Rapid breeding method for grafted seedlings of durian

Abstract

The invention discloses a rapid breeding method for grafted seedlings of durian, and relates to the technical field of durian culture. The efficient breeding of the durian grafted seedlings is realized through the measures of seedbed building, disinfection, germination acceleration, laser separation planting, matrix nutrition bag component optimization, environment monitoring by a sensor, breeding parameter optimization by combining a random forest model and a genetic algorithm and the like. According to the method, the breeding period is shortened, the survival rate and the growth speed of the grafted seedlings are increased, a large number of high-quality grafted seedlings are provided for durian farmers, and the durian cultivation efficiency is remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of durian cultivation, and in particular to a method for rapid propagation of durian grafted seedlings. Background Art

[0002] Durian is a giant tropical evergreen tree of the Malvaceae family. It has oblong leaves with pointed tips, cymes, light yellow flowers, football-sized fruits, firm peels, dense triangular thorns, and flesh composed of aril flesh bags. The flesh is light yellow, sticky and juicy. It is a fruit with great economic value. According to the phenological characteristics of durian, it can be planted in areas with an average daily temperature of more than 22°C and no frost. It can be developed in some hot areas of Hainan Province, Yunnan Province and Guangxi Zhuang Autonomous Region in China. Because durian needs a high temperature climate all year round to grow and bear fruit, even in the equatorial region, durian cannot be planted or bear fruit in highlands above 600m due to the drop in temperature. The listing season is from April to January of the following year, and there is also production in other months.

[0003] For a long time, people have often used seeds to propagate durian. However, due to the large genetic variability of durian seeds, each seed may grow fruits of different qualities, and the good and the bad are mixed. Good varieties of durian seeds may grow inferior varieties of durian, while inferior durian may breed very excellent durian varieties. Therefore, if you choose to use seeds for durian propagation, its genetic variability is relatively large, and it may grow fruits of different qualities. In order to cultivate excellent varieties, grafting is the best propagation method.

[0004] The current method of propagating durian by grafting fruit trees requires a long cycle. To this end, we propose a method for rapid propagation of grafted durian seedlings to solve the above problems. Summary of the invention

[0005] The main purpose of the present application is to provide a method for rapid propagation of durian grafted seedlings, aiming to solve the technical problem of the long cycle required for the current method of propagating durian by grafting fruit trees, and to a certain extent can greatly shorten the breeding cycle of durian grafted seedlings.

[0006] To achieve the above object, the present application provides a method for rapid propagation of grafted durian seedlings, comprising the following steps:

[0007] Step S101, constructing a seedling bed, laying clean river sand with a thickness of 15-25 cm on the constructed seedling bed, disinfecting the seedling bed by spraying a first disinfectant, and cleaning the disinfected seedling bed;

[0008] Step S102, based on the collected clean, fresh and full durian seeds, soaking them in a second disinfectant for at least 30 minutes, cleaning the soaked durian seeds, and germinating the washed durian seeds, wherein the first disinfectant is implemented as 0.3% potassium permanganate disinfectant, and the second disinfectant is implemented as 800 times liquid of 50% carbendazim wettable powder;

[0009] Step S103, arranging a plurality of laser generators staggered and spaced 10 cm apart from each other on the upper surface of the seedling bed, wherein the lasers generated by the laser generators are used to separate the upper surface of the seedling bed into a plurality of cultivation areas, each of which is used to plant a durian seed after germination;

[0010] Step S104, selecting durian seeds with hypocotyls 1-3 cm long at the hilum as seedlings to be cultivated, burying each seedling in river sand in rows of 10×10 cm with the hypocotyl facing downward, spraying the second disinfectant, covering with greenhouse film, and setting a sunshade net with a shading degree of 40-70%;

[0011] Step S105, extending the detection end of the sensor into the river sand, the depth of the detection end of the sensor extending into the river sand is greater than the depth of the seedlings to be cultivated buried in the river sand, using the sensor to detect the environment in the seed shed, maintaining the temperature of the environment in the seed shed at 25-35°C and the humidity at 70-90%, and burying each seedling to be cultivated in the river sand with the radicle facing downward and 10×10 cm in rows, further comprising: after covering the surface of each seedling to be cultivated with river sand, evenly spreading a layer of coconut bran on the surface of the river sand , use a sprayer to spray clean water on the river sand covered with coconut bran once, select seedlings with a height of 10-20 cm in 25-28 days, pull them out and use them as young rootstocks for grafting, transplant the grafted seedlings into a preset matrix nutrient bag, water and fertilize them regularly, remove suckers, and prevent and control diseases and pests. When the buds of the scion grow to 7.5-20 cm, they are used as durian seedlings that have completed breeding. The matrix nutrient bag includes: perlite, vermiculite, coconut bran, humus, peat soil, garden soil and organic fertilizer.

[0012] Furthermore, in the method for rapid propagation of grafted durian seedlings of the present invention, the matrix nutrient bag comprises: 15% to 25% perlite, 5% to 10% vermiculite, 10% to 20% coconut bran, 10% to 20% humus, 10% to 20% peat soil, 25% to 35% garden soil and 15% to 25% organic fertilizer;

[0013] The pulling out and grafting as a juvenile rootstock comprises:

[0014] The pulled out seedlings are used as young rootstocks, the young rootstocks are cut off with a grafting knife at a position 5-10 cm away from the sowing mark, and the young rootstocks are split longitudinally along the length direction to form a V-shaped knife mark with a depth of 1.5-3 cm;

[0015] Healthy, high-quality, pest-free durian buds with full buds are collected as scions. Each scion is cut into small segments with 1-3 buds, each segment has 1-2 leaves and 2 / 3 of the leaves are cut off, and the lower end of the scion is cut into a wedge with a wedge length of 1-3 cm;

[0016] The scion is grafted onto the young stock by cleft grafting, the scion and the young stock are tightly bound with a grafting film, and the scion is bagged;

[0017] Before transplanting the grafted seedlings into the preset substrate nutrient bag, the method further comprises:

[0018] crushing the lumps in the matrix nutrient bag and uniformly mixing the components in the matrix nutrient bag;

[0019] Use a sprayer to water the substrate nutrient bag;

[0020] After watering the substrate nutrient bag, the substrate nutrient bag is placed for 5-10 minutes;

[0021] The method of crushing the lumps in the matrix nutrient bag and uniformly mixing the components in the matrix nutrient bag comprises:

[0022] Put the components in the matrix nutrient bag into a grinder, and use the grinder to grind the lumps in the matrix nutrient bag;

[0023] The crushed components in the matrix nutrient bag are put into a blender, and the blender is used to mix the components in the matrix nutrient bag.

[0024] Furthermore, the method for rapid propagation of grafted durian seedlings of the present invention comprises accelerating germination of durian seeds, comprising:

[0025] Set up a first moisturizing layer horizontally in a weak light place, place durian seeds on the first moisturizing layer at intervals, and then cover the durian seeds with a second moisturizing layer;

[0026] The second moisturizing layer is sprayed evenly and intermittently with water using a sprayer to collect the humidity between the first moisturizing layer and the second moisturizing layer, and the durian seeds are kept moist based on the humidity to fully germinate the durian seeds.

[0027] The cleaning of the sterilized seedling bed comprises:

[0028] The nozzle of the sprayer is arranged on the seedling bed after the first disinfectant is sprayed, and clean water is sprayed evenly above the seedling bed;

[0029] The clean water flowing out of the seedling bed is obtained, and the content of the first disinfectant in the clean water flowing out of the seedling bed is detected. When the content of the first disinfectant in the clean water flowing out of the seedling bed is reduced to 2%-3%, spraying the seedling bed is stopped.

[0030] Furthermore, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0031] Temperature sensors and humidity sensors are installed in the seed shed to monitor and maintain the optimal temperature and humidity range in the seed shed in real time. The optimal temperature range is 25-35°C and the optimal humidity range is 70-90%.

