Distributed heat supply and heat collection tank layered control system and method based on image recognition and PLC

Abstract

The invention provides a distributed heat supply heat collecting groove layered control system and method based on image recognition and a PLC. According to the method, rough adjustment and fine adjustment are combined, the position of a heat collecting groove is collected through an absolute value encoder, the target tracking angle of the heat collecting groove is calculated based on the SPA algorithm, and the heat collecting groove is controlled by the PLC through a driving mechanism to rotate for rough adjustment. Meanwhile, a machine vision image processing technology is used for recognizing the projection contour of the plastic rod on a receiving panel, image preprocessing is carried out through image denoising, graying and binarization, edge detection is carried out through a Canny algorithm, the center of projection of the carbon rod is determined through a minimum circumcircle method, and the tracking angle and the rotation direction of the heat collecting groove are calculated; and the heat collecting groove is controlled to rotate for fine adjustment, so that the sun is always focused on the heat collecting pipe. According to the invention, the tracking precision and stability of the trough collector to the sun position can be improved, and the utilization efficiency of solar energy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed heating systems, and in particular to a distributed heating collector trough hierarchical control system and method based on image recognition and PLC. Background Art

[0002] Traditional heating methods, especially centralized heating with coal as the main fuel, will produce a large amount of pollutants. Pollutants such as sulfur dioxide, nitrogen oxides, and particulate matter will be released during the combustion of coal. These pollutants are important factors in causing environmental problems such as acid rain and haze. Distributed clean energy heating can greatly reduce the emission of these pollutants.

[0003] As the global energy crisis and environmental pollution become increasingly serious, solar energy, as a clean and renewable energy source, has received more and more attention. Solar energy is an almost inexhaustible clean energy source. Distributed solar thermal heating systems directly use solar energy to obtain heat, reducing dependence on traditional fossil energy (such as coal and natural gas).

[0004] Distributed solar thermal heating systems are relatively independent, and the heating network composed of multiple distributed systems will not cause large-scale heating interruptions due to single-point failures (such as damage to large central heating equipment). In extreme weather or tight energy supply, it can be used as a supplementary heating method to improve the safety and reliability of the entire society's energy supply. For some companies or users, distributed energy systems can also achieve energy self-sufficiency and reduce the cost of purchasing energy from the power grid or heating pipeline network. Distributed clean energy heating systems can be flexibly configured according to user needs. In some remote areas, where centralized heating pipelines are difficult to cover, distributed clean energy heating can provide independent heating solutions for these areas to meet residents' heating needs.

[0005] Distributed solar thermal heating generally uses a collector trough to track the sun's position, that is, the solar collector converts solar radiation energy into heat energy for building heating. It achieves the use of solar energy by focusing sunlight to heat the fluid in the collector tube. However, since the position of the sun is constantly changing, in order to improve the solar energy utilization efficiency of the trough collector, the sun's position needs to be accurately tracked.

[0006] There are two main methods for tracking the sun's position in traditional trough collectors: one is a closed-loop tracking method based on sensors, and the other is an open-loop tracking method based on astronomical algorithms. Sensor-based tracking methods usually use sensors such as photoelectric sensors and photoresistors to detect the position of the sun, and then adjust the angle of the collector trough through a control system so that the sun is always focused on the collector tube. The advantage of this method is high tracking accuracy, but the sensor is easily affected by environmental factors such as clouds and dust, which leads to reduced tracking accuracy. The tracking method based on astronomical algorithms usually determines the angle of the collector trough through calculation according to the relationship between the sun's position and time, and then adjusts the angle of the collector trough through a control system so that the sun is always focused on the collector tube. The advantage of this method is that it is not affected by environmental factors, but the calculation is large and the tracking accuracy is low.

[0007] The comparison with the existing technology is as follows:

[0008] Technical comparison with patent CN111207525A "Intermittent driving method and device for the heat collecting trough of trough-type solar thermal solar system"

[0009] The present invention adopts a hierarchical control method based on image recognition and PLC, performs coarse adjustment through an absolute value encoder, and uses machine vision for fine adjustment to achieve accurate tracking of the sun's position; while CN111207525A uses an intermittent drive method, focuses on the judgment and control of the real-time tracking angle, and emphasizes staged intermittent drive. There are essential differences between the two in terms of control strategy and implementation method.

[0010] The present invention focuses on using machine vision technology for image processing and fine adjustment, focusing on improving the efficiency of solar energy utilization; while CN111207525A improves the control stability of the heat collecting tank by calculating the angle difference and combining the dead zone and hysteresis angle, focusing on reducing energy consumption and ensuring stability. There are obvious differences in the focus and purpose of the two technologies.

[0011] The present invention improves the tracking accuracy and stability of the trough collector to the sun's position through image processing and PLC control, emphasizing the efficient use of solar energy; while CN111207525A reduces energy consumption and improves control stability through intermittent driving, focusing on the energy efficiency management of the system. There are essential differences between the two in terms of the final effect and goal. Summary of the invention

[0012] In order to solve the above technical problems, the present invention proposes a distributed heating collector trough hierarchical control system and method based on image recognition and PLC. The method adopts a combination of coarse adjustment and fine adjustment, collects the collector trough position through an absolute encoder, calculates the collector trough target tracking angle based on the SPA algorithm, and uses PLC to control the collector trough rotation through a driving mechanism for coarse adjustment. At the same time, machine vision image processing technology is used to identify the projection contour of the plastic rod on the receiving panel, and the rotation angle of the collector trough is controlled for fine adjustment, so that the sun is always focused on the collector tube, improving the tracking accuracy and stability of the trough collector to the sun's position, and improving the utilization efficiency of solar energy. Thereby improving the safe, stable and reliable operation level of the distributed solar thermal heating system.

