Search

CN-122017193-A - Intelligent detection method and system for vegetable and fruit planting environment parameters based on sensor

CN122017193ACN 122017193 ACN122017193 ACN 122017193ACN-122017193-A

Abstract

The application provides a vegetable and fruit planting environment parameter intelligent detection method and system based on a sensor, which relate to the technical field of intelligent agriculture, and enable soil to reach a preset saturated stable state by applying controllable physical intervention to a sensor monitoring area in a selected time window so as to acquire an original output signal of the sensor in the stable state; based on the quantized performance drift quantity, the mapping relation between the sensor output value and the real soil moisture physical quantity is reconstructed, and the real-time output value of the sensor is converted by adopting the reconstructed mapping relation in a non-intervention period, so that calibrated soil moisture data are obtained, and the problem of measurement data deviation caused by long-term operation, environmental change and self-aging of the sensor in the prior art is effectively solved.

Inventors

  • CHEN LIPING
  • REN SHU

Assignees

  • 浙江锦农农业开发有限公司

Dates

Publication Date
20260512
Application Date
20260129

Claims (10)

  1. 1. The intelligent detection method for the vegetable and fruit planting environment parameters based on the sensor is characterized by comprising the following steps of: Applying controllable physical intervention to a sensor monitoring area within a selected time window to enable soil to reach a preset saturated stable state; acquiring an original output signal of the sensor in the saturated stable state, and comparing the original output signal with a corresponding signal in an initial reference state in a cross-time dimension manner so as to quantify the performance drift amount of the sensor; reconstructing a mapping relation between the sensor output value and the real soil moisture physical quantity based on the performance drift quantity; And in the non-intervention period, converting the real-time output value of the sensor by adopting the reconstructed mapping relation to obtain calibrated soil moisture data.
  2. 2. The intelligent detection method for vegetable and fruit planting environment parameters based on sensors according to claim 1, wherein the controllable physical intervention comprises: Injecting supersaturated water into the monitored area to reach the saturated stable state of the soil, or Dividing the monitoring area into a core area and a buffer isolation area, injecting supersaturated water into the core area, and simultaneously applying ventilation and/or heating operation to the buffer isolation area to inhibit conduction interference of external environment variables to the core area.
  3. 3. The intelligent detection method for vegetable and fruit planting environment parameters based on a sensor according to claim 1, wherein the step of obtaining the original output signal of the sensor in the saturated stable state and comparing the original output signal with the corresponding signal in the initial reference state in a cross-time dimension manner to quantify the performance drift amount of the sensor comprises the following steps: acquiring an original output signal of the sensor in the saturated stable state; and calculating an algebraic difference between the original signal in the current saturation state and the reference signal in the initial base state as a performance drift amount.
  4. 4. The intelligent detection method for vegetable and fruit planting environment parameters based on a sensor according to claim 1, wherein the step of obtaining the original output signal of the sensor in the saturated stable state and comparing the original output signal with the corresponding signal in the initial reference state in a cross-time dimension manner to quantify the performance drift amount of the sensor comprises the following steps: Applying a preset electric excitation signal to the sensor in the saturated stable state; collecting and analyzing an output response signal of the sensor under the action of the electric excitation signal; extracting at least one characteristic parameter characterizing sensor performance from the output response signal; And comparing the currently extracted characteristic parameters with corresponding characteristic parameters acquired in an initial reference state, and calculating a difference quantity as the performance drift quantity.
  5. 5. The intelligent detection method for vegetable and fruit planting environment parameters based on the sensor according to claim 1, wherein the step of reconstructing the mapping relationship between the sensor output value and the real soil moisture physical quantity based on the performance drift amount comprises: Identifying soil micro-environmental characteristics of a monitoring area of the sensor; dividing the monitoring area into a plurality of microenvironment management units according to the soil microenvironment characteristics; Establishing a soil micro-environment file for each micro-environment management unit; And reconstructing a mapping relation between the sensor output value and the real soil moisture physical quantity according to the performance drift quantity and the soil microenvironment file of the microenvironment management unit which is currently self-calibrated.
  6. 6. The intelligent detection method for vegetable and fruit planting environment parameters based on sensors according to claim 5, wherein the step of establishing a soil micro-environment profile for each micro-environment management unit comprises: monitoring soil characteristic data of the microenvironment management unit in real time, wherein the soil characteristic data at least comprises soil conductivity, pH value and oxidation-reduction potential; triggering archival updating when detecting that the variation of the soil characteristic data relative to the archival record value exceeds a preset threshold value; Based on the current soil characteristic data, updating parameters related to the moisture adsorption capacity, the permeation rate and the oxygen exchange sensitivity in the soil micro-environment file.
  7. 7. The intelligent detection method for vegetable and fruit planting environment parameters based on the sensor according to claim 1, wherein the mapping relation between the reconstructed sensor output value and the real soil moisture physical quantity comprises any one of the following modes: the performance drift amount is used as a fixed offset to linearly correct the sensor output value, or And forming new data points by the sensor output value in the saturated stable state and the real soil moisture physical quantity, and re-fitting the mapping function by combining the historical data point set.
  8. 8. The intelligent detection method for vegetable and fruit planting environment parameters based on sensors according to claim 1, wherein the step of obtaining calibrated soil moisture data by converting sensor real-time output values using reconstructed mapping relations in a non-intervention period comprises: and in the non-intervention period, when the reconstructed mapping relation is adopted to convert the real-time output value of the sensor, selecting the corresponding mapping relation to convert the real-time output value according to the position of the micro-environment management unit.
  9. 9. The intelligent detection method for vegetable and fruit planting environment parameters based on the sensor according to claim 1, wherein the determining manner of the selected time window comprises: acquiring growth stage information of a current planted crop; And dynamically adjusting the trigger frequency of the selected time window according to the mapping relation between the preset crop growth stage and the calibration period, wherein the trigger frequency of the fruiting period is higher than that of the seedling period.
  10. 10. Sensor-based intelligent detection system for vegetable and fruit planting environment parameters, which is characterized by comprising: the physical intervention module is used for applying controllable physical intervention to the sensor monitoring area within a selected time window so as to enable the soil to reach a preset saturated stable state; the signal acquisition module is used for acquiring an original output signal of the sensor in the saturated stable state; the drift amount calculation module is used for comparing the original output signal with a corresponding signal in an initial reference state in a cross-time dimension mode so as to quantify the performance drift amount of the sensor; The mapping relation reconstruction module is used for reconstructing a mapping relation between the output value of the sensor and the real soil moisture physical quantity based on the performance drift quantity; And the real-time monitoring module is used for converting the real-time output value of the sensor by adopting the reconstructed mapping relation in a non-intervention period to obtain calibrated soil moisture data.

