for shading and cooling.

[0067] Since the temperature response inside the vehicle gradually changes over time, the change in the projected area of ​​the sunshade will continuously affect the amount of solar radiation entering the vehicle and be reflected in the subsequent temperature drop. By maximizing the projected area of ​​the sunshade, heat input can be minimized within the temperature-sensitive range, making the sunshade action more effectively converted into actual cooling benefits.

[0068] After obtaining two types of perspectives, weighted fusion is no longer performed because weighting is essentially a compromise and cannot fully utilize the characteristics of each. A time-driven function is then used instead. By implementing structural coupling and considering the electromechanical delay between the motor receiving the command and its actual position, a forward compensation is applied to the time axis to advance the target angle for the future moment. This ensures that the actuator arrives at the desired position at the expected time, thereby suppressing the disruption of synchronization caused by lag and outputting a dynamic angle command. ; in, For the response delay time of the motor actuator, Indicates in The optimal angle for power generation that should be reached at any given time. To achieve the optimal shading angle, time-shift compensation is used to resynchronize the control commands with actual environmental changes.

[0069] By directly applying the time control structure to the tilt angle output, when the system is in the power generation priority stage, the shading canopy no longer takes into account the shading projection area, but directly adopts the optimal photovoltaic angle; when the system is in the shading priority stage, the shading canopy no longer takes into account the instantaneous power generation efficiency, but directly adopts the optimal shading angle. In this way, the control objectives in each stage remain single and clear, avoiding the weakening of objectives caused by weighted compromise.

[0070] At the same time, introduce To compensate for the response delay between receiving control commands and the actual positioning of the regulating mechanism, ensuring that the moment the sunshade reaches the target angle matches the corresponding control stage, the system calculates the angle not only based on the current moment but also based on the future moment after the mechanism's response. This guarantees that "the power generation angle is reached during the power generation stage, and the sunshade angle is reached during the sunshade stage." To prevent the mechanical structure from being impacted by a large step change in angle caused by stage switching, the variation range of the output angle between adjacent control cycles is limited to keep the inertial load within the structure's tolerance range, thereby suppressing the impact energy caused by excessive angular velocity. A variation constraint is introduced: ; in, and These represent the command angles for two adjacent control cycles. To limit the maximum allowable angle change amplitude, the structural impact caused by rapid changes is suppressed.

[0071] Considering the overturning effect of wind loads on large-area sunshades, when the windward area of ​​the panel is too large, strong winds may overload the mechanism. By combining the current wind speed and direction, a safe tilt angle upper limit is calculated in real time, and the upper limit is forcibly truncated at the output end to suppress any movement that may exceed the structural safety boundary, thereby avoiding the risk of wind-induced overload. Hard constraints are applied. ; in, Based on the current wind speed and wind direction The upper limit of the safe angle calculated in real time represents the maximum allowable tilt angle that makes the wind load on the slab surface lower than the safe threshold. This imposes a hard constraint on the system output, thereby ensuring structural safety.

[0072] Based on the independent physical principles of power generation and shading, the optimal single target angle for each is derived. These two angles are usually quite different and irreconcilable. Instead of seeking a numerical compromise between the two, the time-driven signal constructed in the previous step is used to fully pursue the optimal angle for power generation during the power generation priority period and the optimal angle for shading during the cooling priority period. Both objectives can be achieved almost to the extreme in their respective time segments, while achieving overall balance on the time scale. The forward compensation for motor delay, the rate limit for step impact, and the hard cut-out of wind safety margin constitute three layers of protective constraints. These constraints act on the time advance of command generation, action rate, and final output boundary, respectively, to ensure the safety and smoothness of physical execution in a progressive manner. This allows the sunshade to pursue maximum energy efficiency without causing structural risks due to ignoring mechanical boundaries.

[0073] A continuous control chain of "time phase recognition - stage target determination - physical optimal angle output" is formed. The time phase difference is used to determine the response sequence of different targets in the time dimension, the time driving function is used to determine the current control stage, and the photovoltaic optimal angle and the shading optimal angle are used to achieve single-target optimal control in each stage, so that the power generation target and the shading and cooling target reach the optimal in their respective more sensitive time intervals, rather than compromising with each other at the same time.

[0074] S4. The drive adjustment mechanism performs tilt adjustment, and during the adjustment process, it identifies the risk of obstruction based on changes in motor current and resistance torque, triggering reverse retraction and safety lock.

