A dynamic weighing control method in long strip bare seedling transplanting process

Abstract

The application discloses a kind of long strip bare seedling transplanting process in dynamic weighing control method, it is related to agricultural intelligent equipment technical field, including S1, obtain environmental humidity data and fog concentration data, whether the data is judged to exceed normal range by preset threshold value, obtain sensor surface condensation risk assessment result under high humidity fog state, S2, according to the condensation risk assessment result, using heating module activates sensor surface temperature control mechanism, determines temperature rise value to prevent water droplet formation and maintain sensor surface dry state;The long strip bare seedling transplanting process in dynamic weighing control method, closed-loop control is realized.The present application has significantly improved the precision and survival rate of nursery stock transplanting, improves agricultural production efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural intelligent equipment, in particular to a dynamic weighing control method in long strip bare seedling transplanting process. BACKGROUND

[0002] Weighing technology is crucial in agricultural production, especially in the process of seedling transplanting. Precise weight data directly affects planting decisions and survival rates. By accurately measuring seedling weight, soil ratio, irrigation volume, and transplanting timing can be optimized, thereby improving the efficiency and quality of agricultural production. However, the influence of high humidity and fog environment on weighing precision becomes a pressing problem, especially in greenhouses, coastal areas, or rainy regions. Ensuring the stability and accuracy of weighing equipment in complex environments is the key to promoting agricultural intelligence and refinement. Existing weighing equipment often performs poorly in high humidity and fog environments, mainly due to the susceptibility of sensors to environmental interference. Traditional protective measures, such as simple sealed casings, can prevent water from entering to some extent, but cannot cope with the condensation of fog on the sensor surface. This condensation not only adds extra weight to the sensor surface, but also changes the sensitivity of the sensor, leading to distorted measurement data. In addition, existing devices often lack the ability to dynamically adapt to environmental changes, such as when humidity changes rapidly, protective devices cannot adjust in time, leading to decreased precision. The core technical difficulty lies in effectively preventing the influence of fog condensation on the sensor while maintaining high sensitivity and stability of the equipment. First, fog condensation forms water droplets on the sensor surface, changing its force state and causing measurement errors. For example, in a greenhouse, the sensor surface may record additional water droplet weight due to fog condensation when weighing seedlings, resulting in higher data and misleading soil ratio decisions during the transplanting process. Second, sensors need to maintain sensitivity in high humidity environments for a long time, while moisture may penetrate into the device, affecting the stability of electronic components, and thus leading to inaccurate measurements or even equipment damage. The contradiction between moisture condensation and sensor sensitivity is the key to technological breakthrough. SUMMARY

[0003] The purpose of the present application is to provide a dynamic weighing control method in long strip bare seedling transplanting process, which solves the problems existing in the prior art.

[0004] To achieve the above purpose, the present application provides the following technical scheme: a dynamic weighing control method in long strip bare seedling transplanting process, the method comprising the following steps:

[0005] S1, obtaining environmental humidity data and fog concentration data, judging whether the data exceeds the normal range through a preset threshold, and obtaining the sensor surface condensation risk assessment result under high humidity and fog state;

[0006] S2, according to the condensation risk assessment result, a heating module is used to activate a sensor surface temperature control mechanism, and a temperature increase value is determined to prevent water droplet formation and maintain a dry state of the sensor surface;

[0007] S3, an initial seedling weight signal is obtained from the sensor in a dry state, the signal is processed by a signal filtering algorithm, whether the filtered signal is stable is judged, and weight data free from noise interference is obtained;

[0008] S4, for the weight data, a calibration algorithm is used to compare a preset standard weight curve, a deviation value is determined and data output is adjusted, and an accurate seedling weight value is obtained for transplanting decision-making;

[0009] S5, soil proportioning parameters are calculated through the accurate seedling weight value, if the calculated parameters exceed an optimization threshold, an adjustment instruction is triggered, a proportioning correction scheme is obtained to match irrigation optimization requirements;

[0010] S6, an irrigation amount index is extracted from the correction scheme, a prediction model is used to simulate survival rate trend, whether the simulated trend reaches an expected level is judged, and a final transplanting timing parameter is determined;

[0011] S7, according to the transplanting timing parameter, all data streams are integrated to generate a comprehensive report data package, the data package is transmitted to an agricultural system interface through the data package, and closed-loop control of the entire weighing process is completed.

[0012] Preferably, the environmental humidity data and fog concentration data are obtained by an environment acquisition module arranged around the sensor, the environment acquisition module includes a humidity sensor and a fog sensor, the humidity sensor and the fog sensor respectively monitor air humidity and aerosol concentration in real time, and the detection results are sent to a central control unit through a wireless communication module for condensation risk judgment.