[0013] To achieve the above object, the technical solution adopted by the present invention is:

[0014] The distributed heating collector trough hierarchical control system based on image recognition and PLC includes: a collector trough, a PLC controller, an absolute value encoder, a black carbon rod, a shadow receiving resin panel, an image acquisition device, an image processing module, and a driving mechanism. The collector trough is driven to rotate by a driving mechanism installed at its bearing. The PLC, as the control center, calculates the sun's position and the target tracking angle of the collector trough based on the solar position algorithm SPA, while receiving data from the absolute value encoder and adjusting the angle of the collector trough through the driving mechanism. The black carbon rod, the shadow receiving resin panel, and the image acquisition device are fixed directly above the collector trough through a bracket. The image acquisition device is used to capture the projection of the carbon rod on the shadow receiving resin panel, and identify the projection position through the image processing module. Image denoising, graying, and binarization are used for image preprocessing, and the Canny algorithm is used for edge detection. The minimum circumscribed circle method is used to determine the center of the carbon rod projection, calculate the collector trough tracking angle and rotation direction, and transmit them to the PLC controller to control the rotation of the collector trough.

[0015] The image processing module uses a Raspberry Pi development board, and the camera is inserted into the CSI interface on the Raspberry Pi mainboard through a flexible cable;

[0016] The shadow receiving resin panel must be arranged parallel to the opening plane of the heat collecting tank, and the midline of the shadow receiving resin panel is located in the plane determined by the normal vector of the heat collecting tank and the heat collecting tube. The shadow receiving resin panel has a rectangular shape and a light-colored outer surface. The carbon rod must be arranged parallel to the heat collecting tube of the heat collecting tank and will not be blocked by any nearby objects.

[0017] The driving mechanism is a DC motor or a hydraulic mechanism. The DC motor has at least two speed adjustment gears, high speed and low speed. For coarse adjustment, the driving motor rotates at high speed. For fine adjustment, the driving motor rotates at low speed to improve control accuracy.

[0018] The image processing module communicates with the PLC via a network connection, and the communication protocol is Modbus TCP;

[0019] The length and diameter of the carbon rod can be adjusted according to actual conditions;

[0020] The installation position and angle of the camera of the image acquisition device can be adjusted according to actual conditions to ensure that the image on the receiving panel can be clearly acquired.

[0021] The present invention provides a control method for a distributed heating collector trough layering system based on image recognition and PLC, and the specific steps are as follows:

[0022] S1: When the system starts working, it first performs coarse adjustment, obtains the current real-time position of the solar collector through the absolute encoder, and calculates the target tracking angle according to the SPA algorithm. The PLC drives the solar collector to rotate roughly to the target angle.

[0023] S2: When the actual position of the solar collector is close to the target tracking angle, the PLC determines whether the fine adjustment conditions are met. If the coarse adjustment error is within the set threshold range, the system automatically switches to the fine adjustment stage;

[0024] S3: The system starts the camera to capture the projection image of the plastic rod on the receiving panel. The image processing system calculates the deviation of the projection and controls the heat collecting tank for fine adjustment.

[0025] S4: During the fine-tuning process, the system continuously collects and processes images, and fine-tunes the angle of the heat collecting trough in real time to ensure that the sunlight is focused on the heat collecting tube;

[0026] S5: If the tracking error exceeds the preset range due to external factors, such as cloud cover and wind speed, the system will automatically stop fine adjustment and switch back to coarse adjustment mode, perform large angle adjustment before entering fine adjustment.

[0027] As a further improvement of the control method of the present invention, in step S1, the PLC is used to roughly control the heat collecting tank to roughly align with the sun. The system uses an absolute encoder to collect the current position information of the heat collecting tank, and calculates the azimuth and elevation of the sun through the SPA algorithm to obtain the target tracking angle. The PLC controls the driving mechanism to rotate the heat collecting tank to near the target tracking angle;

[0028] Specific steps:

[0029] S101: Use an absolute encoder to collect the current position information and current tracking angle of the heat collecting tank, and feed it back to the PLC in real time;

[0030] S102: Calculate the azimuth and elevation of the sun on the day through the SPA algorithm to obtain the target tracking angle of the solar trough;

[0031] S103: According to the difference between the target tracking angle and the current tracking angle, the PLC drives the motor to control the rotation of the heat collecting tank to complete the rough adjustment;

[0032] S104: When the difference of the tracking angle of the solar collector is reduced to a critical switching value, the system enters a switching judgment phase.

[0033] As a further improvement of the control method of the present invention, the specific steps of step S102: SPA algorithm are as follows:

[0034] S201: Establishing a solar position model: establishing a relationship model between the solar position and time and geographical location based on astronomical knowledge;

[0035] S202: Collecting current time and geographical location information: Collecting current time and geographical location information through a clock chip and a GPS module;

[0036] S203: Calculate the sun position: Calculate the sun's position in the sky based on the sun position model and the current time and geographic location information;

[0037] S204: Calculate the target tracking angle of the solar trough: Calculate the target tracking angle of the solar trough according to the sun's position and the geometric parameters of the solar trough.

[0038] As a further improvement of the control method of the present invention, the goal of the fine adjustment in step S4 is to ensure that the solar beam is always accurately focused on the heat collecting tube. The specific steps are as follows:

[0039] S401: The camera collects the projection image of the plastic rod on the receiving panel in real time;

[0040] S402: The image processing module performs image preprocessing, edge monitoring and projection center calculation;

[0041] S403: Calculating the required angle adjustment value according to the distance between the projection center and the center of the rectangular shadow receiving resin panel;

[0042] S404: When the required angle adjustment value is greater than the dead zone angle, the PLC controls the driving device to rotate the heat collecting tank according to the calculated fine-tuning angle, and further adjusts the angle of the heat collecting tank until the projection is accurately aligned with the heat collecting tube.

[0043] During the fine-tuning process, the system analyzes the changes in projection in real time and continuously makes fine adjustments to ensure that the solar trough and solar beam always maintain the best tracking accuracy.

