An LED array acquisition method and plant cultivation system based on improved genetic algorithm optimization

Abstract

The present invention discloses an LED array acquisition method and plant cultivation system based on an improved genetic algorithm optimization. The method analyzes and acquires the number of LED beads based on the light intensity required by the plant, a multi-bead light irradiation model, and the area of ​​the target plane, with the goal of achieving high uniformity of the light intensity required by the target plane. Each LED bead is constrained within a coordinate range, with a random initial coordinate value of each LED bead within its constrained coordinate range as input and a fitness function as a judgment condition for evolutionary selection. The ratio of the standard error of the target plane light field irradiance to the average irradiance is used as an evaluation function to evaluate the uniformity of the target plane light field. A smaller value indicates a more uniform light field. When a set fitness value is reached or a set number of consecutive iterations are repeated and the fitness value is stable, the optimized coordinates of each LED bead are output. The present invention can solve the problem of uneven light distribution in the crop canopy and improve plant production efficiency in plant factories.

Description

Technical Field

[0001] The present invention belongs to the technical field of facility agriculture and relates to light regulation during plant growth, in particular to a high-light-efficiency plant growth lamp optimized based on an improved genetic algorithm and its application. Background Art

[0002] Light is one of the most fundamental environmental factors for plant growth and plays an important role in plant growth, development, and physiological metabolism. It not only provides energy for photosynthesis but also signals for various physiological responses. It has been established that an optimal light environment can achieve high-quality plant production.

[0003] Facility-based horticulture environments, such as plant factories, offer significant potential for stable and efficient production of horticultural crops, as they are less susceptible to extreme weather conditions. Here, artificial light sources are used to replace natural sunlight to provide the lighting conditions required by plants. Light-emitting diodes (LEDs) are widely used due to their long lifespan, low power consumption, and adjustable light spectrum. It is well known that light quality (wavelength), brightness (intensity), duration (duration), and uniformity and direction of light are all factors that determine light intensity.

[0004] When light is insufficient, plants cannot fully photosynthesize, but respiration still occurs, resulting in a net reduction in plant biomass. However, as light intensity increases, both plant growth and photosynthesis are impaired. Literature has demonstrated that plants grown under low light are more susceptible to light suppression. For weak-light artificial light sources, the linear, concentrated distribution of light within the cultivation layer is compounded by the current design of traditional light panels and the secondary optical design of LED lamp beads. When the central light intensity is optimally adjusted, the surrounding light intensity decreases; conversely, when the surrounding light intensity is optimally adjusted, the central light intensity becomes excessively strong. Uneven light distribution across the target receiving surface can lead to individual plant developmental variations, resulting in uneven plant growth across the plant population and potentially failing to meet plant growth requirements. Uniform plant growth is crucial for standardized production and harvesting management in plant factories. Uneven light distribution compromises standardized cultivation and automated harvesting standards, making standardized lighting and uniform light irradiance essential.

[0005] From an experimental research perspective, to ensure consistency in experimental materials and reduce experimental errors, plants must be manually relocated randomly or systematically to mitigate the defects caused by uneven light intensity. This not only increases the complexity of experimental management, but also causes more or less damage to the plants during the relocation process, affecting the experimental results. Obviously, this is not a viable option for plant factories and controlled environments. At the same time, ensuring that all plants indoors receive uniform intensity and spectrum is important, which is a necessary condition for indoor experiments studying specific gene expression. In general, uniform light distribution in the planting layer has a significant impact on experimental research, production management, standardized cultivation, and yield. Therefore, the design of uniform artificial light sources is one of the key issues that need to be addressed in production and experiments.

[0006] At present, some achievements have been made in achieving uniform LED lighting, which can be roughly divided into two directions. The first is to improve the structure of LED lamp beads to achieve uniform light intensity. For example, parabolic reflectors and free-form lenses are used to design secondary optical devices to achieve uniform lighting with small emission angles. There is also the use of the quadratic surface method to design optical components to produce uniform lighting, but this method increases production costs. The second is to use intelligent algorithms to optimize the design of LED arrays to achieve uniform lighting. Existing literature includes but is not limited to the use of simulated annealing algorithms and particle swarm algorithms to design LED arrays with uniform lighting. The second method has problems such as large randomness in the position of LED lamp beads, certain differences between the results and optimization, and increased time costs as the number of lamp beads increases. However, in the design of LED uniform light arrays, there is little research on the artificial light sources required for plant growth, especially the growth of plants under weak light conditions in plant factories. Summary of the Invention

[0007] In order to overcome the shortcomings of the above-mentioned prior art, the purpose of the present invention is to provide a high-light-efficiency plant growth lamp optimized based on an improved genetic algorithm and its application, so as to effectively solve the problem of uneven light distribution in the crop canopy and improve the plant production efficiency and standardization in plant factories.