[0013] Preferably, the heating module includes a flexible heating sheet and a temperature control unit, the temperature control unit controls the flexible heating sheet to work according to the condensation risk evaluation result, and the flexible heating sheet is attached to the surface of the weighing sensor shell to maintain the shell temperature higher than the environmental dew point temperature.

[0014] Preferably, the signal filtering algorithm is a Kalman filtering algorithm or a low-pass filtering algorithm, the filtering algorithm continuously processes the initial seedling weight signal according to a set time window, removes transient fluctuation terms, and outputs a relatively stable net weight signal as the weight data.

[0015] Preferably, the calibration algorithm includes a regression model established based on historical measurement samples, the regression model inputs current weight data and environmental parameters, outputs a deviation value from the standard weight curve, and realizes data correction according to the deviation value by calling a preset weight compensation function.

[0016] Preferably, the prediction model is a combination structure of integrated support vector regression model SVR and random forest model, which takes accurate seedling weight value, corrected irrigation amount and historical survival rate data as input and outputs expected survival rate trend of seedlings under different transplanting time.

[0017] Preferably, the comprehensive report data package includes environmental parameter data, dynamic weighing data, irrigation proportioning parameter, transplanting time parameter and predicted survival rate curve, which is packaged in JSON format and uploaded to the agricultural Internet of Things platform interface through the MQTT protocol.

[0018] Preferably, the temperature rise value is determined based on the difference between the current surface temperature of the sensor and the environmental dew point temperature, the required power is obtained by looking up the temperature rise-power mapping table, and the corresponding power is output by the heating module to realize temperature regulation.

[0019] Preferably, the standard weight curve is preset according to different seedling species, the seedling species is identified and classified by a visual recognition module, and the visual recognition module includes an industrial camera and a seedling image database, which is used to automatically match the corresponding weight curve template.

[0020] Preferably, the generation of the proportioning correction scheme is based on the current seedling weight value, the target growth period, the seedbed substrate type and the historical proportioning optimization data, and the adjustment instruction is generated by the central processing unit based on multi-parameter cross comparison.

[0021] From the above technical solution, the present application has the following beneficial effects:

[0022] The dynamic weighing control method in the long strip bare seedling transplanting process solves the problem of accurate transplanting under the influence of multiple factors such as humidity, fog, weight signal noise and soil proportioning, etc. in view of the complexity of seedling weighing, environmental control and transplanting decision in agricultural production. The present application collects environmental humidity and fog concentration data, evaluates the condensation risk of the sensor surface in combination with the preset threshold value, activates the heating module to dynamically adjust the temperature, prevents water droplets from forming, and ensures the stable operation of the sensor; then the initial seedling weight signal is processed by a signal filtering algorithm to eliminate noise interference, and the standard weight curve is compared by a calibration algorithm to output accurate weight data; based on this data, the soil proportioning parameter is calculated, the optimization adjustment instruction is triggered, and the proportioning correction scheme is generated; finally, the prediction model simulates the irrigation amount and the survival rate trend to determine the best transplanting time, integrates the data stream to generate a comprehensive report, and transmits it to the agricultural system interface to realize closed-loop control. The present application significantly improves the accuracy and survival rate of seedling transplanting and optimizes the efficiency of agricultural production. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1The flow chart of the dynamic weighing control method in the long strip bare seedling transplanting process. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0025] As shown in the drawings, Figure 1 The present application provides a technical solution: a dynamic weighing control method in the long strip bare seedling transplanting process, which comprises the following steps:

[0026] S1, obtaining environmental humidity data and fog concentration data, judging whether the data exceeds the normal range through a preset threshold, and obtaining a sensor surface condensation risk assessment result under a high humidity and fog state;

[0027] S2, according to the condensation risk assessment result, using a heating module to activate a sensor surface temperature control mechanism, determining a temperature rise value to prevent water droplet formation and maintain a dry state of the sensor surface;

[0028] S3, obtaining an initial seedling weight signal from the sensor in a dry state, processing the signal through a signal filtering algorithm, judging whether the filtered signal is stable, and obtaining weight data free of noise interference;

[0029] S4, for the weight data, using a calibration algorithm to compare with a preset standard weight curve, determining a deviation value and adjusting data output, and obtaining an accurate seedling weight value for transplanting decision;

[0030] S5, calculating soil proportioning parameters through the accurate seedling weight value, triggering an adjustment instruction if the calculated parameters exceed an optimization threshold, obtaining a proportioning correction scheme to match irrigation optimization requirements;

[0031] S6, extracting an irrigation amount index from the correction scheme, using a prediction model to simulate a survival rate change trend, judging whether the simulated trend reaches an expected level, and determining a final transplanting timing parameter;

[0032] S7, integrating all data streams according to the transplanting timing parameter, generating a comprehensive report data package, transmitting the data package to an agricultural system interface through the data package, and completing the closed-loop control of the entire weighing process.