[0044] As a further improvement of the control method of the present invention, the image processing step S402 is as follows:

[0045] (1) Image preprocessing;

[0046] Image denoising: Use Gaussian filtering or median filtering to denoise the image. Gaussian filtering can smooth the image and reduce random noise, while median filtering is better for removing salt and pepper noise.

[0047] Grayscale: The original image is usually a color image, and the color image is converted into a grayscale image. This makes image processing more efficient;

[0048] Binarization: By setting a threshold, the grayscale image is converted into a binary image. The adaptive threshold method can dynamically set the threshold according to the change of local image brightness, and perform better under different lighting conditions;

[0049] (2) Edge detection;

[0050] The Canny edge detection method is used to extract significant contour features in the projection image and identify the projection contour of the plastic rod. It has high edge detection accuracy and can effectively reduce false edges. Through the Canny algorithm, the projection contour of the plastic rod can be clearly detected, providing a basis for subsequent projection center calculation;

[0051] After edge detection, the coordinates of the edge pixels are stored in a data structure, which is an array or list. These points will be used for the subsequent minimum circumscribed circle calculation.

[0052] (3) Projection center calculation:

[0053] The minimum circumscribed circle method is used to determine the center of the projection. The minimum circumscribed circle that contains the shape is found. The center of the circle is the center of the shape. The algorithm for determining the minimum circumscribed circle is based on the calculation of geometric distance. By continuously adjusting the radius and center position of the circle, the circle can contain all edge points and have the smallest radius.

[0054] (4) Adjustment of the tracking angle of the heat collecting trough:

[0055] According to the center position of the determined shape, the distance between the center of the projection and the center of the rectangular shadow receiving resin panel is calculated. The angle formed between the normal vector of the receiving panel and the normal vector of the heat collecting tank is calculated as: The system calculates the required adjustment angle, and the PLC controls the driving mechanism to rotate the heat collecting tank to make a small angle rotation adjustment. The rotation direction should reduce the angle θ formed between the normal vector of the receiving panel and the normal vector of the heat collecting tank.

[0056] (5) Real-time feedback and fine-tuning loop:

[0057] During the adjustment process, the image processing system will continuously capture and analyze the projected image to form a closed-loop control. After each angle adjustment, the system will re-capture the projected image, perform the same image preprocessing, edge detection and shape matching steps, and evaluate the adjustment effect. If the matching degree reaches the preset threshold, the system will stop adjusting and consider the fine-tuning process to be complete. If the matching degree still does not meet the requirements, the system will continue to fine-tune until the projection contour is aligned with the standard position, that is, the distance between the projection center and the center of the rectangular shadow receiving resin panel is less than the set threshold.

[0058] As a further improvement of the control method of the present invention, the detection steps of the Canny edge detection method in the edge detection step are as follows:

[0059] S501: Gaussian blur: Gaussian blur is performed on the binary image again to smooth out noise and reduce false positives in edge detection;

[0060] S502: Calculate gradient: Calculate the gradient of the image in the X-axis and Y-axis directions respectively by using the Sobel operator, and combine the gradient information in these two directions to obtain the edge gradient strength of the image;

[0061] S503: Non-maximum suppression: non-maximum suppression is performed on edge gradients, and only edge points with the largest gradient intensity are retained, thereby ensuring that the detected edges are more accurate;

[0062] S504: Dual threshold detection: Set two thresholds, high and low, to further refine the edge and eliminate noise interference. Pixels with gradient strength higher than the high threshold are marked as strong edges, and pixels with gradient strength lower than the low threshold are suppressed.

[0063] As a further improvement of the control method of the present invention, the projection center calculation method is as follows:

[0064] S601: Assume that the edge point set of the shape is, P = {(x1, y1), ..., (x n ,y n )}, first determine an initial circle center (x0, y0) and radius r0;

[0065] S602: Calculate the distance from each edge point to the center of the circle

[0066] S603: Adjust the center position and radius so that r = max(d i ) is minimized, and the final determined center position of the circle is the center position of the shape.

[0067] As a further improvement of the control method of the present invention, the method for determining the initial center position and radius size in the projection center is as follows: x Coordinates and yThe maximum and minimum values ​​of the coordinates, with the center of the rectangle determined by these values ​​as the initial center, and the distance from the center to the farthest edge point as the initial radius;

[0068] Let the minimum x coordinate of the edge point be x min , the maximum x coordinate is x max , the minimum y coordinate is y min , the maximum y coordinate is y max , the initial center coordinates (x0, y0) are estimated to be The initial radius r0 is obtained by calculating the distance from (x0, y0) to the farthest point in the edge point set, that is, Where (x i ,y i ) is a point where edge points are concentrated;

[0069] Adjust the center and radius according to the distance. If the distance of an edge point is greater than the current radius, it means that the point is outside the circle. The center position and radius size need to be adjusted so that the circle can contain this point. The gradient descent optimization algorithm is used to adjust the center position.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] By combining PLC control and image recognition technology, high-precision tracking of the sun's position can be achieved, significantly improving the efficiency of the solar collector system. The system can automatically perform coarse and fine adjustments without manual intervention, reducing manual operation errors and improving the stability of the system. The system is not only suitable for solar tracking control under normal weather conditions, but also can continue to accurately track the sun's position when there are many clouds and unstable lighting conditions. The hardware equipment required for this system is low in cost and has a long service life, reducing the operation and maintenance costs of the overall system. The switching mechanism ensures seamless connection between coarse and fine adjustments, and intelligently selects the appropriate control method by automatically judging the deviation between the collector trough and the target angle. The seamless switching mechanism between coarse and fine adjustments ensures that the system can efficiently track the sun under any circumstances, reducing operation errors and improving system stability. The system can automatically switch the control mode according to the actual tracking error to adapt to different working environments, especially in unstable weather conditions. The automatic switching between coarse and fine adjustments greatly simplifies the system control logic, reduces complexity, and reduces dependence on hardware. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0073] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0074] like Figure 1 As shown, the present invention provides a distributed heating collector trough hierarchical control method and system based on image recognition and PLC, which specifically includes the following steps:

[0075] (I) System composition

[0076] The tracking control system of the present invention is mainly composed of the following parts:

[0077] Solar trough: used to focus sunlight and heat the fluid in the collector tube.