[0075] Furthermore, by collecting and adjusting motor current signals, mechanism position feedback, and resistance torque estimation data, the operating status of the adjusting mechanism is comprehensively reflected, and joint analysis is performed to identify abnormal resistance characteristics and form a blocking judgment result.

[0076] When a risk of obstruction is detected, the control and adjustment mechanism executes reverse drive and locks the position to achieve safe retraction and protection.

[0077] It should be noted that during the automatic adjustment of the sun visor, if there is an external obstacle in the movement path, the obstruction cannot be directly identified by relying solely on the angle and position feedback, because the position sensor can only indicate the current angle and cannot distinguish between "normal arrival" and "being blocked by a foreign object". By utilizing the electromagnetic and motion relationship inside the drive motor, an obstruction identification mechanism that does not rely on external sensors is constructed.

[0078] In the basic dynamic equations of a permanent magnet DC motor, the motor output torque is proportional to the armature current, while the load resistance torque is related to the speed and friction. Based on the original torque balance relationship, considering that the inertia of the sun visor is small and the movement is slow during normal operation, the acceleration torque term can be ignored, and the friction and normal transmission resistance are merged into the part proportional to the speed. To highlight the external abnormal resistance, the difference between the current term representing the driving force and the speed term representing normal consumption is taken, so that the difference is kept low during normal unobstructed operation. Once an external obstacle appears, the current rises sharply and the speed drops sharply, and the difference is amplified drastically. Thus, a resistance torque estimation expression that is highly sensitive to abnormal obstruction is constructed: ; in, For the motor armature current, This refers to the motor speed. As a torque constant, the current is mapped to the driving torque. The damping coefficient maps the rotational speed to normal resistance torque consumption. This is the estimated external additional resistance torque. By subtracting the driving force from the normal consumption, the numerical representation of abnormal loads is amplified.

[0079] To distinguish between persistent obstruction and instantaneous impact, and to avoid misjudgment caused by momentary spikes, a time-difference calculation of the drag torque is performed. The current drag torque is compared with its value from a short time ago. Only when the sudden increase exceeds a set threshold is it considered persistent jamming. This amplifies the characteristics of continuous stress accumulation while suppressing short-term interference, thus establishing the discrimination criteria. ; in, This is the estimated value of the resistance torque at the current moment. For a tiny time interval The condition is used to identify persistent resistance, rather than transient disturbances.

[0080] Upon detecting persistent obstruction, a protective retraction action is immediately executed. This involves actively rotating the sun visor angle slightly in the opposite direction along its original path to release contact pressure and transfer mechanical stress from the jamming point, thereby preventing structural damage. ; in, For each small angular step of regression, From the current command angle, the process does not rely on external sensors, but only achieves obstruction recognition and structural self-protection through the coupling analysis of motor current and speed.

[0081] The motor is viewed as both a driver and a "tactile sensor"—the combination of current and speed can indirectly reflect abnormal changes in the load without the need to add pressure switches or infrared proximity sensors to the outside of the sun visor.

[0082] By building This estimation formula, which combines electrical and mechanical quantities, allows the system to extract an estimated value of the resisting torque in real time. Under normal micro-movement conditions, this value is close to zero. However, when jamming occurs, the current surges dramatically and the rotational speed drops sharply due to the obstruction of an object, immediately amplifying the difference. Therefore, a time difference is introduced. It effectively filters out current spikes or short-term fluctuations caused by accidental collisions at the moment of motion start-up, and only reacts to continuous jamming, thus ensuring high sensitivity while having strong anti-interference ability.

[0083] Step length of the retreating motion To achieve flexible protection that is far below the limit of mechanical damage, allowing for "instant retraction" upon contact, this approach utilizes the physical variables of the motor itself for sensorless obstacle detection. It transforms the classic motor torque balance equation into a real-time obstacle detector, simplifying the system hardware and improving anomaly response speed.

[0084] S5. After the adjustment is completed, collect the actual cooling effect, actual power generation and actual regulation energy consumption, and adaptively update the decision model based on the actual cooling effect, actual power generation and actual regulation energy consumption.