[0008] In order to achieve the above object, the technical solution adopted by the present invention is:

[0009] In a first aspect of the present invention, a method for obtaining an LED array based on an improved genetic algorithm is provided, wherein the coordinates of each LED lamp bead in the LED array are optimized by the following steps:

[0010] The first step is to analyze and determine the number of LED lamp beads based on the light intensity required by the plants, the multi-lamp light irradiation model, and the area of ​​the target plane, with the goal of achieving high uniformity of light intensity on the target plane; the target plane is the crop canopy plane.

[0011] The second step is to constrain the coordinate range of each LED lamp bead. A random initial coordinate value of each LED lamp bead within its constrained coordinate range is used as input, and the fitness function is used as the judgment condition. The population size, maximum number of iterations, crossover rate and mutation rate are set, and then evolutionary selection is performed; the individuals in the population are the coordinate values ​​of the LED lamp bead;

[0012] In the third step, the ratio of the standard error of the irradiance of the target plane light field to the average value of the irradiance is used as an evaluation function to evaluate the uniformity of the target plane light field. The smaller the value, the more uniform the light field. When the set fitness value is reached or the set number of continuous iterations is reached and the fitness value is stable, the optimized coordinates of each LED lamp bead are output.

[0013] A second aspect of the present invention provides a plant cultivation system comprising an LED array and an automatic control system, wherein the LED array is coordinate optimized using the LED array acquisition method based on improved genetic algorithm optimization described in the first aspect, and the automatic control system is used to achieve real-time detection and automatic regulation of the light intensity of the LED array after the LED lamp bead coordinates are optimized.

[0014] Compared with the prior art, the present invention has the following beneficial effects:

[0015] Firstly, the LED lamp array is optimized based on the improved genetic algorithm to improve the uniformity of light intensity in the crop canopy plane.

[0016] Second aspect: Compared with the traditional square LED array, more and stronger pea sprouts can be obtained under the optimized LED array.

[0017] The third aspect: The optimized LED array significantly improves the efficiency of light energy utilization and has certain economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is the experimental design process of the present invention.

[0019] Figure 2 This is the Lambert distribution of a single lamp bead of the present invention.

[0020] Figure 3 This is the LED array lighting model of the present invention.

[0021] Figure 4 The improved method of the present invention and the corresponding IGA flow chart

[0022] Figure 5 36 grid points of the measurement plane of the present invention (the light intensity in the red area is set to 1668 lux).

[0023] Figure 6The present invention relates to a growth and culture process of pea sprouts (Dark: germination in darkness for two days; Light: daily illumination treatment of 14 h / 10 h (day / night) for eight days).

[0024] Figure 7 The cultivation light environment of the present invention is provided by IGA and square LED array.

[0025] Figure 8 The optimized LED array arrangement (a) and square LED array arrangement (b) of the present invention are shown.

[0026] Figure 9 1. The irradiance diagram (a) and cross-sectional diagram (b) of the optimized array obtained by IGA of the present invention.

[0027] Figure 10 1. The irradiance diagram (a) and cross-sectional diagram (b) of the square array of the present invention.

[0028] Figure 11 3D distribution of irradiance of the IGA optimized array (a) and the square array (b) of the present invention.

[0029] Figure 12 1 is a comparison of growth indicators under the two LED arrays of the present invention, namely plant height (a), stem diameter (b), yield (c) and health index (d).

[0030] Figure 13 Comparison results of the total output (a), total illumination (b), light energy utilization efficiency (c), and energy efficiency (d) of the IGA of the present invention and the square LED array.