[0032] Install light sensors to obtain light data during germination and growth;

[0033] A soil moisture sensor and a nutrient sensor are set in the matrix nutrient bag. The soil moisture sensor is used to collect soil moisture data, and the nutrient sensor is used to collect soil fertilization data;

[0034] Industrial cameras are installed in the greenhouse and on the soil to collect image data during the growth process of durian;

[0035] The data collected by the temperature sensor, humidity sensor, light sensor, soil moisture sensor, nutrient sensor and industrial camera are transmitted to a preset database for storage, and the durian grafted seedling rapid breeding experimental data generated in real time by the user mobile terminal and the background control terminal are stored in the preset database.

[0036] Furthermore, the method for rapid propagation of grafted durian seedlings of the present invention obtains historical data of rapid propagation of grafted durian seedlings from a preset database, and the historical data of rapid propagation of grafted durian seedlings includes historical data of seedling bed disinfection, historical data of durian seed treatment, historical data of seedling growth environment, historical data of seedling growth, historical data of grafting, historical data of the number of components in substrate nutrient bags, historical data of the number of grafted seedlings grown, and historical data of experimental data;

[0037] The historical data of rapid propagation of durian grafted seedlings is divided into a training set and a validation set, and the random forest model is trained using the training set and the validation set to obtain a simulation model for rapid propagation of durian grafted seedlings;

[0038] Obtain the start time of durian grafting in the area monitored by the industrial camera, use the real-time data of durian in the area monitored by the industrial camera to substitute the real-time data of durian in the area monitored by the industrial camera and the start time of durian grafting in the area monitored by the industrial camera into the durian grafting rapid breeding simulation model, and output the expected growth data of durian;

[0039] The starting time of the durian grafting seedlings in the monitoring area of ​​the industrial camera is matched in the preset durian grafting seedlings breeding knowledge base to obtain the standard durian growth data and durian growth error range value corresponding to the starting time of the durian grafting seedlings in the monitoring area of ​​the industrial camera;

[0040] The standard flow growth data is compared with the expected growth data of durian to obtain the expected growth error data of real-time flow, and the expected growth error data of real-time flow is compared with the durian growth error range value. If the expected growth error data of real-time flow exceeds the durian growth error range value, the durian breeding parameters in step S101, step S102, step S103, step S104 and step S105 of the rapid breeding method for grafted seedlings of durian are optimized to obtain optimized durian breeding parameters, and the optimized durian breeding parameters are used to replace the various durian breeding parameters in step S101, step S102, step S103, step S104 and step S105.

[0041] Furthermore, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0042] The specific steps of optimizing the durian breeding parameters in step S101, step S102, step S103, step S104 and step S105 in the rapid breeding method of durian grafted seedlings using a genetic algorithm to obtain the optimized durian breeding parameters, and replacing the durian breeding parameters in the original steps with the optimized durian breeding parameters include:

[0043] Receive and determine the optimization goals of the genetic algorithm, which include improving the breeding efficiency of durian grafted seedlings, shortening the breeding cycle, and improving the quality of durian;

[0044] Encoding the various durian breeding parameters in steps S101 to S105 to form individuals in a genetic algorithm, each individual comprising a set of breeding parameter values, the breeding parameter values ​​comprising the type of disinfectant, the concentration of the disinfectant, the soaking time, the setting of the laser generator, the burying depth of the seedling, the detection frequency of the sensor, and the component ratio of the matrix nutrient bag;

[0045] A certain number of individuals are randomly generated to form an initial population, each of which represents a possible combination of breeding parameters;

[0046] According to the optimization goal, the fitness function is defined. The fitness function is calculated based on the length of the breeding cycle, the quality of the durian seedlings, and the survival rate.

[0047] According to the result of fitness function, individuals with higher fitness are selected as parents for crossover and mutation operations, and the parents are selected by tournament selection method;

[0048] Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals;

[0049] Mutation operation, performing mutation operation on offspring individuals;

[0050] Replace the parent individuals with lower fitness with the generated offspring individuals and update the population;

[0051] The operations of selection, crossover, mutation and population update are repeatedly performed until a preset number of iterations or optimization target is reached. When the genetic algorithm reaches the stopping condition, the optimized durian breeding parameters are output. The optimized durian breeding parameters include the specific parameter values ​​involved in steps S101 to S105.

[0052] The optimized durian breeding parameters are used to replace the original breeding parameters in steps S101 to S105, and the durian grafted seedlings are rapidly bred according to the optimized durian breeding parameter values.

[0053] Furthermore, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0054] The specific steps of encoding the various durian breeding parameters in steps S101 to S105 to form individuals in the genetic algorithm, each individual including a set of breeding parameter values, are as follows:

[0055] Receive and determine the encoding method, the encoding method includes real number encoding;

[0056] List the breeding parameters involved in steps S101 to S105, including:

[0057] The types of disinfectants were converted into numerical codes: 1 for potassium permanganate and 2 for carbendazim;

[0058] A real number encoding the disinfectant concentration;

[0059] Soaking time real number encoding, in minutes;

[0060] Encode the laser intensity and frequency of the laser generator with real numbers;

[0061] The burying depth of the seedlings is coded as a real number in centimeters;

[0062] The detection frequency of the sensor is encoded as a real number, in units of times / hour or minute;

[0063] The component ratio of the substrate nutrient bag, each component ratio is coded with a real number;

[0064] For each breeding parameter, set the value range and construct an individual structure to store a set of breeding parameter values. Under real number coding, the individual is represented as a vector or array, and each element corresponds to the value of a breeding parameter.

[0065] For each breeding parameter, according to its value range and encoding method, it is converted into individual representation in the genetic algorithm;

[0066] After the individual structure and encoding method are determined, a certain number of individuals are randomly generated to form the initial population. Each individual represents a set of breeding parameter combinations, and its parameter values ​​are randomly selected within a certain range.

[0067] Furthermore, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0068] According to the optimization goal, the fitness function is defined. The fitness function is calculated based on the length of the breeding cycle, the quality of the durian seedlings, and the survival rate index, including the following steps:

[0069] Determine the evaluation indicators, quantify each evaluation indicator, and base the evaluation indicators on the quantified ones;

[0070] The formula to define the fitness function is:

[0071] Fitness = w1*(1 / breeding cycle)+w2*quality score+w3*survival rate;

[0072] Among them, w1, w2, w3 are weight coefficients;

[0073] Receive an individual as input, calculate the breeding cycle, quality and survival rate according to the individual's breeding parameter values, and then substitute them into the fitness function formula to calculate the fitness value.

[0074] The method for rapid propagation of grafted durian seedlings proposed in the present application embodiment has at least the following advantages:

[0075] The present invention optimizes the seedling embedding method and growth environment, and uses a laser generator to maintain the seedling spacing. The present invention effectively improves the growth efficiency and migration ability of the seedlings, greatly shortening the period of seedlings forming rootstocks. The grafted seedlings are transplanted into a special matrix nutrient bag, further shortening the rootstock maintenance time, and the overall breeding cycle is shortened by more than 30% compared with traditional methods.

[0076] During the growth of seedlings, the use of the second disinfectant effectively eliminated pests and diseases and enhanced the disease resistance of grafted seedlings. The proportioned matrix nutrient bags provided comprehensive nutritional support for grafted seedlings and promoted their healthy growth. The cleft grafting method was used for grafting, and the scion and rootstock were strictly disinfected and protected to ensure the high-quality healing of the grafting interface, thereby improving the survival rate of grafted seedlings.

[0077] The invention effectively saves manpower, pesticide and fertilizer input and improves resource utilization through scientific environmental control and nutrition management. The large number of high-quality grafted seedlings bred quickly provides a reliable seedling source for growers, shortens the production cycle of durian trees, and significantly improves economic benefits.

[0078] By installing a variety of sensors and industrial cameras, real-time monitoring and intelligent management of the durian growth environment are achieved. Advanced random forest models and genetic algorithms are used to optimize and analyze breeding data, and breeding parameters are adjusted according to real-time data, realizing dynamic optimization of the breeding process and further improving breeding efficiency.

[0079] In summary, the rapid breeding method of durian grafted seedlings provided by the present invention not only significantly shortens the breeding cycle and improves the quality and survival rate of grafted seedlings, but also saves resources and improves economic benefits. It is also simple to operate and easy to promote, has important promotion value and application prospects, and provides strong support for the sustainable development of the durian planting industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 A schematic diagram of the process of a method for rapid propagation of grafted durian seedlings provided in an embodiment of the present application. DETAILED DESCRIPTION

[0081] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. On the contrary, these embodiments are provided to make the disclosure of the present application more thorough and complete, and to fully convey the scope of the disclosure of the present application to those skilled in the art.