[0085] Furthermore, statistics were collected on the changes in vehicle interior temperature, photovoltaic power generation output, and regulation energy consumption after sunshade adjustment to obtain actual operating data. This data was then compared and analyzed with the corresponding predicted benefits to calculate the benefit deviation.

[0086] The parameters of the decision-making model are adjusted based on the profit deviation, thereby improving the accuracy and adaptability of subsequent control strategies.

[0087] It should be noted that in dynamic environments, changes in light intensity, temperature fluctuations, and vehicle usage patterns all drift over time. If control parameters remain fixed, long-term operation will inevitably deviate from the optimal state. By introducing an adaptive update mechanism based on benefit feedback, the shading strategy can have the ability to self-adjust under time-varying environments.

[0088] The prediction model refers to the revenue prediction model pre-stored in the controller. The inputs are the current environmental state vector, target attitude, constrained optimal tilt angle and control duration, and the outputs are the predicted cooling revenue, predicted power generation revenue and predicted regulation energy consumption corresponding to this shading adjustment action within the prediction time window.

[0089] The predictive model does not exist independently of the control process, but corresponds to the decision model. It is used to estimate the potential benefits of different candidate control actions before performing sunshade attitude adjustment. The predictive model establishes a mapping relationship between environmental state, sunshade attitude and actual operation effect based on historical control data, thereby predicting the comprehensive benefits that a certain target attitude may bring within the current control cycle.

[0090] To avoid the prediction model outputting only a single abstract benefit value, the benefit is broken down into three parts: cooling benefit, power generation benefit, and energy consumption regulation benefit, which are calculated separately. For the actual cooling benefit, instead of using a certain instantaneous temperature difference as the evaluation result, the temperature change inside the vehicle is continuously collected within the observation time window after the sunshade completes its attitude adjustment. The difference between the baseline temperature before adjustment and the real-time temperature after adjustment is accumulated over time to obtain the actual cooling benefit that can reflect the continuous cooling effect.

[0091] Considering that instantaneous values ​​of control effects are easily affected by accidental factors, such as a brief temperature drop caused by a single gust of wind, they cannot represent the true quality of the control strategy. To establish a robust evaluation standard, the cooling effect is integrated over time, continuously accumulating the temperature drop over each minute interval. This is equivalent to amplifying the cumulative effect of long-term continuous cooling, suppressing evaluation distortion caused by short-term fluctuations, and thus ensuring that the cooling benefit truly reflects the strategy's thermal management capability over the complete cycle. The following is constructed: ; in, The interior temperature before implementing the sunshade strategy, The temperature after execution. This is the single-step cooling amount. The cumulative cooling gain after time integration is expressed as temperature multiplied by time.

[0092] For actual power generation revenue, within the same observation time window after the sunshade adjustment is completed, the real-time output power of the photovoltaic modules is collected, and the output power is integrated over time to obtain the actual power generation. This can reflect the continuous power generation capability of the photovoltaic modules under the target posture, rather than judging the posture quality based solely on a certain instantaneous power.

[0093] By integrating the power generation process over time and accumulating the real-time photovoltaic power over time, the power generation benefit is unified with the cooling benefit into a time-cumulative form in the energy dimension. This provides a unified benchmark for direct comparison between the two heterogeneous objectives in subsequent evaluations, thus constructing a power generation benefit index: ; in, For real-time power generation, This refers to the cumulative energy generated.

[0094] The actual energy consumption for adjustment is calculated by collecting the voltage, current, and duration of the adjustment motor during the attitude adjustment process. Since the sunshade adjustment action itself consumes power battery energy, the energy consumption needs to be included as a revenue deduction item in the actual comprehensive revenue calculation. The expression is as follows: ; in, The moment when the regulating mechanism begins to operate. The timing of the adjustment mechanism completing its action. To adjust the real-time drive voltage of the motor, To adjust the real-time drive current of the motor, To effectively regulate energy consumption, treating energy consumption regulation as a deduction item can prevent the system from frequently adjusting the shading posture to obtain smaller cooling or power generation benefits, thereby improving the actual usability of the control strategy. After obtaining the actual cooling benefit, actual power generation benefit, and actual regulation energy consumption, these three are combined to form the actual comprehensive benefit. Since the core of this scheme does not rely on fixed weights to achieve a trade-off between power generation and cooling, the actual comprehensive benefit is mainly used for model feedback correction, rather than replacing the aforementioned time-phased control structure. The expression is: ; in, For actual comprehensive benefits, For actual cooling benefits, For actual power generation revenue, To actually regulate energy consumption, the actual comprehensive benefits are used to evaluate the true execution effect of this control action and serve as the data basis for subsequent model correction.