[0031] Figure 14 It is a structural schematic diagram of the plant cultivation system of the present invention. DETAILED DESCRIPTION

[0032] The embodiments of the present invention are described in detail below with reference to the accompanying drawings and examples.

[0033] In facility lighting, light distribution is one of the important factors affecting plant growth. However, traditional lighting has the problem of uneven light distribution, which increases the difficulty of cultivation management and also brings challenges to the standardized production of plants. Achieving an optimal lighting environment is of great significance for efficient production. In order to effectively solve the problem of uneven light distribution in the planting layer, the present invention adopts a form of comparison with the traditional LED array and proposes an LED array acquisition method based on improved genetic algorithm optimization. The canopy plane light intensity distribution of crops is simulated, and the difference between the light source and the traditional light source is analyzed by spectrometer measurement. Pea sprouts are used as the research object to study the effect of optimized LED uniformity on the growth and development of pea sprouts, so as to further study and explore the uniformity of the light source through cultivation applications. More specifically, the present invention has three purposes: (1) to design a uniform light LED array that can benefit crop growth; (2) to measure some growth indicators of crops based on the optimized LED array; (3) to compare the growth quality of pea sprouts under the optimized LED array and the traditional square LED array environment. The present invention provides a new idea for crop production and lighting optimization strategy.

[0034] In terms of technical solutions, the present invention uses an improved genetic algorithm (IGA) to optimize the artificial light source for sprout cultivation, obtains a uniform light LED array solution, and compares and verifies the light source parameters through simulation and actual light field measurement; designs a cultivation light control system based on the above LED array solution, uses an artificial climate chamber to build a cultivation environment to conduct a comparative experiment on sprout cultivation, verifies the actual cultivation effect and economic benefit analysis of the optimized LED array, and the overall experimental process is as follows: Figure 1 .

[0035] The main steps of optimizing the coordinates of each LED lamp bead in the LED array of the present invention include:

[0036] The first step is to solve the problem of uneven illumination on the traditional light source plane while taking into account the lighting efficiency of the target plane. Based on the light intensity required by the plant, the multi-lamp light irradiation model and the area of ​​the target plane, the present invention analyzes and obtains the number of LED lamp beads with the goal of high uniformity of the light intensity required by the target plane. Here, the target plane refers to the canopy plane of the crop.

[0037] This step analyzes the method and principle of obtaining the number of LED lamp beads as follows:

[0038] Since LED lamp beads can be approximately equivalent to point light sources, there are obvious differences in light intensity at different divergence angles. The light intensity at the center of the point light source is the largest and gradually decreases towards the surroundings. Figure 2As shown. In actual cultivation environments, traditional matrix light sources have a higher central light intensity and a lower peripheral light intensity due to the superposition of light from multiple lamp beads. Therefore, it is necessary to optimize the LED array according to the LED lighting principle. According to Lambert's theorem, the relationship between the luminous intensity of a single lamp bead and the luminous angle satisfies formula (1):

[0039] I=I0cos m θ (1)

[0040] Where θ is the viewing angle of the LED lamp bead, and I0 is the luminous intensity perpendicular to the light source surface. The number of m depends on the luminous half-angle width θ 1 / 2 , which is usually provided by the manufacturer and is defined as the viewing angle when the irradiance in the vertical direction is reduced to half of the value, expressed as formula (2):

[0041]

[0042] The target plane light intensity is related to the light source distance z, LED luminous parameters (I, θ) and the number of lamp beads, such as Figure 3 As shown. The light intensity generated by a single LED lamp bead P(X, Y, 0) on the LED light source plane at the point Q(x, y, z) on the target plane can be expressed as formula (3):

[0043]

[0044] The target plane light field is formed by the superposition of the discrete LED lamp beads of the light source. Assuming that each LED lamp bead is an ideal Lambertian light source, when there are N lamp beads on the LED array surface, based on the multi-lamp bead light irradiation model, the total light intensity of N LED lamp beads at point Q (x, y, z) on the target plane is obtained, as shown in formula (4):

[0045]

[0046] Where z is the vertical distance between the LED array plane and the target plane, (X n ,Y n ,z) is the coordinate of the nth LED bead in the LED array. The setting of the total light intensity must meet the light intensity requirements and have a certain margin.