[0033] The method first obtains environmental humidity data and fog concentration data through the environmental sensing module arranged in the transplanting area. The humidity data is obtained by a relative humidity sensor and is in percentage. The fog concentration data is collected by a scattering fog concentration monitor in real time and is in milligrams per cubic meter. The system presets a humidity threshold of 85% and a fog concentration threshold of 20 milligrams per cubic meter. When any real-time data exceeds the threshold, the system determines that the environment is in a high-humidity fog state. At this time, the system automatically starts the condensation risk assessment module. The module compares the current environmental data, the real-time collection results of the sensor material thermal conductivity and the surface temperature. If the surface temperature is lower than the environmental dew point temperature by more than 2 degrees Celsius, it is determined that there is a surface condensation risk.

[0034] Under the premise that the condensation risk is established, the system drives the heating module connected to the surface of the sensor. The heating module uses a thermistor to control the current size and uses the resistance heating principle to raise the surface temperature of the sensor. Specifically, the system sets the target surface temperature to be 5 degrees Celsius higher than the current dew point temperature. The heating module collects the surface temperature in real time and controls the temperature rise rate to be 0.5 degrees Celsius per second until the target temperature is maintained constant for more than 5 seconds, confirming that the surface reaches a dry state. After entering the dry state, the system starts collecting the seedling weight signal. The weight signal is derived from a resistance strain type weighing sensor installed below the weighing platform, and its output is a continuous analog voltage signal. The system converts the analog signal to a digital signal through an analog-to-digital conversion module at a frequency of 100 times per second and continuously collects a data sequence of not less than 5 seconds.

[0035] Subsequently, the system performs signal filtering processing on the collected weight signal. The filtering process uses a sliding mean filtering algorithm. Specifically, 500 consecutive data points are divided into 10 groups in chronological order, with each group containing 50 data points. The arithmetic mean of each group of data is calculated as the representative value of that time period. Then, it is determined whether the difference between the adjacent two average values is less than 0.5 grams. If this stability condition is met, the filtering result is confirmed to be stable. If not, the next batch of data is continuously collected and the above steps are repeated until the condition is met.

[0036] When a stable weight signal is obtained, the system calls the calibration algorithm and compares it with the preset standard weight curve. The standard curve is established from historical data and corresponds to the target weight range defined for the seedling size. For example, for seedlings with a height of 30 centimeters, the standard weight range is 180 grams to 220 grams. The calibration process compares the current filtered weight with the nearest point on the standard curve. If the deviation is greater than 5 grams, a correction value is calculated. The correction value is obtained by weighting the historical error mean and the current deviation. The system adds the correction value back to the original filtered result to obtain the final accurate seedling weight value.

[0037] The system further calculates the soil ratio parameters for transplanting according to the accurate seedling weight value. The soil ratio includes three materials: nutrient soil, peat soil, and perlite, and the proportion is determined according to the weight of the seedling. For example, for a seedling weighing 200 grams, the recommended total amount of soil is set to 1000 grams, and the initial recommended ratio of the three components is 5:3:2, i.e. 500 grams, 300 grams, and 200 grams respectively. The system compares the recommended ratio with the recommended standard associated with the current actual measured weight, and if the deviation of any component calculated value exceeds 10%, the ratio correction mechanism is triggered. The ratio correction is based on the ratio adjustment record in the database corresponding to the weight interval, combined with the actual deviation ratio for weighted correction, outputting new component mass values to form a ratio correction scheme.

[0038] The system then extracts the irrigation amount index from the correction scheme, which is the total mass of the soil multiplied by the set water content coefficient. The water content coefficient is determined according to the seedling variety and the current environmental temperature, for example, when the environmental temperature is 25 degrees Celsius, the recommended water content coefficient for this seedling is 0.35, then the corrected total mass of the soil is 1050 grams, corresponding to an irrigation water amount of 367.5 grams. The system inputs this irrigation amount into the survival rate prediction model, which is a multivariate regression model established based on historical samples, the input variables include seedling weight, soil ratio, irrigation amount and environmental temperature, and the output variable is the predicted survival rate percentage. The system calculates the survival rate curve trend under the current parameters, if the survival rate prediction value is higher than 90%, it is determined that the current transplanting scheme reaches the expected level, and the current system time, environmental conditions and parameter combination are extracted as the final transplanting opportunity parameters.