[0078] Collector tube: used to transmit and store heated fluid.

[0079] Absolute encoder: installed on the rotating shaft of the solar collector, used to collect the position information of the solar collector.

[0080] PLC: It is used to receive the solar collector tank position information collected by the absolute encoder and the solar collector tank target tracking angle calculated by the SPA algorithm, and then output the control signal to control the driving mechanism to rotate the solar collector tank.

[0081] Driving mechanism: used to rotate the heat collecting tank according to the control signal output by the PLC.

[0082] Camera: used to collect images on the receiving panel.

[0083] Image processing module: used to perform preprocessing, contour extraction, contour analysis and other operations on the images collected by the camera, and then calculate the position and angle of the plastic rod. Finally, according to the position and angle of the plastic rod, the rotation angle and direction of the heat collecting tank are calculated, and then the control signal is output through the PLC to control the driving mechanism to rotate the heat collecting tank.

[0084] (II) Working Principle

[0085] The working principle of the tracking control system of the present invention is as follows:

[0086] Coarse adjustment:

[0087] (1) The absolute encoder collects the position information of the heat collector and sends the position information to the PLC.

[0088] (2) The PLC receives the solar collector position information collected by the absolute encoder and calculates the target tracking angle of the solar collector according to the SPA algorithm.

[0089] (3) The PLC calculates the rotation angle and direction of the solar collector based on the solar collector position information and the target tracking angle.

[0090] (4) The PLC outputs a control signal to control the driving mechanism to rotate the heat collecting tank so that the angle of the heat collecting tank gradually approaches the target tracking angle.

[0091] Fine Tuning:

[0092] (1) The camera captures the image on the receiving panel and sends the image to the image processing module.

[0093] (2) The image processing module preprocesses the collected images, including denoising, enhancement, binarization and other operations.

[0094] (3) The image processing module calculates the projection shape center of the preprocessed image.

[0095] (4) The image processing module calculates the target tracking angle of the solar collector.

[0096] (5) Calculate the rotation angle and direction of the solar collector, and then output a control signal through the PLC to control the drive mechanism to rotate the solar collector so that the sun is always focused on the solar collector tube.

[0097] (III) Implementation steps

[0098] The implementation steps of the trough collector solar position tracking control method based on image recognition and PLC of the present invention are as follows:

[0099] Install the system: install the solar collector tank, solar collector tube, absolute encoder, PLC, drive mechanism, camera, image processing module and other components in the appropriate position, and connect and debug them.

[0100] Initialize the system: Set system parameters, such as geographic location, time, initial angle of the solar collector, etc.

[0101] Start System: Start the system and begin sun tracking.

[0102] Coarse adjustment: Use an absolute encoder to collect the position information of the solar collector tank, calculate the target tracking angle of the solar collector tank based on the SPA algorithm, and use PLC to control the rotation of the solar collector tank through the drive mechanism for coarse adjustment.

[0103] Switching link: When the switching conditions are met, the system switches from coarse adjustment to fine adjustment.

[0104] Fine-tuning: Use machine vision image processing technology to identify the projection outline of the plastic pole on the receiving panel, and control the rotation angle of the solar collector to make fine adjustments so that the sun is always focused on the solar collector tube.

[0105] Monitoring system: Real-time monitoring of the system's operating status, such as the angle of the collector trough, the position of the sun, the temperature of the collector tube, etc.

[0106] Adjust the system: According to the monitored system operating status, adjust the system parameters, such as the angle of the solar collector, the position and angle of the camera, etc., to improve the performance of the system.

[0107] Stop system: When sun tracking is not needed, stop the system.

[0108] (IV) Precautions

[0109] When implementing the trough collector solar position tracking control method based on image recognition and PLC of the present invention, the following points need to be noted:

[0110] When installing the system, make sure that each component is installed in the correct position and angle to ensure the performance of the system.

[0111] When initializing the system, it is necessary to set the correct parameters, such as geographic location, time, initial angle of the solar collector, etc.

[0112] After starting the system, it is necessary to monitor the operating status of the system in real time, such as the angle of the collector trough, the position of the sun, the temperature of the collector tube, etc., so as to adjust the system parameters in time.

[0113] When making a rough adjustment, the drive mechanism should be controlled to rotate the heat collector according to the target tracking angle of the heat collector calculated by the SPA algorithm, so that the angle of the heat collector gradually approaches the target tracking angle. When making a fine adjustment, the machine vision image processing technology should be used to identify the projection outline of the plastic rod on the receiving panel, and the rotation angle of the heat collector should be controlled for fine adjustment so that the sun is always focused on the heat collector tube.

[0114] During the switching process, it is necessary to ensure the stability and accuracy of the system to avoid errors and failures.

[0115] When adjusting system parameters, it is necessary to reasonably adjust system parameters based on the monitored system operating status, such as the angle of the solar collector, the position and angle of the camera, etc., to improve system performance.

[0116] When stopping the system, shut down the drive mechanism first and then other components to ensure the safety of the system.