[0047] In an embodiment of the present invention, pea sprouts are taken as an example for further research. Previous studies have shown that blue LED light can increase the content of phenolic compounds in pea sprouts. Under a long photoperiod (22h / 2h), using a red to blue light ratio of 2:1 can improve the quality of pea sprouts, increase the soluble sugar content and antioxidant capacity. For some plants that are photosensitive crops, slight changes in light intensity may lead to changes in plant growth. When pea sprouts are exposed to light intensity of about 1668lux, they contain higher levels of soluble sugars and soluble proteins. However, when the light is uneven, its growth effect is greatly reduced.

[0048] In the embodiment of the present invention, the vertical distance z is set to 50 cm based on the traditional cultivation conditions. In order to meet the light intensity requirements of pea sprouts and set a certain margin, the total light intensity is set to 2000 lux. According to the above formula, the number of LED lamp beads N is 12.

[0049] The second step is to constrain the coordinate range for each LED lamp bead. A random initial coordinate value of each LED lamp bead in its constrained coordinate range is used as input, and the fitness function is used as the judgment condition. The population size, maximum number of iterations, crossover rate and mutation rate are set, and then evolutionary selection is performed; the individuals in the population are the coordinate values ​​of the LED lamp beads.

[0050] In the genetic algorithm, the present invention sets the population size to 100, the maximum number of iterations to 500, the crossover rate to 0.8, the mutation rate to 0.1, and uses the roulette wheel principle for evolutionary selection, ultimately obtaining an LED array with a higher fitness. The basic algorithm flow chart is shown in Figure 4 As shown in (b). Furthermore, mutations are often random, requiring considerable time to obtain appropriate mutation results. Therefore, an improved genetic algorithm (IGA) is proposed by initializing the population and constraining the mutation range.

[0051] The method for constraining the coordinate range of each LED bead in this step is as follows:

[0052] For ease of calculation, in the present invention, the number N of LED lamp beads is taken as an even number. For example, when N calculated in the first step is an odd number, it can be increased by 1. In order to avoid the concentrated distribution of LED lamp beads affecting the uniformity of light intensity, and at the same time speed up the convergence of the algorithm and quickly obtain the coordinate position of the LED lamp beads, constraints are added during the population initialization and mutation process. That is, the width of the LED array plane is divided into 2 rows by N LED lamp beads along the vertical axis (y-axis), and the length of the LED array plane is divided into N / 2 columns by the LED lamp beads on each row along the horizontal axis (x-axis). Finally, a symmetrical constraint range is formed, as shown in FIG. Figure 4As shown in a, the blue in the figure represents LED lamp beads. Each LED lamp bead determines its position in its own red area, and then the accurate LED lamp bead position is obtained through the algorithm. Figure 8 In (a), each LED is mutated within this designed range, allowing for rapid optimization of the LED coordinates while achieving high illumination uniformity. The purpose of this range-delimited step is: 1. To obtain an LED array with high illumination uniformity; 2. Compared to the unmodified genetic algorithm, the improved algorithm can achieve this in a shorter time.

[0053] For pea sprouts, the first step has determined that the number of its LED lamp beads is 12, so this step divides its LED array into 2 rows and 6 columns, thereby constructing the constraint range of each LED lamp bead.

[0054] In the third step, the ratio of the standard error of the irradiance of the target plane light field to the average value of the irradiance is used as an evaluation function to evaluate the uniformity of the target plane light field. The smaller the value, the more uniform the light field. When the set fitness value is reached or the set number of continuous iterations is reached and the fitness value is stable, the optimized coordinates of each LED lamp bead are output.

[0055] In the optimization of this step, the evaluation function is expressed as formula (5):

[0056]

[0057] Where x i ,y i Indicates the coordinates of the i-th LED lamp bead, n indicates the number of LED lamp beads, which has a different meaning from N indicating the actual number of lamp beads. The value of n can be N. Represents the average value of the target plane light field irradiance, σ represents the standard error of the target plane light field irradiance, the target plane is gridded and divided into U×V grids. The larger the values ​​of U and V, the more grids the plane is divided into, and the more accurate the average light intensity. The average light intensity of all grids on the target plane is the average value of the target plane light field irradiance. The standard deviation of the illumination intensity of the target plane light field is the standard error σ of the irradiance of the target plane light field. and σ can be expressed as Equation (6) and Equation (7) respectively:

[0058]

[0059]

[0060] In the formula, E(x p ,y q ,z) represents the target plane coordinate (x p ,y q,z). z is the vertical distance between the LED array plane and the target plane. For pea sprouts, the vertical distance is set to 50 cm based on traditional cultivation conditions.