[0039] Finally, the system packages all the key data points in the above processes, including environmental monitoring data, temperature control log, weight collection and correction results, soil ratio adjustment scheme, irrigation amount and survival rate prediction value, into a unified format data package. The data package is packaged in binary structured format and uploaded to the central agricultural decision platform through the agricultural system communication interface, and the platform records, reviews and further instructs feedback after receiving the data, marking the completion of the dynamic weighing control process of the seedling.

[0040] The environmental humidity data and fog concentration data are obtained by an environmental acquisition module arranged around the sensor, which includes a humidity sensor and a fog sensor that respectively monitor air humidity and aerosol concentration in real time, and send the detection results to the central control unit through a wireless communication module for condensation risk judgment.

[0041] In this embodiment, a set of environmental acquisition modules are fixedly arranged outside the sensor structure of the dynamic weighing control system. The module structure includes a humidity sensor and a mist sensor, which are respectively used to collect the relative humidity data and aerosol particle concentration in the air. The humidity sensor selects a digital sensor based on the principle of capacitance change, with an accuracy of ±2%, a response time of less than 3 seconds, and a measurement unit of percentage; the mist sensor adopts a laser scattering type sensing technology, with a detection range of 0 to 100 milligrams per cubic meter, and a response time of less than 2 seconds. The two sensors continuously collect target parameters in the air at a frequency of 5 times per second through a high-speed sampling module, and transmit the detected real-time data through an internally integrated wireless communication module. The wireless communication module adopts a low-power wireless transmission technology based on the 2.4 GHz frequency band, with automatic channel switching and anti-interference functions to ensure data integrity.

[0042] The collected data is sent to the central control unit via the wireless communication module. The central control unit is the core processor of the system. After receiving the real-time air humidity and aerosol concentration data, it immediately compares with the preset threshold standard. The humidity threshold is set to 85%, and the aerosol concentration threshold is set to 20 milligrams per cubic meter. When the data feedback by any sensor exceeds the corresponding threshold, the system determines that the sensor surface is in a high-humidity mist state. This judgment result is directly used as the trigger condition for starting the subsequent condensation risk assessment mechanism, and enters the temperature regulation process of the heating module to prevent water droplets from condensing in a high-humidity environment and causing errors to the weighing sensor, thereby realizing the pre-protection of the stability of the sensor.

[0043] The heating module includes a flexible heating sheet and a temperature regulation unit. The temperature regulation unit controls the operation of the flexible heating sheet according to the condensation risk assessment result. The flexible heating sheet is attached to the surface of the weighing sensor's shell to maintain the shell temperature above the environmental dew point temperature.

[0044] In this embodiment, a heating module composed of a flexible heating sheet and a temperature regulation unit is set up to solve the condensation problem that may occur on the surface of the sensor in a high-humidity environment. The flexible heating sheet is a resistance heating element made of polyimide or silicone rubber material, with a thickness not exceeding 1 millimeter, which can tightly adhere to the surface of the metal shell of the weighing sensor, covering at least more than 70% of the sensor shell. The temperature regulation unit includes a thermistor, a temperature control chip and a control circuit, which collects the temperature data of the heating sheet surface and compares it with the real-time calculated environmental dew point temperature. The environmental dew point temperature is calculated by the central control unit according to the real-time environmental temperature and relative humidity. When the temperature of the sensor shell surface is within 3 degrees Celsius of the dew point temperature, the system considers that there is a condensation risk.

[0045] At this time, the temperature regulation unit sends a heating instruction to the flexible heating sheet, adjusts the current to control the heating sheet to heat up, and the heating rate is controlled at 0.3 degrees Celsius per second, and the target temperature is set to be 5 degrees Celsius higher than the ambient dew point temperature. The heating process continues until the surface temperature reaches the target value and is stably maintained for more than 10 seconds, and the system confirms that the sensor shell surface is dry, that is, the weighing signal collection can be started. In the heating state, the temperature regulation unit continuously monitors the surface temperature, and if the temperature fluctuation exceeds ±1 degree Celsius or the external humidity changes significantly, the system automatically adjusts the heating current to maintain the set temperature difference, so as to ensure that the shell surface is always in a non-condensation state, and to ensure the stability and reliability of the weighing data.

[0046] The signal filtering algorithm is a Kalman filtering algorithm or a low-pass filtering algorithm, which continuously processes the initial seedling weight signal according to a set time window, removes the instantaneous fluctuation term, and outputs a relatively stable net weight signal as the weight data.

[0047] In the present embodiment, in order to improve the accuracy and stability of the weighing data, the system uses a Kalman filtering algorithm or a low-pass filtering algorithm to continuously process the original signal after collecting the initial seedling weight signal. The original weight signal is output in the form of an analog voltage by the weighing sensor, and after analog-to-digital conversion, a discrete time series data is formed, and the data sampling frequency is 100 times per second. The system sets a time window for the filtering algorithm, and the time length is set to 5 seconds, that is, each batch processing contains 500 sampling points. The filtering processing is operated and processed in time sequence on these 500 data points.