[0095] Corresponding to the actual comprehensive benefit, before the control action is executed, the prediction model outputs the predicted cooling benefit, predicted power generation benefit, and predicted regulation energy consumption based on the current environmental state vector, target attitude, constrained optimal tilt angle, and expected control duration, and combines them to obtain the predicted comprehensive benefit, expressed as follows: ; in, To predict overall returns, To predict the benefits of cooling, To predict power generation revenue, To predict and regulate energy consumption.

[0096] The predicted cooling benefits are estimated by the prediction model based on the current in-vehicle temperature, light intensity, shading posture, and historical cooling response data. The predicted power generation benefits are estimated by the prediction model based on the current light intensity, solar incidence direction, photovoltaic module posture, and historical power generation data. The predicted regulation energy consumption is estimated by the prediction model based on the current angle, target angle, and historical motor energy consumption data.

[0097] By comparing the actual comprehensive return with the predicted comprehensive return, the expression for the return deviation is obtained as follows: ; in, For actual cumulative revenue, The cumulative return predicted based on the current parameter model. This is the profit deviation, when A value greater than zero indicates that the actual effect of the control action is better than the predicted result; when... When the value is less than zero, it indicates that the actual effect of the current control action is lower than the predicted result. The profit deviation reflects the estimation error of the prediction model in the relationship between the current environmental state and the shading posture.

[0098] The sign and magnitude of the error directly reflect whether the strategy under the existing parameters is better or worse than expected, that is, the degree of mismatch between the model and the real environment. Through the difference, the evaluation information of the adaptability of the control parameters to the external environment is amplified, enabling the system to perceive the direction of the superiority or inferiority of its own strategy.

[0099] Using the deviation as a driving signal, the key angle bias or threshold in shading control is iteratively corrected. When the actual benefit is better than the prediction, the error is positive, and the parameters are fine-tuned in the direction of the current strategy to solidify the advantage. When the actual benefit is worse than the prediction, the error is negative, and the parameters are adjusted in the opposite direction to weaken the undesirable bias. Through repeated iterations, the parameters gradually converge towards a better equilibrium point, and an update mechanism is introduced. ; in, These are the original control parameters. To update the step size and control the magnitude of each adjustment, These are the updated parameters.

[0100] Instead of relying on precise mathematical models of the external environment or pre-calibrated operating condition tables, it smooths short-term fluctuations into long-term cumulative effects through continuous time integration, thereby obtaining a stable and reliable profit evaluation.

[0101] Utilizing the deviation between actual returns and model-predicted returns The system obtains a quantitative indication of whether the current strategy is still applicable, internalizing the slow drift of the environment into a continuous error signal, and then updating it proportionally. Gradually move the parameters.

[0102] By adjusting the internal parameters of the control strategy, the overall strategy evolves slowly in a direction that can generate better cumulative returns. Unlike feedback that relies solely on instantaneous errors, the long-term evaluation constructed through integration filters out the interference of short-term abnormal events, making the parameter adjustment direction more robust. This enables continuous performance self-improvement during long-term operation without the need for manual recalibration.

[0103] The predictive model provides a predicted comprehensive benefit before the action is executed. The system calculates the actual comprehensive benefit after the action is executed, and then corrects the predictive model parameters by the benefit deviation between the two. The adaptive update is not simply adjusting the angle based on a certain instantaneous temperature or instantaneous power, but rather performing closed-loop correction based on the cooling benefit, power generation benefit, and energy consumption adjustment within the complete control cycle, so that the model gradually adapts to the real operating state under different seasons, different lighting conditions, and different vehicle usage habits.

[0104] This embodiment also provides an intelligent sunshade control system for electric tricycles, including: an environmental perception module, a strategy decision-making module, an attitude calculation module, a safety execution module, and a model optimization module.

[0105] The environmental perception module is used to acquire information such as light intensity, interior temperature, vehicle speed, remaining battery power, and ambient wind speed through onboard sensors, and to construct an environmental state vector.