[0061] In order to optimize the LED array layout and obtain a lamp arrangement with high illumination uniformity, the evaluation function f must be minimized. The smaller f is, the smaller the difference in illumination intensity at different locations on the target illumination surface, i.e., the higher the uniformity. In the GA algorithm, the fitness function value is required to be maximized. Therefore, the fitness value of the present invention is defined as formula (8):

[0062] fit=1 / f (8)

[0063] At this time, the higher the fitness fit, the higher the uniformity of the illumination. In addition, after GA reaches a certain number of iterations, its fitness function is difficult to change. For this reason, the present invention also designs that if the fitness function is consistent for 100 consecutive generations, it will be directly output, again achieving the purpose of reducing time cost, such as Figure 4 As shown in Figure b, the method of combining coordinate area division can more quickly obtain the optimized LED coordinates and produce a highly uniform lighting distribution LED lamp array design.

[0064] The uniform light LED array obtained in this study was verified and compared with a conventional array light source through optical simulation and actual light field measurement. The optical simulation was performed using Trace Pro software. To accommodate actual cultivation environments, the simulated receiving surface was set to 50 cm × 50 cm and then scaled down to a target surface of 40 cm × 40 cm to calculate irradiance uniformity.

[0065] The actual light field is measured using a PS-300 spectrometer in an artificial climate chamber. The distance between the LED array and the target receiving surface is 50 cm, and the size of the receiving surface is 40 cm × 40 cm. During the experiment, the target receiving surface is divided into 6 × 6 measurement points. Figure 5 The light intensity at the points marked in the figure was set to 1668 lux to ensure that the light intensity at any point on the receiving surface was sufficient for the growth of pea sprouts. Finally, the PS-300 was used to measure and record the 36 grid points of the two light sources one by one.

[0066] In order to evaluate the illumination uniformity on the target plane, the illumination uniformity parameter η is used for analysis, and its calculation formula is as follows:

[0067]

[0068] Among them, E max is the maximum illumination intensity of all grid points on the target plane

[0069] The seeds of pea were sown in 40×40 cm seedling trays, with one seed placed in each hole of the tray (the planting density was about 1.36 kg / m 2 ), add water until the seeds are half submerged. Then place them in an artificial climate box (RGL-P500D-CO2, Hefei Youke scientific equipment co. LTD), with the cultivation plane located 50 cm below the LED array. The production method of the pea sprouts of the present invention is as follows Figure 6 Observe the growth of the pea sprouts daily and photograph their horizontal surface. Observe them daily and replenish water promptly and perform meticulous management.

[0070] In order to explore the effect of light intensity uniformity on plant growth, a light uniformity experiment with controlled environment was conducted in an incubator. In order to meet the weak light intensity requirements for the growth of pea sprouts, the light intensity of the LED array in the artificial climate chamber was always kept consistent with the light intensity during the uniform light measurement. Figure 7 shown.

[0071] After eight days of LED array illumination, the pea sprouts were harvested when they reached approximately 10 centimeters. The pea sprouts were cut 2 to 3 centimeters from the root and measured for plant height, stem diameter, and yield.

[0072] The seedling health index was determined by the following formula: health index = stem diameter / stem height × dry weight.

[0073] Energy efficiency is determined by the following formula: Energy efficiency = dry weight / power consumed by all LEDs.

[0074] The total illuminance (TIL) of light consumed and the illuminance (r) required to produce a unit mass of pea sprouts during the growth of pea sprouts are calculated as follows:

[0075]

[0076] where TIL is the total light intensity (lux) measured on the receiving surface, and T is the cumulative time of the pea sprout growth photoperiod (s).

[0077]

[0078] Where r is the illuminance required to produce unit mass of pea sprouts, in lux g-1, and TY is the total yield of pea sprouts (g).

[0079] In addition, the economic benefits can be analyzed by adding the yield of the sample and the remaining pea sprouts.