[0048] When the Kalman filtering algorithm is used, the system first sets the initial state estimation value and the estimation error covariance, and then uses the prediction and update two-step iterative process to correct the weight state value at the current time after receiving the actual weight observation value at each sampling point. The prediction step calculates the prior estimation value at the current time by using the state estimation at the last time and the control input; the update step corrects the state estimation by weighting the difference between the actual observation value and the prior estimation value, and finally outputs the current optimal net weight estimation value. This process is executed in a loop in the entire time window, and finally a set of filtered and smoothed weight curves are output, and the mean value in the last stable interval is extracted as the current net weight data.

[0049] If the low-pass filtering algorithm is used, the system sets the filtering cutoff frequency to 1 Hz, and applies a first-order discrete low-pass filter to suppress high-frequency interference. The filter weights the ratio between the current input value and the last filtering output value through a weight coefficient, thereby effectively smoothing the signal mutation part and filtering out high-frequency fluctuations. After processing, the same continuous data segment with a fluctuation amplitude of not more than 0.5 grams in the last few seconds is extracted, and the average value is calculated and output as the final net weight signal.

[0050] The entire filtering process is provided with a stability judgment module, and the judgment condition is that the fluctuation amplitude of the output signal is less than 0.5 grams within 3 seconds, if the condition is established, the system considers that the net weight data is stable and reliable, and can be used for subsequent weight calibration and decision calculation. If the condition is not established, the system continues to collect the next time window data and repeats the processing until the stability standard is met.

[0051] The calibration algorithm includes a regression model established based on historical measurement samples, which inputs the current weight data and environmental parameters, outputs the deviation value from the standard weight curve, and realizes data correction according to the deviation value by calling a preset weight compensation function.

[0052] In the embodiment, after completing the preliminary weight signal acquisition and filtering process, the dynamic weighing system further calls the calibration algorithm to finely correct the obtained net weight data, so as to improve the weighing accuracy and eliminate the system error caused by environmental changes. The calibration algorithm is constructed based on a large number of historical measurement samples, and a regression modeling method is used to statistically model the related parameters including the real weight of the seedling, the weighing output value, the environmental temperature, the humidity, the wind speed and the like in the samples. The model takes the current filtered net weight data and real-time environmental parameters as input variables, estimates by using a linear regression or a polynomial regression method, and outputs the deviation value of the current data point from the standard weight curve.

[0053] The standard weight curve is preset according to the variety and specification of the seedling, and defines the weight range that should be possessed under a specific seedling height or seedling diameter. For example, the standard weight curve of the bare seedling with a seedling height of 30 cm varies between 190 grams and 210 grams. The deviation value output by the regression model represents the difference of the current weighing data from the standard interval, and the unit is gram. The system calls a preset weight compensation function according to the deviation value, and the function is a piecewise linear function structure, which sets the correction amplitude according to different deviation intervals. For example, if the deviation is within ±5 grams, the original value is processed, if the deviation is between ±5 grams and ±15 grams, the actual deviation is corrected by 80%, and if the deviation is greater than ±15 grams, the maximum correction value is limited to 12 grams to prevent error amplification caused by excessive correction.

[0054] The data correction process is automatically executed by the system, specifically: the current net weight data is added to the output value of the correction function to generate the final weight value. After the correction is completed, the system will also perform a boundary comparison with the standard curve, if the corrected data is still outside the standard range, the system records the data as an abnormal weight and sends a prompt signal for subsequent manual verification or automatic re-measurement. The entire correction process takes no more than 1 second, which is suitable for real-time dynamic weighing process.

[0055] The prediction model is a combination of a support vector regression (SVR) model and a random forest model. The prediction model takes the accurate seedling weight value, the corrected irrigation amount, and the historical survival rate data as inputs, and outputs the expected survival rate trend of the seedlings under different transplanting time conditions.

[0056] In this embodiment, to improve the scientificity and timeliness of seedling transplanting decision-making, the system introduces an integrated prediction model after the weighing data acquisition and soil ratio correction are completed to evaluate the expected survival rate trend of seedlings under different transplanting time conditions. The prediction model consists of two parts, namely, a support vector regression (SVR) model and a random forest model. The SVR model is suitable for regression prediction of small samples with high-dimensional features, and can capture the nonlinear relationship between seedling physiological indicators and survival rate. The random forest model has good anti-overfitting ability and feature importance identification ability, and can handle scenarios with large fluctuations in environmental variables. The system uses a weighted integration strategy to fuse the output results of the two models, improving the overall prediction stability and accuracy.