[0117] Furthermore, the tracking control system of the present invention includes: a heat collecting tank, a PLC controller, an absolute encoder, a black carbon rod, a shadow receiving resin panel, an image acquisition device, an image processing module, and a driving mechanism. The heat collecting tank is driven to rotate by a driving mechanism installed at its bearing. The PLC, as a control center, calculates the position of the sun and the target tracking angle of the heat collecting tank based on the solar position algorithm (SPA), while receiving data from the absolute encoder and adjusting the angle of the heat collecting tank through the driving mechanism. The black carbon rod, the shadow receiving resin panel, and the image acquisition device are fixed directly above the heat collecting tank by a bracket. The image acquisition device is used to capture the projection of the carbon rod on the shadow receiving resin panel, and identify the projection position through an image processing algorithm. After calculating the target tracking angle of the heat collecting tank, it is transmitted to the PLC controller for fine-tuning the heat collecting tank.

[0118] The shadow receiving resin panel must be arranged parallel to the opening plane of the heat collecting tank, and the midline of the shadow receiving resin panel is located in the plane determined by the normal vector of the heat collecting tank and the heat collecting tube. The shadow receiving resin panel has a rectangular shape and a light-colored outer surface. The carbon rod must be arranged parallel to the heat collecting tube of the heat collecting tank and will not be blocked by any nearby objects.

[0119] The position of the solar collector tank is collected by the absolute encoder, the target tracking angle of the solar collector tank is calculated based on the SPA algorithm, and the PLC is used to control the rotation of the solar collector tank through the drive mechanism for coarse adjustment. At the same time, the machine vision image processing technology is used to identify the projection outline of the plastic rod on the receiving panel, and the rotation angle of the solar collector tank is controlled for fine adjustment.

[0120] A dual control strategy is adopted: first, the real-time position data of the solar collector is collected through an absolute encoder, and the precise position and tracking angle of the sun are calculated using the solar position algorithm (SPA); second, the PLC is used to perform coarse adjustment control on the solar collector to ensure that the solar collector is roughly aligned with the sun. Subsequently, machine vision technology is used to further adjust the rotation angle of the solar collector through image recognition of the projection contour of the plastic rod on the receiving panel, and fine adjustment is performed to ensure that the solar beam is always accurately focused on the solar collector tube. By combining coarse adjustment with fine adjustment, high-precision tracking of the sun's position is achieved to ensure that the sunlight is always focused on the solar collector tube, thereby improving the heat collection efficiency.

[0121] The main steps are as follows:

[0122] S1: When the system starts working, it first performs coarse adjustment to obtain the current real-time position of the solar collector through the absolute encoder, and calculates the target tracking angle according to the SPA algorithm. The PLC drives the solar collector to rotate roughly to the target angle.

[0123] S2: When the actual position of the solar collector is close to the target tracking angle, the PLC determines whether the fine adjustment conditions are met. If the coarse adjustment error is within the set threshold range (such as within 1 degree), the system automatically switches to the fine adjustment stage.

[0124] S3: The system starts the camera to capture the projection image of the plastic rod on the receiving panel. The image processing system calculates the deviation of the projection and controls the heat collecting tank for fine adjustment.

[0125] S4: During the fine-tuning process, the system continuously collects and processes images, and fine-tunes the angle of the solar collector in real time to ensure that the sunlight is focused on the collector tube.

[0126] S5: If the tracking error exceeds the preset range due to external factors (such as cloud cover, wind speed, etc.), the system will automatically stop fine adjustment and switch back to coarse adjustment mode, perform large angle adjustment before entering fine adjustment.

[0127] The main task of coarse adjustment is to control the solar collector to roughly align with the sun through PLC. The system uses an absolute encoder to collect the current position information of the solar collector, and calculates the azimuth and elevation of the sun through the SPA algorithm to obtain the target tracking angle. The PLC controls the drive mechanism to rotate the solar collector to near the target tracking angle.

[0128] Specific steps:

[0129] S101: Use an absolute encoder to collect the current position information and current tracking angle of the heat collecting tank, and feed it back to the PLC in real time;

[0130] S102: Calculate the azimuth and elevation of the sun on the day through the SPA algorithm to obtain the target tracking angle of the solar trough;

[0131] S103: According to the difference between the target tracking angle and the current tracking angle, the PLC drives the motor to control the rotation of the heat collecting tank to complete the rough adjustment;

[0132] S104: When the difference of the tracking angle of the solar collector is reduced to a critical switching value, the system enters a switching judgment phase.

[0133] The specific steps of step S102: SPA algorithm are as follows:

[0134] S201: Establishing a solar position model: establishing a relationship model between the solar position and time and geographical location based on astronomical knowledge;

[0135] S202: Collecting current time and geographical location information: Collecting current time and geographical location information through a clock chip and a GPS module;

[0136] S203: Calculate the sun position: Calculate the sun's position in the sky based on the sun position model and the current time and geographic location information;

[0137] S204: Calculate the target tracking angle of the solar trough: Calculate the target tracking angle of the solar trough according to the sun's position and the geometric parameters of the solar trough.

[0138] The SPA algorithm can calculate the sun's position. The main calculation formula is as follows. The Julian day calculation is as follows:

[0139] JD=INT(365.25(Y+4716))+INT(30.6001(M+1))+D+B-1524.5

[0140] In the formula, INT means rounding the parameter; Y means year; M means month; D means day; when the Julian calendar is used, B is 0;

[0141] When using the Gregorian calendar, the value of B is as follows:

[0142] (2-INT(Y / 100)+INT(Y / 400))

[0143] The sun altitude angle is calculated as follows:

[0144] α=e0+Δe

[0145]

[0146] In the formula, represents latitude, δ' represents declination, H' represents hour angle, P represents annual mean air pressure, and T represents annual mean temperature.

[0147] The solar azimuth angle is calculated as follows:

[0148]

[0149] The switching mechanism between coarse adjustment and fine adjustment is as follows: After the coarse adjustment is completed, the system determines whether it needs to switch to the fine adjustment mode based on the deviation between the collector and the target angle. When the deviation between the current tracking angle of the collector and the target tracking angle is within a certain range (for example, within 1 degree), the system automatically switches to the fine adjustment mode; after the switch is completed, the coarse adjustment mechanism stops temporarily and the fine adjustment phase starts.