[0080] Furthermore, the present invention uses a digital multimeter (UT890C) to measure the voltage and current of the LED array circuit, calculates the electrical power and electricity cost of the two LED arrays, and compares and analyzes the energy consumption of the two LED arrays and the economic benefits of pea sprouts.

[0081] The present invention uses optical simulation software to import the coordinates of the optimized LED array and the traditional square LED array into the optical simulation software. Figure 8 As shown. Figure 9 (right) and Figure 10 (Right) It is found that the range of light intensity fluctuation at each point on the IGA target receiving surface in the horizontal and vertical directions is smaller than that of the square. Figure 9 (left) and Figure 10 (Left) shows a 50cm×50cm square light spot. It can be seen that the IGA has better uniformity of illumination intensity than the traditional square array. In addition, after reducing the plane to the target plane of 40cm×40cm, the uniformity of the IGA optimized array and the square array is calculated to be 91.72% and 85.74% respectively. In three-dimensional space, the irradiance distribution of the LED array optimized by IGA in the vertical direction of space has less fluctuation and is more flat and uniform, such as Figure 11 In general, the simulation results of the LED array designed by IGA in the present invention are better than those of the traditional square array and are in line with reality.

[0082] Analysis and calculations yield a comparison of the two light panels (Table 1). The range of the traditional square LED array is approximately three times that of the IGA-optimized LED array, and the standard deviation is approximately 2.89 times, further demonstrating the reduced fluctuation in light intensity achieved by the IGA-optimized LED array. Furthermore, the spectrometer-measured uniformity of the IGA-optimized light panel reached 92.40%, surpassing the 80.11% uniformity of the traditional square array.

[0083] Table 1: Comparison of actual measured light intensity of two LED arrays.

[0084]

[0085] The pea sprouts with higher light uniformity received the highest yield at harvest, indicating that low light uniformity in the planting layer during the growth period of pea sprouts will lead to lower yields at the harvest period. Figure 12 In addition, compared with the LED array with uneven light intensity, the pea sprouts illuminated by the LED array with higher light intensity uniformity have a higher health index, as shown in Figure c. Figure 12 As shown in d

[0086] In terms of energy efficiency, the total light intensity on the traditional square LED array is the highest ( Figure 13b), but the total yield of pea sprouts under the IGA optimized array increased by 8.58% ( Figure 13 This suggests that higher light intensity may have a negative impact on plant yield. In addition, the IGA-optimized LED array reduced the light intensity required to produce pea sprouts per unit mass by 26.46% ( Figure 13 Figure c) shows that higher light energy utilization efficiency leads to higher pea sprout yields. Notably, compared to traditional square LED arrays, this method achieves a slight improvement in energy efficiency. This method, which uses higher energy efficiency to produce more healthy pea sprouts, is undoubtedly a wise and feasible choice. Therefore, the LED optimization array designed in this invention, based on an improved genetic algorithm, has considerable economic benefits and practical application value.

[0087] In terms of economic benefits, the optimized LED array costs 0.04 yuan more in electricity than the traditional square LED array. Taking the crop price of 4 yuan per kilogram as an example, if calculated at the lowest price, the optimized LED array needs to increase yield by at least 5.00g to break even compared to the square LED array. In this experiment, compared to the traditional square LED array, the pea sprouts obtained by the IGA optimized LED array increased yield by an average of 21.57g ( Figure 13 (a) This method increases pea sprout production at a very low electricity cost. The LED array wastes very little energy, resulting in more, stronger pea sprouts at a relatively high cost.

[0088] Table 2:

[0089]

[0090] Therefore, in an experiment using pea sprouts as an example, the results showed that the illumination uniformity of the optimized LED array and the traditional square LED array was 91.72% and 85.74% respectively during the simulation phase. The actual illumination uniformity measured by the spectrometer was 92.40% and 80.11% respectively. Furthermore, a comparison of the effects of the two LED arrays on the growth of pea sprouts revealed that the total light intensity of the optimized LED array was reduced by 20.12%, while the total output of the improved LED array increased by 8.58%. The light intensity required to produce pea sprouts per unit mass was reduced by 26.46%, while also improving energy and economic efficiency.