[0082] Reference Figure 1 The present application provides a method for rapid propagation of grafted durian seedlings, which comprises:

[0083] Building a seedling bed, laying clean river sand with a thickness of 15-25 cm on the built seedling bed, disinfecting the seedling bed by spraying a first disinfectant, and cleaning the disinfected seedling bed;

[0084] Based on the collected clean, fresh and full durian seeds, soak them in a second disinfectant for at least 30 minutes, wash the soaked durian seeds, and germinate the washed durian seeds;

[0085] A plurality of laser generators are arranged on the upper surface of the seedling bed in a staggered manner with an adjacent spacing of 10 cm, and the lasers generated by the laser generators are used to separate the upper surface of the seedling bed into a plurality of cultivation areas, each of which is used to plant a durian seed after germination;

[0086] Select durian seeds with hypocotyls 1-3 cm long growing from the hilum as seedlings to be cultivated, bury each seedling in river sand in a row of 10×10 cm with the hypocotyl facing downward, spray the second disinfectant, cover with greenhouse film, and set a sunshade net with a shading degree of 40-70%;

[0087] Extend the detection end of the sensor into the river sand, the depth of the detection end of the sensor into the river sand is greater than the depth of the seedlings to be cultivated buried in the river sand, use the sensor to detect the environment in the seed shed, and maintain the temperature of the environment in the seed shed at 25-35°C and the humidity at 70-90%;

[0088] On day 25-28, select seedlings with a height of 10-20 cm, pull them out and use them as juvenile rootstocks for grafting;

[0089] The grafted seedlings are transplanted into a preset matrix nutrient bag, and watering, fertilization, sucker removal, and pest and disease control are carried out regularly. When the buds of the scion grow to 7.5-20 cm, the durian seedlings are considered to have completed breeding. The matrix nutrient bag includes: perlite, vermiculite, coconut bran, humus, peat soil, garden soil and organic fertilizer.

[0090] Exemplarily, the first disinfectant is implemented as 0.3% potassium permanganate disinfectant. The second disinfectant is implemented as 800 times liquid of 50% carbendazim wettable powder.

[0091] The size of the seedling bed is 1.2×10m. Clean river sand with a thickness of 15-25cm is evenly spread on the seedling bed. 0.3% potassium permanganate disinfectant is used as the first disinfectant to spray the seedling bed. An arch shed is built on the seedling bed and covered with shed film for disinfection for more than 24 hours.

[0092] Pull out and use it as a young rootstock for grafting. The durian grafting rapid breeding method provided by the present application comprises:

[0093] The pulled out seedlings are used as young rootstocks, the young rootstocks are cut off with a grafting knife at a position 5-10 cm away from the sowing mark, and the young rootstocks are split longitudinally along the length direction to form a V-shaped knife mark with a depth of 1.5-3 cm;

[0094] Healthy, high-quality, pest-free durian buds with full buds are collected as scions. Each scion is cut into small segments with 1-3 buds, each segment has 1-2 leaves and 2 / 3 of the leaves are cut off, and the lower end of the scion is cut into a wedge with a wedge length of 1-3 cm;

[0095] The scion is grafted onto the young rootstock using the cleft grafting method, the scion and the young rootstock are tightly bound using a grafting film, and the scion is bagged.

[0096] Before transplanting the grafted seedlings into a preset substrate nutrient bag, the method provided by the present application further includes:

[0097] crushing the lumps in the matrix nutrient bag and uniformly mixing the components in the matrix nutrient bag;

[0098] Use a sprayer to water the substrate nutrient bag;

[0099] After watering the substrate nutrient bag, the substrate nutrient bag is left standing for 5-10 minutes.

[0100] The matrix nutrient bag comprises: 15%-25% of perlite, 5%-10% of vermiculite, 10%-20% of coconut bran, 10%-20% of humus soil, 10%-20% of peat soil, 25%-35% of garden soil and 15%-25% of organic fertilizer.

[0101] The sensors used include temperature sensors and humidity sensors. The detection ends of the temperature sensor and the humidity sensor are inserted into the river sand, and the depth of the detection ends of the temperature sensor and the humidity sensor extended into the river sand is greater than the depth of the seedlings to be cultivated buried in the river sand, thereby realizing the detection of the ambient temperature and humidity of the river sand, ensuring that the growth of the seedlings in the seed shed is in a suitable ambient temperature and humidity.

[0102] The method of accelerating germination of durian seeds comprises:

[0103] Set up a first moisturizing layer horizontally in a weak light place, place durian seeds on the first moisturizing layer at intervals, and then cover the durian seeds with a second moisturizing layer;

[0104] The second moisturizing layer is sprayed evenly and intermittently with water using a sprayer to collect the humidity between the first moisturizing layer and the second moisturizing layer, and the durian seeds are kept moist based on the humidity to fully germinate the durian seeds.

[0105] Exemplarily, the first moisturizing layer and the second moisturizing layer can both be implemented as newspapers, a layer of newspapers is horizontally laid out in a weak light indoors, the durian seeds are laid flat on the newspapers, and another layer of newspapers is covered on the flat durian seeds;

[0106] Use a sprayer to spray water on the second layer of newspaper intermittently to keep the durian seeds moist and fully germinate them.

[0107] The cleaning of the sterilized seedling bed comprises:

[0108] The nozzle of the sprayer is arranged on the seedling bed after the first disinfectant is sprayed, and clean water is sprayed evenly above the seedling bed;

[0109] The clean water flowing out of the seedling bed is obtained, and the content of the first disinfectant in the clean water flowing out of the seedling bed is detected. When the content of the first disinfectant in the clean water flowing out of the seedling bed is reduced to 2%-3%, spraying the seedling bed is stopped.

[0110] When each seedling to be cultivated is buried in river sand in a row of 10×10 cm with the hypocotyl facing downward, the method further comprises:

[0111] After covering the surface of each seedling to be cultivated with river sand, evenly spread a layer of coconut bran on the surface of the river sand;

[0112] Use a sprayer to spray clean water on the river sand covered with coconut bran.

[0113] The method for rapid propagation of grafted durian seedlings proposed in the embodiment of the present application has at least the following advantages:

[0114] By burying the radicle of the seedlings to be cultivated downward in river sand, the growth migration of the seedlings to be cultivated is improved, and excellent seedlings to be cultivated with fast growth speed are screened;

[0115] During the growth of the seedlings to be cultivated, the second disinfectant is used to eliminate durian pests and diseases, thereby greatly improving the resistance of the grafted seedlings to durian pests and diseases;

[0116] The laser light generated by the laser generator is used to establish multiple cultivation areas with an area of ​​10×10 cm, so as to maintain the spacing between adjacent seedlings to be cultivated and reduce the mutual interference between adjacent seedlings to be cultivated during the growth process, thereby maintaining the growth efficiency of each seedling to be cultivated;

[0117] Maintain the temperature in the seed shed at 25-35℃ and the humidity at 70-90% to ensure the growth environment of the seedlings and shorten the period for the seedlings to form rootstocks;

[0118] By transplanting the grafted seedlings into a preset matrix nutrient bag, the maintenance time of the rootstock during the durian grafted seedling breeding process is shortened, effectively saving the input of manpower, pesticides and fertilizers, while improving the resource utilization rate, solving the technical problem of the long cycle required for the current method of breeding durian by grafting fruit trees, and achieving the purpose of greatly shortening the breeding cycle of the durian grafted seedlings.

[0119] Embodiment 1:

[0120] In this embodiment, the matrix nutrient bag includes: 15% perlite, 5% vermiculite, 20% coconut bran, 10% humus, 10% peat soil, 25% garden soil and 15% organic fertilizer.