[0057] The model input parameters include the accurate seedling weight value, the corrected irrigation amount, and the historical survival rate data. The accurate seedling weight value is obtained from the final weight result after filtering and calibration, and the unit is grams. The corrected irrigation amount is calculated by the system according to the corrected ratio, and the unit is grams. The historical survival rate data is obtained by statistical analysis of the survival conditions of seedlings of the same variety and specification in different transplanting periods in the database, and the unit is percentage. The above three input parameters are standardized and input into the SVR and random forest models respectively. The SVR model outputs the continuous survival rate prediction curve under the current conditions, and the random forest model outputs the distributed survival rate estimates at multiple possible time points. The system fuses the output results of the two models according to the empirical weights, with the SVR model weight set to 0.6 and the random forest model weight set to 0.4.

[0058] The fusion output is an expected survival rate trend curve, with the horizontal axis representing the candidate transplanting time points in the future and the vertical axis representing the survival rate prediction values at the corresponding time points. The system further analyzes the trend curve to identify the time point corresponding to the maximum survival rate, which is returned as the current best transplanting time parameter and used for subsequent generation of transplanting recommendation reports or automatic control instructions. The prediction process is completed by the central control unit, with an average processing time of no more than 2 seconds, which can realize fast response and prediction support for dynamic field data.

[0059] The comprehensive report data package includes environmental parameter data, dynamic weighing data, irrigation ratio parameters, transplanting time parameters, and prediction survival rate curves. The data package is packaged in JSON format and uploaded to the agricultural Internet of Things platform interface through the MQTT protocol.

[0060] In this embodiment, to realize the data closed-loop management of the seedling dynamic weighing process, the system integrates all key data to form a unified data package after completing data acquisition, processing and decision calculation in each stage. The data package contains five core contents: the first is environmental parameter data, including real-time collected air temperature, relative humidity, fog concentration and calculated dew point temperature; the second is dynamic weighing data, including original weight signal, filtered net weight, calibration correction value and final confirmed weight value; the third is irrigation ratio parameter, specifically recording the mass ratio of three materials of nutrient soil, peat soil and perlite and the corresponding irrigation water quantity calculation value; the fourth is transplanting opportunity parameter, including the optimal transplanting time point and its corresponding time stamp judged by the system; and the fifth is the predicted survival rate curve, listing the survival rate prediction values of future continuous multiple candidate transplanting days in time sequence, constituting complete trend data.

[0061] All the above data are packaged by the central control unit after each weighing control process is completed, and the packaging format is standard JSON structure. Each field adopts a clear key-value pair form, and the field name adopts the naming specification of lower case letters and underscores, for example, “env_temperature”, “net_weight”, “irrigation_ratio”, “transplant_time” and “survival_rate_curve”, to ensure data structure consistency and platform parsing compatibility.

[0062] The packaged JSON data package is uploaded by the communication module integrated in the system. The communication process adopts the MQTT protocol, which is a lightweight message publishing and subscribing mechanism, has the characteristics of low bandwidth occupation, high reliability and support for QoS level, and is suitable for the communication needs of a large number of low-power devices in agricultural scenarios. The data package is published to the configured MQTT topic path, corresponding to the collection interface node of the agricultural Internet of Things platform. After the platform receives the data, it can automatically parse the JSON structure and write it into the database, while triggering subsequent tasks such as data visualization, expert system suggestion generation or automatic execution instructions, to realize remote collaborative management of the whole system.

[0063] The determination method of temperature rise value is: based on the difference between the current surface temperature of the sensor and the environmental dew point temperature, the required power is obtained by looking up the temperature rise-power mapping table, and the corresponding power is output by the heating module to realize temperature regulation.

[0064] In this embodiment, to achieve accurate control of the sensor surface temperature, the system needs to calculate the temperature rise value that the sensor surface should reach after determining the risk of condensation. The specific process is as follows: the central control unit first collects the current sensor shell surface temperature in real time through the temperature sensor, and combines the environmental temperature and relative humidity to calculate the current environmental dew point temperature using the humidity conversion formula. Then, the system performs a difference operation between the dew point temperature and the sensor surface temperature. If the difference is less than 5 degrees Celsius, the system determines that the temperature needs to be raised, and the temperature rise value is set to the sum of the difference and a safety margin, which is fixed at 3 degrees Celsius.

[0065] Once the temperature rise value is obtained, the system looks up the pre-established temperature rise-power mapping table, which is generated based on the heating module structure, the thermal conductivity of the material, and the experimental data of the heat conduction efficiency under the working environment. The mapping table lists the required power supply power under different environmental temperatures and target temperature rise values. For example, when the environmental temperature is 20 degrees Celsius and the target temperature rise value is 8 degrees Celsius, the mapping table indicates that the required power is 4 watts. The system controls the PWM modulation signal to adjust the actual output power of the flexible heating sheet according to the power value found. The heating module uses a constant power control mode and continuously works at the target power level until the sensor surface temperature probe feedbacks that the surface temperature reaches the target value and remains stable for more than 5 seconds.