[0150] The key to the switching mechanism is to ensure that the position of the collector tank after coarse adjustment is close enough to the target position, but still retain a certain adjustment space for the fine adjustment system to perform high-precision angle correction. Through this switching mechanism, excessive coarse adjustment errors can be avoided and the overall response speed of the system can be improved.

[0151] During the fine-tuning phase, machine vision technology is used to perform more precise angle adjustments. The goal of fine-tuning is to ensure that the solar beam is always accurately focused on the collector tube. The specific steps are as follows:

[0152] S301: The camera collects the projection image of the plastic rod on the receiving panel in real time;

[0153] S302: The image processing module performs image preprocessing, edge monitoring and projection center calculation;

[0154] S303: Calculating the required angle adjustment value according to the distance between the projection center and the center of the rectangular shadow receiving resin panel;

[0155] S304: When the required angle adjustment value is greater than the dead zone angle, the PLC controls the driving device to rotate the heat collecting tank according to the calculated fine-tuning angle, and further adjusts the angle of the heat collecting tank until the projection is accurately aligned with the heat collecting tube.

[0156] During the fine-tuning process, the system analyzes the changes in projection in real time and continuously makes fine adjustments to ensure that the solar trough and solar beam always maintain the best tracking accuracy.

[0157] Fine-tuning collects the projected image on the receiving panel through the camera, uses the image processing algorithm to analyze the projection contour position of the plastic rod, and finally adjusts the angle of the heat collecting trough to ensure that the solar beam is focused on the heat collecting tube. The image processing algorithm includes the following main steps:

[0158] 1. Image preprocessing

[0159] Image preprocessing is the first step in image recognition, and its purpose is to improve the quality of the image and make the subsequent processing steps more accurate. This step includes image denoising, grayscale conversion, binarization, and other operations.

[0160] Image denoising: Since the camera may be affected by light changes, noise, etc. during the shooting process, the original image collected may contain noise. In order to improve the accuracy of subsequent recognition, use Gaussian filtering or median filtering to denoise the image. Gaussian filtering can smooth the image and reduce random noise, while median filtering is more effective in removing salt and pepper noise.

[0161] Grayscale: The original image is usually a color image, containing RGB channels. In order to simplify the processing steps, the color image is converted into a grayscale image. Grayscale makes image processing more efficient by taking a weighted average of each pixel in the image and merging the RGB values ​​into a single grayscale value.

[0162] Binarization: By setting a threshold, the grayscale image is converted into a binary image, and each pixel in the image is either black or white (0 or 1). The purpose of binarization is to separate the projection outline from the background, making the outline clearer and facilitating subsequent edge detection. The adaptive threshold method can dynamically set the threshold according to the change in local image brightness, and performs better under different lighting conditions.

[0163] 2. Edge Detection

[0164] The Canny edge detection method is used to extract significant contour features in the projection image and identify the projection contour of the plastic rod. It has high edge detection accuracy and can effectively reduce false edges. Through the Canny algorithm, the projection contour of the plastic rod can be clearly detected, providing a basis for subsequent projection center calculation.

[0165] The main steps of the test are as follows:

[0166] S501: Gaussian blur: Gaussian blur is performed on the binary image again to smooth out noise and reduce false positives in edge detection.

[0167] S502: Calculate gradient: Calculate the gradient of the image in the X-axis and Y-axis directions respectively by using the Sobel operator, and combine the gradient information in these two directions to obtain the edge gradient strength of the image.

[0168] S503: Non-maximum suppression: Perform non-maximum suppression on edge gradients, retaining only edge points with the largest gradient intensity, thereby ensuring that the detected edges are more accurate.

[0169] S504: Dual threshold detection: Set two thresholds, high and low, to further refine the edge and eliminate noise interference. Pixels with gradient strength higher than the high threshold are marked as strong edges, and pixels with gradient strength lower than the low threshold are suppressed.

[0170] After edge detection, the coordinates of the edge pixels are stored in a data structure, such as an array or a list, and these points will be used for subsequent minimum circumscribed circle calculations.

[0171] 3. Projection center calculation

[0172] The minimum circumscribed circle method is used to determine the center of the projection. The minimum circumscribed circle that contains the shape is found, and its center is the center of the shape. The algorithm for determining the minimum circumscribed circle is based on the calculation of geometric distance. By continuously adjusting the radius and center position of the circle, the circle can contain all edge points and have the smallest radius.

[0173] S601: Assume that the edge point set of the shape is, P = {(x1, y1), ..., (x n ,y n )}, first determine an initial center (x0, y0) and radius r0.

[0174] S602: Calculate the distance from each edge point to the center of the circle

[0175] S603: Adjust the center position and radius so that r = max(d i ) is minimized, and the final determined center position of the circle is the center position of the shape.

[0176] Furthermore, the method for determining the initial center position and radius size is as follows: find the coordinates of all edge points and the maximum and minimum values ​​of the coordinates, use the center of the rectangle determined by these maximum values ​​as the initial center, and use the distance from the center to the farthest edge point as the initial radius.

[0177] Let the minimum x coordinate of the edge point be x min , the maximum x coordinate is x max , the minimum y coordinate is y min , the maximum y coordinate is y max The initial center coordinates (x0, y0) can be estimated as The initial radius r0 can be obtained by calculating the distance from (x0, y0) to the farthest point in the edge point set, that is, Where (x i ,y i ) is a point where edge points are concentrated.

[0178] Adjust the center and radius according to the distance. If the distance of an edge point is greater than the current radius, it means that the point is outside the circle, and the center position and radius need to be adjusted so that the circle can contain this point. Use the gradient descent optimization algorithm to adjust the center position of the circle. For example, according to the distribution of edge points whose distance is greater than the radius, move the center of the circle toward the center of gravity of these points by a certain step length, and update the radius to the maximum distance.