[0121] Transplant the grafted durian seedlings into the corresponding substrate nutrient bags, plant 30 grafted seedlings in each substrate, set up 3 groups of repetitions, 10 plants in each group, water, fertilize, remove suckers, and control pests and diseases regularly, wait for the scion to germinate, and record the germination time when germination is observed. Record the germination time and number of germinations in each group of repetitions, calculate the germination cycle and survival rate, and evaluate the quality of the grafted seedlings;

[0122] Test results:

[0123] Among the three groups of repetitions of matrix A, the number of germinations in the first and third groups was 9, and the survival rate was 90%. The number of germinations in the second group was 8, and the survival rate was 80%. The average number of germinations in the three groups was 8.67, and the average survival rate was 86.67%. The average number of new buds was 21, and the average bud length was 18.12 cm.

[0124] Table 1A Matrix test results statistics

[0125] repeat Number of grafts (plants) Number of germinations (plants) Survival rate (%) Number of new sprouts Total bud length (cm) 1 10 9 90 21 185.5 2 10 8 80 24 178.4 3 10 9 90 18 179.8 average 10 8.67 86.67 21 181.2

[0126] Embodiment 2:

[0127] In this embodiment, the matrix nutrient bag includes: 25% perlite, 5% vermiculite, 10% coconut bran, 10% humus, 10% peat soil, 25% garden soil and 15% organic fertilizer.

[0128] Among the three groups of repetitions of B matrix, the number of sprouts in the first group was 8, and the survival rate was 80%, the number of sprouts in the second group was 7, and the survival rate was 70%, and the number of sprouts in the third group was 9, and the survival rate was 90%; the average number of sprouts in the three groups was 8, and the average survival rate was 80%, the average number of new sprouts was 22, and the average sprout length was 17.6 cm.

[0129] Table 2B Matrix test results statistics

[0130]

[0131]

[0132] Embodiment 3:

[0133] In this embodiment, the matrix nutrient bag includes: 10% vermiculite, 15% coconut bran, 15% humus, 10% peat soil, 25% garden soil and 25% organic fertilizer.

[0134] Among the three groups of C matrix repetitions, the number of sprouts in the first, second and third groups was 6, and the survival rate was 60%; the average number of new sprouts in the three groups was 12.67, and the average sprout length was 14.61 cm.

[0135] Table 3C Matrix Test Results Statistics

[0136] repeat Number of grafts (plants) Number of germinations (plants) Survival rate (%) Number of new sprouts Total bud length (cm) 1 10 6 60 13 143.6 2 10 6 60 11 145.1 3 10 6 60 14 149.4 average 10 6 60 12.67 146.1

[0137] Embodiment 4:

[0138] In this embodiment, the matrix nutrient bag includes: 25% perlite, 10% coconut bran, 15% humus, 10% peat soil, 25% garden soil and 15% organic fertilizer.

[0139] Among the three groups of repetitions of D matrix, the number of germination in the first group was 6, and the survival rate was 60%. The number of germination in the second and third groups was 7, and the survival rate was 70%. The average number of germination in the three groups was 6.67, the average survival rate was 66.67%, the average number of new sprouts was 15.33, and the average sprout length was 14.3 cm.

[0140] Table 4D Matrix test results statistics

[0141]

[0142]

[0143] Embodiment 5:

[0144] In this embodiment, the matrix nutrient bag includes: 25% perlite, 15% vermiculite, 10% humus, 10% peat soil, 25% garden soil and 15% organic fertilizer.

[0145] Among the three groups of repetitions of E matrix, the number of germinations in the first and second groups was 7, with a survival rate of 70%, and the number of germinations in the third group was 8, with a survival rate of 80%; the average number of germinations in the three groups was 7.3, with an average survival rate of 73.33%, an average number of new buds of 17.33, and an average bud length of 15.61 cm.

[0146] Table 5E Matrix test results statistics

[0147] repeat Number of grafts (plants) Number of germinations (plants) Survival rate (%) Number of new sprouts Total bud length (cm) 1 10 7 70 15 152.1 2 10 7 70 17 153.7 3 10 8 80 20 162.5 average 10 7.3 73.33 17.33 156.1

[0148] Embodiment 6:

[0149] In this embodiment, the matrix nutrient bag includes: 25% perlite, 5% vermiculite, 10% coconut bran, 10% peat soil, 25% garden soil and 15% organic fertilizer.

[0150] Among the three groups of repetitions of F matrix, the number of germination in the first and third groups was 7, and the survival rate was 70%. The number of germination in the second group was 6, and the survival rate was 60%. The average number of germination in the three groups was 6.7, and the average survival rate was 66.67%. The average number of new buds was 17.33, and the average bud length was 15.58 cm.

[0151] Table 6F Matrix test results statistics

[0152]

[0153]

[0154] Embodiment 7:

[0155] In this embodiment, the matrix nutrient bag includes: 25% perlite, 10% vermiculite, 10% coconut bran, 15% humus, 25% garden soil and 15% organic fertilizer.

[0156] Among the three groups of repetitions of G matrix, the number of germination in the first and second groups was 6, and the survival rate was 60%. The number of germination in the third group was 7, and the survival rate was 70%. The average number of germination in the three groups was 6.3, the average survival rate was 63.33%, the average number of new sprouts was 18.33, and the average sprout length was 15.89 cm.

[0157] Table 7G Matrix test results statistics

[0158] repeat Number of grafts (plants) Number of germinations (plants) Survival rate (%) Number of new sprouts Total bud length (cm) 1 10 6 60 18 157.1 2 10 6 60 16 152.4 3 10 7 70 21 167.3 average 10 6.3 63.33 18.33 158.9

[0159] Embodiment 8:

[0160] In this embodiment, the matrix nutrient bag includes: 25% perlite, 5% vermiculite, 10% coconut bran, 15% humus, 15% peat soil and 25% organic fertilizer.

[0161] Among the three groups of repetitions of H matrix, the number of germinations in the first and second groups was 7, and the survival rate was 70%. The number of germinations in the third group was 8, and the survival rate was 80%. The average number of germinations in the three groups was 7.3, and the average survival rate was 73.33%. The average number of new buds was 21.67, and the average bud length was 16.97 cm.

[0162] Table 8H Matrix test results statistics

[0163]

[0164]

[0165] Embodiment 9:

[0166] In this embodiment, the matrix nutrient bag includes: 25% perlite, 5% vermiculite, 10% coconut bran, 10% humus, 15% peat soil and 35% garden soil.

[0167] Among the three groups of repetitions of I matrix, the number of germination in the first group was 6, and the survival rate was 60%. The number of germination in the second and third groups was 7, and the survival rate was 70%; the average number of germination in the three groups was 6.7, and the average survival rate was 66.67%. The average number of new sprouts was 19.33, and the average sprout length was 15.63 cm.

[0168] Table 9I Matrix test results statistics

[0169] repeat Number of grafts (plants) Number of germinations (plants) Survival rate (%) Number of new sprouts Total bud length (cm) 1 10 6 60 22 164.7 2 10 7 70 17 149.6 3 10 7 70 19 154.5 average 10 6.7 66.67 19.33 156.3

[0170] The results show that the method for rapid propagation of grafted durian seedlings provided in the present application uses young durian seedlings as rootstocks and sets the cultivation matrix to include perlite, vermiculite, coconut bran, humus, peat soil, garden soil and organic fertilizer, etc., which is a unique technical means to shorten the breeding cycle (3-6 months) by more than 30%, and conveniently and quickly provide growers with a large number of high-quality grafted seedlings. At the same time, the grafted seedlings bred have a high grafting survival rate and a fast growth rate. They can be transplanted 25 days after grafting, which greatly saves the cultivation time of durian seedlings and improves the cultivation efficiency of durian.

[0171] Specifically, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0172] Temperature sensors and humidity sensors are installed in the seed shed to monitor and maintain the optimal temperature and humidity range in the seed shed in real time. The optimal temperature range is 25-35°C and the optimal humidity range is 70-90%.

[0173] Install light sensors to obtain light data during germination and growth;

[0174] A soil moisture sensor and a nutrient sensor are set in the matrix nutrient bag. The soil moisture sensor is used to collect soil moisture data, and the nutrient sensor is used to collect soil fertilization data;

[0175] Industrial cameras are installed in the greenhouse and on the soil to collect image data during the growth process of durian;

[0176] The data collected by the temperature sensor, humidity sensor, light sensor, soil moisture sensor, nutrient sensor and industrial camera are transmitted to a preset database for storage, and the durian grafted seedling rapid breeding experimental data generated in real time by the user mobile terminal and the background control terminal are stored in the preset database.