[0066] In this process, the temperature control unit collects surface temperature data every second and judges the matching relationship between the temperature rise speed and the power response in real time. If the actual temperature rise speed deviates from the theoretical value by more than 20%, the output power is automatically adjusted, and the adjustment behavior is recorded for subsequent mapping table optimization. The entire temperature control process ensures that the heating module quickly raises the sensor surface temperature to a safe level under the premise of efficiency priority and controllable energy consumption, thereby effectively suppressing the risk of condensation.

[0067] The standard weight curve is preset according to different seedling types, which are identified and classified by a visual recognition module. The visual recognition module includes an industrial camera and a seedling image database, which are used to automatically match the corresponding weight curve template.

[0068] In this embodiment, to ensure that the standard weight curve in the calibration algorithm is accurately matched with the seedling type, the system introduces a visual recognition module to automatically classify the seedlings. This module consists of an image acquisition system composed of an industrial camera installed above the weighing channel. The industrial camera has an image resolution of not less than 5 million pixels and a frame rate of more than 30 frames per second, with automatic exposure and light compensation functions to adapt to changes in field lighting. Before the seedling is placed on the weighing platform, the system starts the visual recognition process, the industrial camera takes one front and one side image of the seedling, and the images are uploaded to the central control unit for analysis.

[0069] The central control unit calls a locally deployed seedling image database, which contains feature image datasets and corresponding standard weight curve templates for common seedling species (e.g., red maple, camphor tree, ginkgo, etc.). The system extracts features from the collected images, including seedling contour shape, leaf texture, branch structure, and other parameters, and classifies them using a convolutional neural network model. The model outputs the seedling species label with the highest recognition probability, and the system matches the preset standard weight curve template based on this label. The standard weight curve takes different specifications (such as seedling height and diameter) of the seedling species as the independent variable and outputs the corresponding weight interval as a calibration reference.

[0070] After completing the identification and curve matching, the system binds the weight template to the sample in the current weighing process and uses it as a reference line for subsequent calibration and deviation calculation, ensuring that each seedling is compared with the standard curve corresponding to its actual species, avoiding weighing errors or calibration failures caused by incorrect seedling species identification. The entire image recognition and template matching process takes an average of no more than 2 seconds, meeting the real-time requirements of the dynamic transplanting process.

[0071] The generation of the ratio correction scheme is based on the current seedling weight value, target growth period, seedbed substrate type, and historical ratio optimization data. The adjustment instruction is generated by the central processing unit based on multi-parameter cross comparison.

[0072] In this embodiment, to achieve accurate soil ratio management during seedling transplanting, the system further introduces a ratio correction mechanism to optimize and adjust the original ratio scheme based on the preliminary calculation of standard ratio parameters. This correction process is performed by the central processing unit, and the core parameters include the current seedling weight value, target growth period, seedbed substrate type, and historical ratio optimization data. The current seedling weight value is output by the dynamic weighing module and is the final accurate seedling weight value after calibration; the target growth period is input by the agricultural management system or user and is in days or weeks, defining the length of time the seedling grows in the seedbed; the seedbed substrate type is set by input, including common peat soil, vermiculite, coconut coir, and perlite, which differ in water retention, air permeability, and nutrient release capacity; the historical ratio optimization data is derived from the database records of past similar seedlings under similar environmental and period conditions, as well as actual growth effect evaluation records.

[0073] After receiving the above parameters, the central processing unit first analyzes the current weight value and the target period, determines the demand intensity of the seedling for water and nutrients, and compares the physical properties of the corresponding substrate to preliminarily evaluate its applicability. On this basis, the system selects successful samples that match the conditions from the historical optimization data of the ratio, extracts the corresponding soil ternary ratio scheme and key indicators such as survival rate and root development score, and performs weighted aggregation to generate a recommended ratio interval. The system then cross-compares the current initial ratio scheme with the recommended interval. If a component (such as peat or perlite) is outside the recommended interval, a ratio correction suggestion is generated, including the component name to be adjusted, the adjustment amplitude (in grams), and the adjustment direction (increase or decrease). All comparison and suggestion generation processes are based on rule engines and statistical optimization algorithms to ensure the executability and optimization expectations of the suggested scheme.

[0074] Finally, the system forms a complete ratio correction scheme and converts it into a standardized format adjustment instruction, including fields such as target component, adjustment value, execution sequence number, and instruction check code. The instruction is transmitted to the filling control module or the manual prompting system simultaneously to achieve automatic or assisted execution of the correction operation. The entire comparison and correction process responds within 1 second, enabling real-time ratio adjustment during dynamic weighing.