[0179] Repeat the above steps until the stopping condition is met. The stopping condition can be that the change of the center position and radius is less than a preset threshold, or the maximum number of iterations is reached.

[0180] After iterative optimization, the final determined center position of the circle is the center position of the shape.

[0181] 4. Adjustment of the tracking angle of the heat collecting trough

[0182] According to the center position of the determined shape, the distance d between the projection center and the center of the rectangular shadow receiving resin panel is calculated. It is known that the distance from the carbon rod to the rectangular shadow receiving resin panel is h, then the angle formed by the normal vector of the receiving panel and the normal vector of the heat collecting tank is:

[0183] The system calculates the required adjustment angle, and the PLC controls the driving mechanism to rotate the solar collector to make a small angle rotation adjustment. The rotation direction should reduce the angle θ formed between the normal vector of the receiving panel and the normal vector of the solar collector.

[0184] 5. Real-time feedback and fine-tuning loop

[0185] During the adjustment process, the image processing system will continue to capture and analyze the projected image to form a closed-loop control. After each angle adjustment, the system will re-capture the projected image and perform the same image preprocessing, edge detection and shape matching steps to evaluate the adjustment effect. If the matching degree reaches the preset threshold, the system stops adjusting and considers the fine-tuning process complete. If the matching degree still does not meet the requirements, the system continues to fine-tune until the projection contour is aligned with the standard position, that is, the distance between the projection center and the center of the rectangular shadow receiving resin panel is less than the set threshold.

[0186] The above description is only a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent change made based on the technical essence of the present invention still falls within the scope of protection required by the present invention.

Claims

1. A distributed heating collector trough hierarchical control system based on image recognition and PLC, characterized by: include: A solar collector tank, a PLC controller, an absolute encoder, a black carbon rod, a shadow receiving resin panel, an image acquisition device, an image processing module, and a driving mechanism. The solar collector tank is driven to rotate by a driving mechanism installed at its bearing. The PLC serves as a control center. It calculates the sun's position and the target tracking angle of the solar collector tank based on the solar position algorithm SPA. It also receives data from the absolute encoder and adjusts the angle of the solar collector tank through the driving mechanism. The black carbon rod, the shadow receiving resin panel, and the image acquisition device are fixed directly above the solar collector tank through a bracket. The image acquisition device is used to capture the projection of the carbon rod on the shadow receiving resin panel and identify the projection position through the image processing module. Image denoising, graying, and binarization are used for image preprocessing. The Canny algorithm is used for edge detection. The minimum circumscribed circle method is used to determine the center of the carbon rod projection. The tracking angle and rotation direction of the solar collector tank are calculated and transmitted to the PLC controller to control the rotation of the solar collector tank. The image processing module uses a Raspberry Pi development board, and the camera is inserted into the CSI interface on the Raspberry Pi mainboard through a flexible cable; The shadow receiving resin panel must be arranged parallel to the opening plane of the heat collecting tank, and the midline of the shadow receiving resin panel is located in the plane determined by the normal vector of the heat collecting tank and the heat collecting tube. The shadow receiving resin panel has a rectangular shape and a light-colored outer surface. The carbon rod must be arranged parallel to the heat collecting tube of the heat collecting tank and will not be blocked by any nearby objects. The driving mechanism is a DC motor or a hydraulic mechanism. The DC motor has at least two speed adjustment gears, high speed and low speed. For coarse adjustment, the driving motor rotates at high speed. For fine adjustment, the driving motor rotates at low speed to improve control accuracy. The image processing module communicates with the PLC via a network connection, and the communication protocol is Modbus TCP; The length and diameter of the carbon rod can be adjusted according to actual conditions; The installation position and angle of the camera of the image acquisition device can be adjusted according to actual conditions to ensure that the image on the receiving panel can be clearly acquired.

2. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 1 is characterized by: The specific steps are as follows: S1: When the system starts working, it first performs coarse adjustment, obtains the current real-time position of the solar collector through the absolute encoder, and calculates the target tracking angle according to the SPA algorithm. The PLC drives the solar collector to rotate roughly to the target angle. S2: When the actual position of the solar collector is close to the target tracking angle, the PLC determines whether the fine adjustment conditions are met. If the coarse adjustment error is within the set threshold range, the system automatically switches to the fine adjustment stage; S3: The system starts the camera to capture the projection image of the plastic rod on the receiving panel. The image processing system calculates the deviation of the projection and controls the heat collecting tank for fine adjustment. S4: During the fine-tuning process, the system continuously collects and processes images, and fine-tunes the angle of the heat collecting trough in real time to ensure that the sunlight is focused on the heat collecting tube; S5: If the tracking error exceeds the preset range due to external factors, such as cloud cover and wind speed, the system will automatically stop fine adjustment and switch back to coarse adjustment mode, perform large angle adjustment before entering fine adjustment.

3. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 2 is characterized by: In step S1, the PLC is used to roughly control the heat collecting tank to roughly align with the sun. The system uses an absolute encoder to collect the current position information of the heat collecting tank, and calculates the azimuth and elevation of the sun through the SPA algorithm to obtain the target tracking angle. The PLC controls the driving mechanism to rotate the heat collecting tank to near the target tracking angle. Specific steps: S101: Use an absolute encoder to collect the current position information and current tracking angle of the heat collecting tank, and feed it back to the PLC in real time; S102: Calculate the azimuth and elevation of the sun on the day through the SPA algorithm to obtain the target tracking angle of the solar trough; S103: According to the difference between the target tracking angle and the current tracking angle, the PLC drives the motor to control the rotation of the heat collecting tank to complete the rough adjustment; S104: When the difference of the tracking angle of the solar collector is reduced to a critical switching value, the system enters a switching judgment phase.

4. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 3 is characterized by: The specific steps of step S102: SPA algorithm are as follows: S201: Establishing a solar position model: establishing a relationship model between the solar position and time and geographical location based on astronomical knowledge; S202: Collecting current time and geographical location information: Collecting current time and geographical location information through a clock chip and a GPS module; S203: Calculate the sun position: Calculate the sun's position in the sky based on the sun position model and the current time and geographic location information; S204: Calculate the target tracking angle of the solar trough: Calculate the target tracking angle of the solar trough according to the sun's position and the geometric parameters of the solar trough.

5. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 1 is characterized by: The goal of the step S4, fine adjustment, is to ensure that the solar beam is always accurately focused on the heat collecting tube. The specific steps are as follows: S401: The camera collects the projection image of the plastic rod on the receiving panel in real time; S402: The image processing module performs image preprocessing, edge monitoring and projection center calculation; S403: Calculating the required angle adjustment value according to the distance between the projection center and the center of the rectangular shadow receiving resin panel; S404: When the required angle adjustment value is greater than the dead zone angle, the PLC controls the driving device to rotate the heat collecting tank according to the calculated fine-tuning angle, and further adjusts the angle of the heat collecting tank until the projection is accurately aligned with the heat collecting tube.

6. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 1 is characterized by: The image processing steps in step S402 are as follows: (1) Image preprocessing; Image denoising: Use Gaussian filtering or median filtering to denoise the image. Gaussian filtering can smooth the image and reduce random noise, while median filtering is better for removing salt and pepper noise. Grayscale: The original image is usually a color image. Converting the color image into a grayscale image makes image processing more efficient. Binarization: By setting a threshold, the grayscale image is converted into a binary image. The adaptive threshold method can dynamically set the threshold according to the change of local image brightness, and perform better under different lighting conditions; (2) Edge detection; The Canny edge detection method is used to extract significant contour features in the projection image and identify the projection contour of the plastic rod. Through the Canny algorithm, the projection contour of the plastic rod can be clearly detected, providing a basis for the subsequent projection center calculation; After edge detection, the coordinates of the edge pixels are stored in a data structure, which is an array or list. These points will be used for the subsequent minimum circumscribed circle calculation. (3) Projection center calculation: The minimum circumscribed circle method is used to determine the center of the projection. The minimum circumscribed circle that contains the shape is found. The center of the circle is the center of the shape. The algorithm for determining the minimum circumscribed circle is based on the calculation of geometric distance. By continuously adjusting the radius and center position of the circle, the circle can contain all edge points and have the smallest radius. (4) Adjustment of the tracking angle of the heat collecting trough: According to the center position of the determined shape, the distance between the projection center and the center of the rectangular shadow receiving resin panel is calculated. The angle formed between the normal vector of the receiving panel and the normal vector of the heat collecting tank is calculated as: The system calculates the required adjustment angle, and the PLC controls the driving mechanism to rotate the heat collecting tank to make a small angle rotation adjustment. The rotation direction should reduce the angle θ formed between the normal vector of the receiving panel and the normal vector of the heat collecting tank. (5) Real-time feedback and fine-tuning loop: During the adjustment process, the image processing system will continuously capture and analyze the projected image to form a closed-loop control. After each angle adjustment, the system will re-capture the projected image, perform the same image preprocessing, edge detection and shape matching steps, and evaluate the adjustment effect. If the matching degree reaches the preset threshold, the system will stop adjusting and consider the fine-tuning process to be complete. If the matching degree still does not meet the requirements, the system will continue to fine-tune until the projection contour is aligned with the standard position, that is, the distance between the projection center and the center of the rectangular shadow receiving resin panel is less than the set threshold.

7. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 6 is characterized by: The detection steps of the Canny edge detection method in the edge detection step are as follows: S501: Gaussian blur, performing Gaussian blur processing on the binary image again to smooth noise and reduce false positives in edge detection; S502: Calculate the gradient, calculate the gradient of the image in the X-axis and Y-axis directions respectively by using the Sobel operator, and combine the gradient information in these two directions to obtain the edge gradient strength of the image; S503: non-maximum suppression, performing non-maximum suppression on edge gradients, retaining only edge points with the largest gradient intensity, thereby ensuring that the detected edges are more accurate; S504: Dual threshold detection, setting high and low thresholds to further refine the edge and eliminate noise interference. Pixels with gradient intensity higher than the high threshold are marked as strong edges, and pixels with gradient intensity lower than the low threshold are suppressed.

8. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 6 is characterized by: The projection center calculation method is as follows: S601: Assume that the edge point set of the shape is, P = {(x1, y1), ..., (x n ,y n )}, first determine an initial circle center (x0, y0) and radius r0; S602: Calculate the distance from each edge point to the center of the circle S603: Adjust the center position and radius so that r = max(d i ) is minimized, and the final determined center position of the circle is the center position of the shape.

9. The distributed heating collector trough hierarchical control method based on image recognition and PLC according to claim 7 is characterized by: The method for determining the initial center position and radius size in the projection center is as follows: find the maximum and minimum values ​​of the x-coordinates and y-coordinates of all edge points, use the center of the rectangle determined by these maximum values ​​as the initial center, and use the distance from the center to the farthest edge point as the initial radius; Let the minimum x coordinate of the edge point be x min , the maximum x coordinate is x max , the minimum y coordinate is y min , the maximum y coordinate is y max , the initial center coordinates (x0, y0) are estimated to be The initial radius r0 is obtained by calculating the distance from (x0, y0) to the farthest point in the edge point set, that is, Where (x i ,y i ) is a point where edge points are concentrated; Adjust the center and radius according to the distance. If the distance of an edge point is greater than the current radius, it means that the point is outside the circle. The center position and radius size need to be adjusted so that the circle can contain this point. The gradient descent optimization algorithm is used to adjust the center position.

Images