[0177] In the rapid propagation method of durian grafted seedlings, the installation and use of sensors and cameras are as follows:

[0178] Temperature sensor: monitors the temperature in the seed shed in real time. Maintains the temperature in the seed shed in the optimal range, i.e. 25-35°C, to promote the growth of durian seedlings.

[0179] Humidity sensor: monitors the humidity in the seed shed in real time. Maintain the humidity in the seed shed in the optimal range, i.e. 70-90%, to provide a suitable humidity environment for durian seedlings.

[0180] Light sensor: monitors light data during germination and growth, allowing durian seeds and seedlings to grow under suitable light conditions and improve breeding efficiency.

[0181] Soil moisture sensor: collects soil moisture data in the matrix nutrient bag. Real-time monitoring of soil moisture ensures that the durian seedlings get the right amount of water to avoid excessive or insufficient water that may adversely affect the growth of the seedlings.

[0182] Nutrient sensor: collects soil fertilization data in the matrix nutrient bag. Monitors soil nutrient content, provides data support for scientific fertilization, and ensures that durian seedlings get adequate nutrition.

[0183] Industrial camera: Collect image data during the growth of durian. Record the growth status of durian seedlings to facilitate monitoring and management, and timely discover and deal with problems in the growth process.

[0184] Data transmission and storage: The data collected by the above sensors are transmitted to the preset database for storage. This facilitates data analysis and management, and stores the durian grafted seedling rapid breeding experimental data generated in real time by the user's mobile terminal and the background control terminal, providing data support for subsequent optimization and improvement.

[0185] In summary, by installing and using these sensors and cameras, the environmental conditions in the seed shed and the growth conditions of the durian seedlings can be monitored and controlled in real time, thereby improving the breeding efficiency and quality of durian grafted seedlings.

[0186] Specifically, the method for rapid propagation of grafted durian seedlings of the present invention obtains historical data of rapid propagation of grafted durian seedlings from a preset database, and the historical data of rapid propagation of grafted durian seedlings includes historical data of seedling bed disinfection, historical data of durian seed treatment, historical data of seedling growth environment, historical data of seedling growth, historical data of grafting, historical data of the number of components in substrate nutrient bags, historical data of the number of grafted seedlings grown, and historical data of experimental data;

[0187] The historical data of rapid propagation of durian grafted seedlings is divided into a training set and a validation set, and the random forest model is trained using the training set and the validation set to obtain a simulation model for rapid propagation of durian grafted seedlings;

[0188] Obtain the start time of durian grafting in the area monitored by the industrial camera, use the real-time data of durian in the area monitored by the industrial camera to substitute the real-time data of durian in the area monitored by the industrial camera and the start time of durian grafting in the area monitored by the industrial camera into the durian grafting rapid breeding simulation model, and output the expected growth data of durian;

[0189] The starting time of the durian grafting seedlings in the monitoring area of ​​the industrial camera is matched in the preset durian grafting seedlings breeding knowledge base to obtain the standard durian growth data and durian growth error range value corresponding to the starting time of the durian grafting seedlings in the monitoring area of ​​the industrial camera;

[0190] The standard flow growth data is compared with the expected growth data of durian to obtain the expected growth error data of real-time flow, and the expected growth error data of real-time flow is compared with the durian growth error range value. If the expected growth error data of real-time flow exceeds the durian growth error range value, the durian breeding parameters in step S101, step S102, step S103, step S104 and step S105 of the rapid breeding method for grafted seedlings of durian are optimized to obtain optimized durian breeding parameters, and the optimized durian breeding parameters are used to replace the various durian breeding parameters in step S101, step S102, step S103, step S104 and step S105.

[0191] Data acquisition: Get the historical data of rapid propagation of durian grafted seedlings from the preset database. These data cover multiple aspects such as seedbed disinfection, durian seed treatment, seedling growth environment, seedling growth, grafting, matrix nutrient bag composition, grafted seedling growth, and experimental data.

[0192] Data division: The acquired historical data of rapid propagation of durian grafted seedlings are divided into a training set and a validation set. The training set is used to train the model, and the validation set is used to verify the accuracy and reliability of the model.

[0193] Model training: The random forest model was trained using the training set and validation set to obtain a simulation model for rapid propagation of durian grafted seedlings. This model can predict the growth of durian grafted seedlings based on historical data.

[0194] Real-time data acquisition: Get the start time of durian grafting in the area monitored by the industrial camera, as well as the durian growth data monitored in real time by the industrial camera. These data include environmental parameters such as temperature, humidity, light, soil moisture, nutrients, and the growth status of durian seedlings.

[0195] Data substitution model: Substitute the real-time durian growth data and the start time of grafting into the durian grafting rapid breeding simulation model. The model will output the expected growth data of durian based on these data, that is, predict the future growth of durian seedlings.

[0196] Standard data matching: The start time of the durian grafting seedlings in the monitoring area of ​​the industrial camera is matched with the preset durian grafting seedling breeding knowledge base. The knowledge base stores the standard durian growth data and durian growth error range values ​​corresponding to different grafting seedling start times. Through matching, the standard growth data and error range corresponding to the current grafting seedling start time can be obtained.

[0197] Data comparison: Compare the expected durian growth data output by the model with the standard durian growth data, and calculate the expected growth error data of real-time traffic. Then, compare this error data with the durian growth error range value to evaluate the error situation in the current breeding process.

[0198] Parameter optimization: if the expected growth error data of real-time flow exceeds the durian growth error range value, it is illustrated that some parameters in the current breeding process may not be suitable for current growth environment or seedling condition. At this point, it is necessary to optimize the various durian breeding parameters (such as the setting of disinfectant type, disinfectant concentration, soak time, laser generator, the embedding depth of seedling, the detection frequency of sensor and the composition ratio of matrix nutrition bag etc., corresponding steps S101 to S105) in the durian grafted seedling rapid breeding method. By adjusting these parameters, the breeding process is more in line with the growth demand of durian seedling, and breeding efficiency and quality are improved.

[0199] The optimized durian breeding parameters will replace the parameters in the original steps to achieve dynamic adjustment and optimization of the breeding process.

[0200] Through this data optimization process, the breeding process of grafted durian seedlings can be controlled more accurately, ensuring that the bred durian seedlings have excellent quality and growth performance.

[0201] Specifically, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0202] The specific steps of optimizing the durian breeding parameters in step S101, step S102, step S103, step S104 and step S105 in the rapid breeding method of durian grafted seedlings using a genetic algorithm to obtain the optimized durian breeding parameters, and replacing the durian breeding parameters in the original steps with the optimized durian breeding parameters include:

[0203] Receive and determine the optimization goals of the genetic algorithm, which include improving the breeding efficiency of durian grafted seedlings, shortening the breeding cycle, and improving the quality of durian;

[0204] Encoding the various durian breeding parameters in steps S101 to S105 to form individuals in a genetic algorithm, each individual comprising a set of breeding parameter values, the breeding parameter values ​​comprising the type of disinfectant, the concentration of the disinfectant, the soaking time, the setting of the laser generator, the burying depth of the seedling, the detection frequency of the sensor, and the component ratio of the matrix nutrient bag;

[0205] A certain number of individuals are randomly generated to form an initial population, each of which represents a possible combination of breeding parameters;

[0206] According to the optimization goal, the fitness function is defined. The fitness function is calculated based on the length of the breeding cycle, the quality of the durian seedlings, and the survival rate.

[0207] According to the result of fitness function, individuals with higher fitness are selected as parents for crossover and mutation operations, and the parents are selected by tournament selection method;

[0208] Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals;

[0209] Mutation operation, performing mutation operation on offspring individuals;

[0210] Replace the parent individuals with lower fitness with the generated offspring individuals and update the population;

[0211] The operations of selection, crossover, mutation and population update are repeatedly performed until a preset number of iterations or optimization target is reached. When the genetic algorithm reaches the stopping condition, the optimized durian breeding parameters are output. The optimized durian breeding parameters include the specific parameter values ​​involved in steps S101 to S105.

[0212] The optimized durian breeding parameters are used to replace the original breeding parameters in steps S101 to S105, and the durian grafted seedlings are rapidly bred according to the optimized durian breeding parameter values.