[0091] Furthermore, the present invention also provides a plant cultivation system, such as Figure 14 As shown, it includes an LED array and an automatic control system. The LED array uses the LED array acquisition method based on improved genetic algorithm optimization of the present invention to optimize the coordinates. The automatic control system is used to realize real-time detection and automatic regulation of the light intensity of the LED array after the LED lamp bead coordinates are optimized.

[0092] The high-light-efficiency plant cultivation system first uses an improved genetic algorithm to obtain an LED array arrangement to achieve higher lighting uniformity, then sets the light intensity target value on the touch screen, and then uses a light radiation sensor to detect the current light intensity and send it to the microcontroller and display it on the touch screen. The main control system calculates the light intensity difference in real time and then adjusts the PWM duty cycle, ultimately controlling the light intensity to the required value.

[0093] The entire system of the present invention is applied to the cultivation of photosensitive crops grown under weak light. By optimizing the position layout of LED lamp beads, the uniformity of light at the planting level is achieved, so that all pea sprouts on the cultivation surface can further obtain a more appropriate light intensity to achieve the purpose of increasing production. In addition, the present invention is also applicable to some environmentally controllable cultivation environments such as plant factories. First, the automatic control system realizes a certain level of intelligence. Secondly, some controllable environments such as plant factories require a lot of electricity costs to maintain their stability. According to cultivation verification, compared with unoptimized light panels, the cultivation system of the present invention can improve light energy utilization efficiency and energy efficiency.

[0094] For example, the automatic control system includes a human-computer interaction module, a light detection module, an LED driver module and a single-chip microcomputer.

[0095] The human-computer interaction module displays the current light intensity in real time, sets the expected light intensity of the target plane and sends it to the microcontroller for control; for example, the human-computer interaction module uses a 2.4-inch Taojingchi HMI smart serial port screen.

[0096] The light detection module is used to detect the real-time light intensity of the target plane and transmit the detection value to the single-chip microcomputer. For example, the light detection module adopts the MAX44009 ambient light radiation sensor.

[0097] The LED driver module uses a wide range of PWM dimming to control the light intensity output by the LED array. For example, the LED driver module uses a PT4123 constant current chip.

[0098] The single-chip microcomputer, as the main control module of the entire system, calculates the light intensity difference in real time and adjusts the PWM duty cycle, ultimately controlling the light intensity to the required value. For example, the single-chip microcomputer uses STM32F103C8T6.

[0099] In a specific embodiment, for ease of application, the system of the present invention further includes a lamp housing module and a power supply module. The lamp housing module serves as the outer shell of the entire LED lamp panel; the power supply module is used to power the entire system. For example, the power supply module uses a 24V / 75W (RSP-75-24) switching power supply.

[0100] In summary, the inventors studied the effect of light uniformity on pea sprouts through a self-designed uniform light plant growth LED array. By comparing with the traditional square LED array, the effect of light uniformity on the growth and quality of pea sprouts was explored. The experimental results show that the LED array optimized by the improved genetic algorithm has higher light uniformity. Both the simulation results and the actual measurement results of the light source show that the uniformity of the optimized LED array is higher than that of the traditional square LED array. Under this optimized LED array, pea sprouts with higher yield, better quality and stronger growth can be obtained. Therefore, it is recommended that pea sprout growers should give certain consideration to light uniformity in order to further achieve the purpose of increasing production and strengthening seedlings. However, the increased cost is extremely low, while energy efficiency and light energy utilization are improved.

Claims

1. A method for obtaining an LED array based on improved genetic algorithm optimization, characterized in that: The coordinates of each LED lamp bead in the LED array are optimized by the following steps: The first step is to analyze and determine the number of LED lamp beads based on the light intensity required by the plants, the multi-lamp light irradiation model, and the area of ​​the target plane, with the goal of achieving high uniformity of the light intensity required on the target plane. The target plane is the crop canopy plane; The second step is to constrain the coordinate range of each LED lamp bead. A random initial coordinate value of each LED lamp bead within its constrained coordinate range is used as input, and the fitness function is used as the judgment condition. The population size, maximum number of iterations, crossover rate and mutation rate are set, and then evolutionary selection is performed; the individuals in the population are the coordinate values ​​of the LED lamp bead; In the third step, the ratio of the standard error of the irradiance of the target plane light field to the average value of the irradiance is used as an evaluation function to evaluate the uniformity of the target plane light field. The smaller the value, the more uniform the light field. When the set fitness value is reached or the set number of continuous iterations is reached and the fitness value is stable, the optimized coordinates of each LED lamp bead are output.