[0075] Although embodiments of the present application have been shown and described, it is to be understood that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A dynamic weighing control method in a long strip bare seedling transplanting process, characterized in that, The method comprises the following steps: S1, obtaining environmental humidity data and fog concentration data, and determining whether the data exceeds the normal range through a preset threshold to obtain a sensor surface condensation risk assessment result under a high humidity and fog state; S2, according to the condensation risk assessment result, a heating module is used to activate the sensor surface temperature control mechanism, and a temperature rise value is determined to prevent water droplet formation and maintain the sensor surface in a dry state; S3, an initial seedling weight signal is obtained from the sensor in a dry state, the signal is processed through a signal filtering algorithm, it is determined whether the filtered signal is stable, and weight data free from noise interference is obtained; S4, for the weight data, a calibration algorithm is used to compare a preset standard weight curve, a deviation value is determined, and data output is adjusted to obtain an accurate seedling weight value for transplanting decision-making; S5, soil ratio parameters are calculated through the accurate seedling weight value, if the calculated parameters exceed the optimization threshold, an adjustment instruction is triggered, a ratio correction scheme is obtained to match the irrigation optimization requirement; S6, an irrigation amount index is extracted from the correction scheme, a prediction model is used to simulate the survival rate trend, it is determined whether the simulated trend reaches the expected level, and the final transplanting time parameter is determined; S7, according to the transplanting time parameter, all data streams are integrated to generate a comprehensive report data package, the data package is transmitted to the agricultural system interface through the data package, and the closed-loop control of the entire weighing process is completed.

2. The dynamic weighing control method for long strip bare seedling transplanting process according to claim 1, characterized in that: The environmental humidity data and fog concentration data are obtained by an environment acquisition module arranged around the sensor, the environment acquisition module comprises a humidity sensor and a fog sensor, the humidity sensor and the fog sensor respectively monitor the air humidity and the aerosol concentration in real time, and the detection results are sent to a central control unit through a wireless communication module for condensation risk judgment.

3. The dynamic weighing control method for long strip bare seedling transplanting process according to claim 1, characterized in that: The heating module comprises a flexible heating sheet and a temperature control unit, the temperature control unit controls the work of the flexible heating sheet according to the condensation risk assessment result, and the flexible heating sheet is attached to the surface of the weighing sensor shell to maintain the shell temperature higher than the environmental dew point temperature.

4. The dynamic weighing control method for long strip bare seedling transplanting process according to claim 1, characterized in that: The signal filtering algorithm is a Kalman filtering algorithm or a low-pass filtering algorithm, the filtering algorithm continuously processes the initial seedling weight signal according to a set time window, removes the instantaneous fluctuation term, and outputs a relatively stable net weight signal as the weight data.

5. The dynamic weighing control method for long strip bare seedling transplanting process according to claim 1, characterized in that: The calibration algorithm comprises a regression model established based on historical measurement samples, the regression model inputs the current weight data and environmental parameters, outputs the deviation value from the standard weight curve, and realizes data correction according to the preset weight compensation function.

6. The dynamic weighing control method for long strip bare seedling transplanting process according to claim 1, characterized in that: The prediction model is a combined structure integrating a support vector regression model SVR and a random forest model, the prediction model takes the accurate seedling weight value, the corrected irrigation amount and the historical survival rate data as input, and outputs the expected survival rate trend of the seedlings under different transplanting time.

7. The method according to claim 1, wherein the method is characterized by: The comprehensive report data package comprises environmental parameter data, dynamic weighing data, irrigation ratio parameters, transplanting time parameters and predicted survival rate curves, the data package is packaged in JSON format and uploaded to the agricultural Internet of Things platform interface through the MQTT protocol.

8. The method according to claim 1, wherein the method is characterized by: The temperature rise value is determined based on the difference between the current surface temperature of the sensor and the ambient dew point temperature, the required power is obtained by looking up the temperature rise-power mapping table, and the corresponding power is output by the heating module to achieve temperature regulation.

9. The method according to claim 1, wherein the method is characterized by: The standard weight curve is preset according to different seedling types, the seedling types are identified and classified by a visual identification module, the visual identification module includes an industrial camera and a seedling image database, and the corresponding weight curve template is automatically matched.

10. The method of claim 1, wherein the method is a dynamic weighing control method for transplanting long strip bare seedlings. The generation of the matching correction scheme is based on the current seedling weight value, the target growth period, the seedbed substrate type and the historical matching optimization data, and the adjustment instruction is generated by the central processing unit based on multi-parameter cross comparison.

Images