[0213] Determine the optimization goals of the genetic algorithm: The main optimization goals include improving the breeding efficiency of durian grafted seedlings, shortening the breeding cycle, and improving the quality of durian. These goals together constitute the direction of genetic algorithm optimization.

[0214] Encoding the breeding parameters: Encode the various durian breeding parameters in steps S101 to S105 (these steps usually cover various links of rapid breeding of durian grafted seedlings, such as seedbed disinfection, durian seed treatment, seedling growth environment setting, grafting operation and configuration of matrix nutrient bags, etc.) to form individuals in the genetic algorithm.

[0215] Each individual contains a set of breeding parameter values, which specifically include the type of disinfectant, disinfectant concentration, immersion time, laser generator settings (if applicable), seedling burial depth, sensor detection frequency, and the composition ratio of the substrate nutrient bag.

[0216] Generate initial population: Randomly generate a certain number of individuals, which together constitute the initial population.

[0217] Each individual represents a set of possible breeding parameter combinations, providing a basis for subsequent genetic algorithm optimization.

[0218] Define fitness function: According to the optimization goal, define a fitness function. This function is used to evaluate the quality of each individual (i.e. each set of breeding parameter combinations).

[0219] The specific calculation method of the fitness function may be based on multiple indicators such as the length of the breeding cycle, the quality of the durian seedlings (such as growth rate, health status, etc.) and the survival rate.

[0220] Select parent individuals: According to the results of the fitness function, select individuals with higher fitness as parents for subsequent crossover and mutation operations. Use selection strategies such as tournament selection to ensure that the selected parent individuals have higher optimization potential.

[0221] Crossover operation: Perform a crossover operation on the selected parent individuals to generate new offspring individuals. This step simulates the gene crossover process in biological genetics.

[0222] Through the crossover operation, individuals with new characteristics can be generated, increasing the diversity of the population.

[0223] Mutation operation: Perform mutation operation on offspring individuals to introduce new gene mutations. This step simulates the gene mutation process in biological genetics. Mutation operation helps to jump out of the local optimal solution and improve the global search ability of the genetic algorithm.

[0224] Update the population: Replace the parent individuals with lower fitness with the generated offspring individuals and update the population. This step ensures that individuals with higher fitness are always maintained in the population.

[0225] Repeat the optimization process: Repeat the selection, crossover, mutation and population update operations until the preset number of iterations or optimization goal is reached. This step is the core process of genetic algorithm optimization. Through continuous iterative optimization, it gradually approaches the optimal solution.

[0226] Output optimized breeding parameters: When the genetic algorithm reaches the stopping condition (such as reaching the preset number of iterations or the optimization target is no longer improved), the optimized durian breeding parameters are output.

[0227] These optimized breeding parameters include the specific parameter values ​​involved in steps S101 to S105, and are important achievements in the rapid breeding method of durian grafted seedlings.

[0228] Replacement of breeding parameters: Replacement of the original breeding parameters in steps S101 to S105 with the optimized durian breeding parameters. Rapid breeding of durian grafted seedlings is performed according to the optimized durian breeding parameter values, in order to obtain higher breeding efficiency and better durian quality.

[0229] Through this series of optimization steps, various breeding parameters in the rapid breeding method of durian grafted seedlings have been effectively optimized and improved, providing strong support for the efficient breeding and quality improvement of durian.

[0230] Specifically, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0231] The specific steps of encoding the various durian breeding parameters in steps S101 to S105 to form individuals in the genetic algorithm, each individual including a set of breeding parameter values, are as follows:

[0232] Receive and determine the encoding method, the encoding method includes real number encoding;

[0233] List the breeding parameters involved in steps S101 to S105, including:

[0234] The types of disinfectants were converted into numerical codes: 1 for potassium permanganate and 2 for carbendazim;

[0235] A real number encoding the disinfectant concentration;

[0236] Soaking time real number encoding, in minutes;

[0237] Encode the laser intensity and frequency of the laser generator with real numbers;

[0238] The burying depth of the seedlings is coded as a real number in centimeters;

[0239] The detection frequency of the sensor is encoded as a real number, in units of times / hour or minute;

[0240] The component ratio of the substrate nutrient bag, each component ratio is coded with a real number;

[0241] For each breeding parameter, set the value range and construct an individual structure to store a set of breeding parameter values. Under real number coding, the individual is represented as a vector or array, and each element corresponds to the value of a breeding parameter.

[0242] For each breeding parameter, according to its value range and encoding method, it is converted into individual representation in the genetic algorithm;

[0243] After the individual structure and encoding method are determined, a certain number of individuals are randomly generated to form the initial population. Each individual represents a set of breeding parameter combinations, and its parameter values ​​are randomly selected within a certain range.

[0244] Determine the encoding method In genetic algorithms, commonly used encoding methods include binary encoding, real number encoding, etc. For durian breeding parameters, since most of them are continuous or discrete but meaningful values, real number encoding can be selected.

[0245] List all breeding parameters involved in steps S101 to S105, including:

[0246] Type of disinfectant (can be converted into a numerical code, such as 1 for potassium permanganate, 2 for carbendazim, etc.);

[0247] Disinfectant concentration (real coded);

[0248] Soaking time (real number encoding, in minutes);

[0249] Laser generator settings (may include laser intensity, frequency, etc., all coded with real numbers);

[0250] The burying depth of seedlings (real number code, unit is centimeter);

[0251] The detection frequency of the sensor (real number encoding, in times / hour or minute)

[0252] The composition ratio of the substrate nutrient bag (each composition ratio is coded as a real number, such as 15% perlite, which is coded as 0.15)

[0253] For each breeding parameter, determine its reasonable value range. This will help ensure that the individual parameter values ​​generated are within a reasonable range when the initial population is generated later.

[0254] Construct an individual structure to store a set of breeding parameter values. Under real number encoding, an individual can be represented as a vector or array, with each element corresponding to a breeding parameter value.

[0255] Implement the encoding process. For each breeding parameter, according to its value range and encoding method, convert it into an individual representation in the genetic algorithm.

[0256] Type of disinfectant: If potassium permanganate is selected as 1 and carbendazim is selected as 2, the code is 1 or 2.

[0257] Disinfectant concentration: If the concentration range is 0.1% to 1%, you can choose a real number (such as 0.3) to represent a concentration of 0.3%.

[0258] Soaking time: If the time range is 30 to 60 minutes, you can choose a real number (such as 45) to represent a soaking time of 45 minutes.

[0259] And so on, encode other parameters.

[0260] After determining the individual structure and encoding method, a certain number of individuals can be randomly generated to form the initial population. Each individual represents a possible combination of breeding parameters, and its parameter value is randomly selected within a certain range.

[0261] Specifically, the method for rapid propagation of grafted durian seedlings of the present invention further comprises:

[0262] According to the optimization goal, the fitness function is defined. The fitness function is calculated based on the length of the breeding cycle, the quality of the durian seedlings, and the survival rate index, including the following steps:

[0263] Determine the evaluation indicators, quantify each evaluation indicator, and base the evaluation indicators on the quantified ones;

[0264] The formula to define the fitness function is:

[0265] Fitness = w1*(1 / breeding cycle)+w2*quality score+w3*survival rate;

[0266] Among them, w1, w2, w3 are weight coefficients;

[0267] Receive an individual as input, calculate the breeding cycle, quality and survival rate according to the individual's breeding parameter values, and then substitute them into the fitness function formula to calculate the fitness value.

[0268] First, clarify the specific indicators for evaluating individual fitness:

[0269] The length of the breeding cycle: The shorter the breeding cycle, the higher the breeding efficiency, and therefore the higher the fitness should be.

[0270] Quality of durian seedlings: Quality can be evaluated by multiple dimensions such as appearance, growth rate, health status, etc. The higher the quality, the higher the fitness should be.

[0271] Survival rate: The survival rate reflects the effectiveness of the breeding method. The higher the survival rate, the more reliable the method is and the higher the fitness should be.

[0272] In order to convert the evaluation indicators into calculable values, each indicator needs to be quantified:

[0273] Breeding cycle: Days or hours can be used directly as quantitative values.