[0106] The strategy decision module is used to call the decision model pre-stored in the controller, input the environmental state vector into the decision model, calculate the shading and cooling benefits, photovoltaic power generation benefits, energy consumption adjustment costs and wind resistance safety constraints based on the multi-objective evaluation function, and output the target attitude.

[0107] The attitude calculation module is used to calculate the constrained optimal tilt angle of the sunshade based on the target attitude and the real-time solar altitude angle.

[0108] The safety execution module is used to drive the adjustment mechanism to perform tilt adjustment, and during the adjustment process, it identifies the risk of obstruction based on changes in motor current and resistance torque, triggering reverse retraction and safety locking.

[0109] The model optimization module is used to collect actual cooling effect, actual power generation, and actual regulation energy consumption after the adjustment is completed, and to adaptively update the decision model based on the actual cooling effect, actual power generation, and actual regulation energy consumption.

[0110] This embodiment also provides a computer device applicable to the intelligent sunshade control method for electric tricycles, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent sunshade control method for electric tricycles as proposed in the above embodiment.

[0111] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0112] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the intelligent sunshade control method for electric tricycles as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0113] In summary, this invention transforms conflicting constraints such as shading and cooling, photovoltaic power generation, energy consumption regulation, and wind resistance safety into a calculable quantitative game by calling a pre-stored decision model and based on a multi-objective evaluation function. This achieves global optimality among comfort, economy, and safety, effectively avoiding the control defects of sacrificing one aspect for another. Furthermore, by combining real-time solar altitude angle, the abstract target attitude is transformed into a constrained optimal tilt angle, ensuring the precise physical attainment and engineering implementation of the theoretical optimal solution.

[0114] 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 intelligent sunshade control of an electric tricycle, characterized in that: include: By acquiring information on light intensity, interior temperature, vehicle speed, remaining battery power, and ambient wind speed through onboard sensors, an environmental state vector is constructed. The decision model pre-stored in the controller is invoked, the environmental state vector is input into the decision model, and the shading and cooling benefits, photovoltaic power generation benefits, energy consumption adjustment costs and wind resistance safety constraints are calculated based on the multi-objective evaluation function, and the target attitude is output. Based on the target attitude and the real-time solar altitude angle, the restricted optimal tilt angle of the sunshade is calculated; The drive adjustment mechanism performs tilt adjustment, and during the adjustment process, it identifies the risk of obstruction based on changes in motor current and resistance torque, triggering reverse retraction and safety lock-up; After the adjustment is completed, the actual cooling effect, actual power generation, and actual regulation energy consumption are collected, and the decision model is adaptively updated based on the actual cooling effect, actual power generation, and actual regulation energy consumption.

2. The intelligent sunshade control method for electric tricycles as described in claim 1, characterized in that: The environment state vector includes: The data on light intensity, in-vehicle temperature, vehicle speed, remaining battery power, and ambient wind force are filtered, smoothed, time-synchronized, and normalized to eliminate interference and unify the dimensions. The wind force information is decomposed into wind speed and wind direction components. The processed parameters are combined and encoded to form an environmental state vector for subsequent decision-making.

3. The intelligent sunshade control method for electric tricycles as described in claim 2, characterized in that: The decision-making model includes: An initial decision model is established by training based on historical environmental state data and corresponding shading adjustment results; features are extracted from the input environmental state vector, and a set of candidate shading postures is obtained by matching through the decision model. The candidate poses are evaluated for comprehensive benefits and the optimal result is selected to determine the target pose.

4. The intelligent sunshade control method for electric tricycles as described in claim 3, characterized in that: The multi-objective evaluation function includes: Based on the current environmental state vector and historical change sequence, the temperature response delay, which characterizes the thermal inertia hysteresis characteristics in the vehicle, and the power generation response delay, which characterizes the instantaneous characteristics of photovoltaic photoelectric conversion, are extracted. Based on the temperature response delay and the power generation response delay, the time phase difference, which reflects the degree of resource competition between thermal regulation and photoelectric conversion in the time dimension, is determined. When the time phase difference exceeds a preset threshold, a time-layered control strategy including a power generation priority zone and a cooling priority zone is constructed. In the power generation priority zone, the first target attitude is output with the constraint of maximizing instantaneous power generation. In the cooling priority zone, the second target attitude is output with the constraint of maximizing cumulative cooling, thereby forming a multi-objective coordinated control strategy based on time layering. Based on the aforementioned time phase difference, feedforward compensation is performed on the evaluation time of the cooling target's benefits to offset the impact of heat transfer lag on control accuracy, so that the cooling effect and power generation effect reach Pareto optimality under a unified time reference. Based on the time-layered control strategy and time compensation mechanism, the target attitude and its switching timing parameters corresponding to each control time interval are determined, thereby outputting a phased control command that includes time interval division, target attitude sequence and switching time.