2. The LED array acquisition method based on improved genetic algorithm optimization according to claim 1, characterized in that: The first step is to analyze and obtain the number of LED lamp beads as follows: Based on the multi-lamp light irradiation model, the total light intensity of N LED lamp beads at point Q (x, y, z) on the target plane is obtained using the following formula: Where I0 is the luminous intensity perpendicular to the surface of the light source, and the number of m depends on the luminous half-angle width θ 1 / 2 , which is defined as the viewing angle at which the irradiance in the vertical direction is reduced to half of its value, as follows: z is the vertical distance between the LED array plane and the target plane, (X n ,Y n ,z) is the coordinate of the nth LED bead in the LED array. The setting of the total light intensity must meet the light intensity requirements and have a certain margin.

3. The LED array acquisition method based on improved genetic algorithm optimization according to claim 1, characterized in that: In the second step, the method for constraining the range of coordinates for each LED bead is as follows: Take N as an even number, divide the width of the LED array plane into 2 rows along the vertical axis with N LED beads, and then divide the length of the LED array plane into N / 2 columns along the horizontal axis with the LED beads in each row, finally forming a symmetrical constraint range.

4. The LED array acquisition method based on improved genetic algorithm optimization according to claim 2 or 3, characterized in that: The plant is pea sprouts. Based on traditional cultivation conditions, the vertical distance z is set to 50 cm, the total light intensity is set to 2000 lux, the number of LED lamp beads is set to 12, the population size is set to 100, the maximum number of iterations is 500, the crossover rate is 0.8, the mutation rate is 0.1, and evolutionary selection is performed using the roulette principle.

5. The LED array acquisition method based on improved genetic algorithm optimization according to claim 1, characterized in that: In the third step, the evaluation function is expressed as follows: Where x i ,y i Indicates the coordinates of the i-th LED lamp bead, n indicates the number of LED lamp beads, Represents the average value of the target plane light field irradiance, σ represents the standard error of the target plane light field irradiance; the target plane is gridded and divided into U×V grids. The larger the values ​​of U and V, the more grids the plane is divided into, and the more accurate the average light intensity. The average light intensity of all grids on the target plane is the average value of the target plane light field irradiance. The standard deviation of the illumination intensity of the target plane light field is the standard error σ of the irradiance of the target plane light field. The calculation formulas for and σ are as follows: In the formula, E(x p ,y q ,z) represents the target plane coordinate (x p ,y q ,z) light intensity.

6. The LED array acquisition method based on improved genetic algorithm optimization according to claim 5, characterized in that: In the improved genetic algorithm, the fitness function value is required to be maximized, and the fitness function is defined as follows: fit = 1 / f The higher the fitness value fit, the higher the lighting uniformity.

7. A plant cultivation system, characterized in that: The invention comprises an LED array and an automatic control system, wherein the LED array is coordinate optimized using the LED array acquisition method based on improved genetic algorithm optimization as described in any one of claims 1 to 6, and the automatic control system is used to realize real-time detection and automatic regulation of the light intensity of the LED array after the LED lamp bead coordinates are optimized.

8. The plant cultivation system according to claim 7, characterized in that: The plant cultivation system further comprises a lamp housing module and a power supply module. The lamp housing module serves as the outer shell of the entire LED lamp panel; the power supply module is used to supply power to the entire system.

9. The plant cultivation system according to claim 7, characterized in that: The automatic control system comprises: The human-computer interaction module displays the current light intensity in real time, sets the expected light intensity of the target plane and sends it to the microcontroller for control; Light detection module, used to detect the real-time light intensity of the target plane and transmit the detection value to the microcontroller; LED driver module, which uses wide-range PWM dimming to control the light intensity output by the LED array; The microcontroller, as the main control module of the entire system, calculates the light intensity difference in real time and adjusts the PWM duty cycle, ultimately controlling the light intensity to the required value.

Images