[0274] Quality of durian seedlings: A quality scoring system can be set up, for example, from 1 to 10 points, and scores can be given based on appearance, growth rate, health status, etc.

[0275] Survival rate: Expressed as a percentage, for example, a 90% survival rate is converted to 0.9.

[0276] Based on the quantified evaluation indicators, define the formula of the fitness function. A common approach is to weight the sum of the various indicators, but other more complex function forms, such as product, exponential, etc., can also be considered, depending on the needs of the specific problem and the optimization goal.

[0277] Assuming we use the weighted sum method, the fitness function can be defined as:

[0278] Fitness = w1*(1 / breeding cycle)+w2*quality score+w3*survival rate

[0279] Among them, w1, w2, and w3 are weight coefficients, which are used to adjust the relative importance of each indicator in the fitness function. The selection of weight coefficients needs to be determined according to the actual problem and optimization goal, and can be set through experiments or experience.

[0280] When implementing programming, you need to define a function to calculate fitness. This function will receive an individual as input, calculate the breeding cycle, quality and survival rate based on the individual's breeding parameter values, and then substitute them into the fitness function formula to calculate the fitness value.

[0281] Finally, the fitness function needs to be debugged and verified to ensure that it can correctly evaluate the fitness of individuals. You can check whether the results of the fitness function meet expectations by calculating a set of individuals with known breeding parameter values. If problems are found, you need to adjust the definition of the fitness function or the setting of the weight coefficient in time.

[0282] Through the above steps, a fitness function based on the length of the breeding cycle, the quality of the durian seedlings, and the survival rate can be defined according to the optimization goal to evaluate the quality of individuals in the genetic algorithm.

[0283] The present invention effectively solves the problem of long durian breeding cycle by fruit tree grafting through a series of technical innovations and refined management measures. The following are specific technical solutions and measures:

[0284] Burying the radicle downwards in river sand: By burying the radicle of the seedlings to be cultivated in river sand downwards, the growth and migration speed of the seedlings is increased, thereby screening out excellent seedlings with fast growth rates, laying the foundation for shortening the breeding cycle.

[0285] Use of the second disinfectant: During the growth of seedlings, the second disinfectant (such as 800 times diluted 50% carbendazim wettable powder) is used to eliminate durian pests and diseases, thereby improving the resistance of grafted seedlings to pests and diseases, reducing the impact of diseases on the growth cycle, and ensuring the healthy growth of seedlings.

[0286] Temperature and humidity control: Maintaining the ambient temperature in the seed shed at 25-35℃ and the humidity at 70-90% provides a suitable growth environment for the seedlings, which is conducive to the rapid growth of the seedlings, thereby shortening the period for the seedlings to form rootstocks.

[0287] Application of matrix nutrient bags: Transplant the grafted seedlings into the preset matrix nutrient bags, which contain a variety of optimized ingredients (such as perlite, vermiculite, coconut bran, humus, peat soil, garden soil and organic fertilizer), providing balanced nutritional support for the seedlings, optimizing the growth environment, and shortening the maintenance time of the rootstock during the breeding process.

[0288] Scientific grafting technology: Using scientific grafting methods (such as cleft grafting) to graft seedlings in time when they grow to a certain height improves the success rate of grafting, shortens the time from sowing to grafting, and further shortens the breeding cycle.

[0289] Real-time monitoring and data optimization: By installing a variety of sensors and industrial cameras, the growth environment of seedlings is monitored in real time, and advanced random forest models and genetic algorithms are used to optimize and analyze breeding parameters, thereby improving the efficiency and accuracy of the breeding process and improving breeding efficiency.

[0290] Dynamically adjust breeding parameters: Based on real-time monitoring data and model prediction results, dynamically adjust various parameters in the breeding process (such as disinfectant type, concentration, soaking time, etc.) to cope with different growth environments and seedling conditions, ensuring the flexibility and effectiveness of the breeding process.

[0291] Nutrient bag matrix optimization: The optimal ratio of matrix nutrient bag components is determined through experiments to provide balanced nutrition for seedlings, promote their rapid growth, and further shorten the breeding cycle.

[0292] Efficient germination technology: Using scientific germination technology to improve the germination efficiency of durian seeds, shorten the germination time, and gain time for the subsequent breeding process.

[0293] Comprehensive management measures: During the breeding process, comprehensive management measures such as regular watering, fertilization, removal of suckers, and disease and pest control are adopted to ensure the healthy growth of seedlings, improve breeding efficiency, and further shorten the breeding cycle.

[0294] In summary, the present invention effectively solves the problem of long durian breeding cycle by fruit tree grafting, improves breeding efficiency and resource utilization, and provides strong support for the sustainable development of durian planting industry.

[0295] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

[0296] In addition, although the operations of the method of the present application are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

Claims

1. A method for rapid propagation of grafted seedlings of durian, characterized in that, The following steps are involved: Step S101, constructing a seedling bed, laying clean river sand with a thickness of 15-25 cm on the constructed seedling bed, disinfecting the seedling bed by spraying a first disinfectant, and cleaning the disinfected seedling bed; Step S102, based on the collected clean, fresh and full durian seeds, soaking them in a second disinfectant for at least 30 minutes, cleaning the soaked durian seeds, and germinating the washed durian seeds, wherein the first disinfectant is implemented as 0.3% potassium permanganate disinfectant, and the second disinfectant is implemented as 800 times liquid of 50% carbendazim wettable powder; Step S103, arranging a plurality of laser generators staggered and spaced 10 cm apart from each other on the upper surface of the seedling bed, wherein the lasers generated by the laser generators are used to separate the upper surface of the seedling bed into a plurality of cultivation areas, each of which is used to plant a durian seed after germination; Step S104, selecting durian seeds with hypocotyls 1-3 cm long at the hilum as seedlings to be cultivated, burying each seedling in river sand in rows of 10×10 cm with the hypocotyl facing downward, spraying the second disinfectant, covering with greenhouse film, and setting a sunshade net with a shading degree of 40-70%; Step S105, extending the detection end of the sensor into the river sand, the depth of the detection end of the sensor extending into the river sand is greater than the depth of the seedlings to be cultivated buried in the river sand, using the sensor to detect the environment in the seed shed, maintaining the temperature of the environment in the seed shed at 25-35°C and the humidity at 70-90%, and burying each seedling to be cultivated in the river sand with the radicle facing downward and 10×10 cm in rows, further comprising: after covering the surface of each seedling to be cultivated with river sand, evenly spreading a layer of coconut bran on the surface of the river sand , use a sprayer to spray clean water on the river sand covered with coconut bran once, select seedlings with a height of 10-20 cm in 25-28 days, pull them out and use them as young rootstocks for grafting, transplant the grafted seedlings into a preset matrix nutrient bag, water and fertilize them regularly, remove suckers, and prevent and control diseases and pests. When the buds of the scion grow to 7.5-20 cm, they are used as durian seedlings that have completed breeding. The matrix nutrient bag includes: perlite, vermiculite, coconut bran, humus, peat soil, garden soil and organic fertilizer.

2. A method for rapid propagation of grafted seedlings of durian as claimed in claim 1, characterized in that, The matrix nutrient bag includes: 15% to 25% perlite, 5% to 10% vermiculite, 10% to 20% coconut bran, 10% to 20% humus, 10% to 20% peat soil, 25% to 35% garden soil and 15% to 25% organic fertilizer; The pulling out and grafting as a juvenile rootstock comprises: The pulled out seedlings are used as young rootstocks, the young rootstocks are cut off with a grafting knife at a position 5-10 cm away from the sowing mark, and the young rootstocks are split longitudinally along the length direction to form a V-shaped knife mark with a depth of 1.5-3 cm; Healthy, high-quality, pest-free durian buds with full buds are collected as scions. Each scion is cut into small segments with 1-3 buds, each segment has 1-2 leaves and 2 / 3 of the leaves are cut off, and the lower end of the scion is cut into a wedge with a wedge length of 1-3 cm; The scion is grafted onto the young stock by cleft grafting, the scion and the young stock are tightly bound with a grafting film, and the scion is bagged; Before transplanting the grafted seedlings into the preset substrate nutrient bag, the method further comprises: crushing the lumps in the matrix nutrient bag and uniformly mixing the components in the matrix nutrient bag; Use a sprayer to water the substrate nutrient bag; After watering the substrate nutrient bag, the substrate nutrient bag is placed for 5-10 minutes; The method of crushing the lumps in the matrix nutrient bag and uniformly mixing the components in the matrix nutrient bag comprises: Put the components in the matrix nutrient bag into a grinder, and use the grinder to grind the lumps in the matrix nutrient bag; The crushed components in the matrix nutrient bag are put into a blender, and the blender is used to mix the components in the matrix nutrient bag.