5. The intelligent sunshade control method for electric tricycles as described in claim 4, characterized in that: The constrained optimal tilt angle includes: The solar incidence direction is calculated based on the real-time solar altitude angle and the vehicle's current attitude, and the initial theoretical tilt angle is determined in conjunction with the phased control commands. Based on the time-layered control strategy, the power generation priority interval and the cooling priority interval are corresponding to different tilt angle control stages; Within the priority power generation zone, the tilt angle of the sunshade is adjusted according to the angle between the surface normal of the photovoltaic module and the direction of solar incidence, so that the angle approaches zero in order to obtain the maximum incident irradiance, thereby determining the first tilt angle; Within the cooling priority zone, the tilt angle of the sunshade is adjusted according to the correspondence between the coverage area of ​​the sunshade projected on the vehicle top and the rate of change of the temperature inside the vehicle, so as to maximize the shaded area inside the vehicle and thus determine the second tilt angle. The switching time between the first tilt angle and the second tilt angle is determined by the time phase difference, and the tilt angle adjustment is performed in stages according to the switching time; Based on the time phase difference and the response time of the adjustment mechanism, the first tilt angle and the second tilt angle are time-corrected so that the environmental state corresponding to the time of adjustment completion matches the target control stage. During the tilt adjustment process, the equivalent wind load on the surface of the sunshade is calculated based on the vehicle speed and ambient wind force, and the tilt angle is limited to the safe tilt angle range corresponding to the wind load. When the wind load exceeds the preset threshold, the tilt adjustment corresponding to the safety constraint is executed first.

6. The intelligent sunshade control method for electric tricycles as described in claim 5, characterized in that: The identification of the obstruction risk includes: Collect and adjust motor current signals, mechanism position feedback and resistance torque estimation data to comprehensively reflect the operating status of the regulating mechanism, perform joint analysis to identify abnormal resistance characteristics, and form a blocking judgment result; When a risk of obstruction is detected, the control and adjustment mechanism executes reverse drive and locks the position to achieve safe retraction and protection.

7. The intelligent sunshade control method for electric tricycles as described in claim 6, characterized in that: The adaptive update includes: The changes in vehicle interior temperature, photovoltaic power generation output, and regulation energy consumption after sunshade adjustment were statistically analyzed to obtain actual operating data, which were then compared with the corresponding predicted benefits to calculate the benefit deviation. The parameters of the decision-making model are adjusted based on the profit deviation, thereby improving the accuracy and adaptability of subsequent control strategies.

8. An intelligent sunshade control system for an electric tricycle, based on the intelligent sunshade control method for an electric tricycle according to any one of claims 1 to 7, characterized in that: It includes an environmental perception module, a strategy decision-making module, an attitude calculation module, a safety execution module, and a model optimization module; The environmental perception module is used to acquire information on light intensity, interior temperature, vehicle speed, remaining battery power, and ambient wind speed through onboard sensors, and to construct an environmental state vector. The strategy decision module is used to call the decision model pre-stored in the controller, input the environmental state vector into the decision model, calculate the shading and cooling benefits, photovoltaic power generation benefits, energy consumption adjustment costs and wind resistance safety constraints based on the multi-objective evaluation function, and output the target attitude. The attitude calculation module is used to calculate the limited optimal tilt angle of the sunshade based on the target attitude and the real-time solar altitude angle. The safety execution module is used to drive the adjustment mechanism to perform tilt adjustment, and during the adjustment process, it identifies the risk of obstruction based on the changes in motor current and resistance torque, triggering reverse retraction and safety locking. The model optimization module is used to collect the actual cooling effect, actual power generation, and actual regulation energy consumption after the adjustment is completed, and to adaptively update the decision model based on the actual cooling effect, actual power generation, and actual regulation energy consumption.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the intelligent sunshade control method for electric tricycles according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the intelligent sunshade control method for electric tricycles according to any one of claims 1 to 7.

Images