3. A method for rapid propagation of grafted seedlings of durian as claimed in claim 1, characterized in that, The method of accelerating germination of durian seeds comprises: Set up a first moisturizing layer horizontally in a weak light place, place durian seeds on the first moisturizing layer at intervals, and then cover the durian seeds with a second moisturizing layer; The second moisturizing layer is sprayed evenly and intermittently with water by a sprayer, and the humidity between the first moisturizing layer and the second moisturizing layer is collected, and the durian seeds are kept moist based on the humidity to fully germinate the durian seeds; The cleaning of the sterilized seedling bed comprises: The nozzle of the sprayer is arranged on the seedling bed after the first disinfectant is sprayed, and clean water is sprayed evenly above the seedling bed; The clean water flowing out of the seedling bed is obtained, and the content of the first disinfectant in the clean water flowing out of the seedling bed is detected. When the content of the first disinfectant in the clean water flowing out of the seedling bed is reduced to 2%-3%, spraying the seedling bed is stopped.

4. A method for rapid propagation of grafted seedlings of durian as claimed in claim 1, characterized in that, Also includes: Temperature sensors and humidity sensors are installed in the seed shed to monitor and maintain the optimal temperature and humidity range in the seed shed in real time. The optimal temperature range is 25-35°C and the optimal humidity range is 70-90%. Install light sensors to obtain light data during germination and growth; A soil moisture sensor and a nutrient sensor are set in the matrix nutrient bag. The soil moisture sensor is used to collect soil moisture data, and the nutrient sensor is used to collect soil fertilization data; Industrial cameras are installed in the greenhouse and on the soil to collect image data during the growth process of durian; The data collected by the temperature sensor, humidity sensor, light sensor, soil moisture sensor, nutrient sensor and industrial camera are transmitted to a preset database for storage, and the durian grafted seedling rapid breeding experimental data generated in real time by the user mobile terminal and the background control terminal are stored in the preset database.

5. A method for rapid propagation of grafted seedlings of durian as claimed in claim 4, characterized in that, From the preset database, the rapid propagation history data of grafted seedlings of durian is obtained. The rapid propagation history data of grafted seedlings of lotus includes the disinfection history data of seedling bed, the treatment history data of durian seeds, the growth environment history data of seedlings, the growth history data of seedlings, the grafting history data, the number of components of substrate nutrient bags, the number of grafted seedlings, and the experimental data history data; The historical data of rapid propagation of durian grafted seedlings is divided into a training set and a validation set, and the random forest model is trained using the training set and the validation set to obtain a simulation model for rapid propagation of durian grafted seedlings; Obtain the start time of durian grafting in the area monitored by the industrial camera, use the real-time data of durian in the area monitored by the industrial camera to substitute the real-time data of durian in the area monitored by the industrial camera and the start time of durian grafting in the area monitored by the industrial camera into the durian grafting rapid breeding simulation model, and output the expected growth data of durian; The starting time of the durian grafting seedlings in the monitoring area of ​​the industrial camera is matched in the preset durian grafting seedlings breeding knowledge base to obtain the standard durian growth data and durian growth error range value corresponding to the starting time of the durian grafting seedlings in the monitoring area of ​​the industrial camera; The standard flow growth data is compared with the expected growth data of durian to obtain the expected growth error data of real-time flow, and the expected growth error data of real-time flow is compared with the durian growth error range value. If the expected growth error data of real-time flow exceeds the durian growth error range value, the durian breeding parameters in step S101, step S102, step S103, step S104 and step S105 of the rapid breeding method for grafted seedlings of durian are optimized to obtain optimized durian breeding parameters, and the optimized durian breeding parameters are used to replace the various durian breeding parameters in step S101, step S102, step S103, step S104 and step S105.

6. A method for rapid propagation of grafted seedlings of durian as claimed in claim 5, characterized in that, Also includes: The specific steps of optimizing the durian breeding parameters in step S101, step S102, step S103, step S104 and step S105 in the rapid breeding method of durian grafted seedlings using a genetic algorithm to obtain the optimized durian breeding parameters, and replacing the durian breeding parameters in the original steps with the optimized durian breeding parameters include: Receive and determine the optimization goals of the genetic algorithm, which include improving the breeding efficiency of durian grafted seedlings, shortening the breeding cycle, and improving the quality of durian; Encoding the various durian breeding parameters in steps S101 to S105 to form individuals in a genetic algorithm, each individual comprising a set of breeding parameter values, the breeding parameter values ​​comprising the type of disinfectant, the concentration of the disinfectant, the soaking time, the setting of the laser generator, the burying depth of the seedling, the detection frequency of the sensor, and the component ratio of the matrix nutrient bag; A certain number of individuals are randomly generated to form an initial population, each of which represents a possible combination of breeding parameters; According to the optimization goal, the fitness function is defined, and the fitness function is calculated based on the length of the breeding cycle, the quality of the durian seedlings, and the survival rate index; According to the result of fitness function, individuals with higher fitness are selected as parents for crossover and mutation operations, and the parents are selected by tournament selection method; Crossover operation: perform crossover operation on the selected parent individuals to generate new offspring individuals; Mutation operation, performing mutation operation on offspring individuals; Replace the parent individuals with lower fitness with the generated offspring individuals and update the population; Repeat the operations of selection, crossover, mutation and population update until a preset number of iterations or optimization target is reached. When the genetic algorithm reaches a stop condition, the optimized durian breeding parameters are output, and the optimized durian breeding parameters include the specific parameter values ​​involved in steps S101 to S105; The optimized durian breeding parameters are used to replace the original breeding parameters in steps S101 to S105, and the durian grafted seedlings are rapidly bred according to the optimized durian breeding parameter values.

7. A method for rapid propagation of grafted seedlings of durian as claimed in claim 6, characterized in that, Also includes: The specific steps of encoding the various durian breeding parameters in steps S101 to S105 to form individuals in the genetic algorithm, each individual including a set of breeding parameter values, are as follows: Receive and determine the encoding method, the encoding method includes real number encoding; List the breeding parameters involved in steps S101 to S105, including: The types of disinfectants were converted into numerical codes: 1 for potassium permanganate and 2 for carbendazim; A real number encoding the disinfectant concentration; Soaking time real number encoding, in minutes; Encode the laser intensity and frequency of the laser generator with real numbers; The burying depth of the seedlings is coded as a real number in centimeters; The detection frequency of the sensor is encoded as a real number, in units of times / hour or minute; The component ratio of the substrate nutrient bag, each component ratio is coded with a real number; For each breeding parameter, set the value range and construct an individual structure to store a set of breeding parameter values. Under real number coding, the individual is represented as a vector or array, and each element corresponds to the value of a breeding parameter. For each breeding parameter, according to its value range and encoding method, it is converted into individual representation in the genetic algorithm; After the individual structure and encoding method are determined, a certain number of individuals are randomly generated to form the initial population. Each individual represents a set of breeding parameter combinations, and its parameter values ​​are randomly selected within a certain range.

8. A method for rapid propagation of grafted seedlings of durian as claimed in claim 6, characterized in that, Also includes: According to the optimization goal, the fitness function is defined. The fitness function is calculated based on the length of the breeding cycle, the quality of the durian seedlings, and the survival rate index, including the following steps: Determine the evaluation indicators, quantify each evaluation indicator, and base the evaluation indicators on the quantified ones; The formula to define the fitness function is: Fitness = w1*(1 / breeding cycle)+w2*quality score+w3*survival rate; Among them, w1, w2, w3 are weight coefficients; Receive an individual as input, calculate the breeding cycle, quality and survival rate according to the individual's breeding parameter values, and then substitute them into the fitness function formula to calculate the fitness value.

Images