Description

Intelligent detection method and system for vegetable and fruit planting environment parameters based on sensor Technical Field The application relates to the technical field of intelligent agriculture, in particular to a vegetable and fruit planting environment parameter intelligent detection method and system based on a sensor. Background In modern agriculture, particularly in the planting of high-value vegetables and fruits, deployment of an intelligent detection system based on a sensor has become a key for ensuring accurate control of environmental parameters and ensuring high-quality output of crops. The system monitors key data such as air temperature and humidity, soil moisture, carbon dioxide concentration and the like in real time. The soil moisture sensor guides accurate irrigation by measuring the electrical characteristics of soil, and is a core component of the system. Organic fertilizers are commonly applied to farms for a long period of time in order to increase soil fertility and improve structure. This process promotes the activity of soil microorganisms. When decomposing organic matters, microorganisms slowly change the microenvironment of soil, and trace organic acids and other slightly corrosive byproducts are generated. These substances are not extremely toxic, but their long-term, sustained presence poses a hidden challenge for sensor metal electrodes (e.g., stainless steel or copper alloys) buried in the soil. In the period of months or years, the trace corrosive substance can react with the electrode surface slowly and electrochemically, and an extremely thin (nano-scale) passivation layer or oxide film is gradually formed. The film is difficult to detect by naked eyes, but can radically change the electrical characteristics between the electrode and soil, and the output signal of the sensor based on the capacitance or impedance principle can slowly drift systematically and nonlinearly. For example, when the actual soil moisture content stabilizes, the raw voltage value output by the sensor may continue to decrease slightly. The problem is that the calibration formula of the intelligent system is a static model built based on standard conditions when the sensor is newly installed. It cannot recognize and compensate for such complex drift caused by long-term physicochemical changes. Therefore, the system continuously receives the original signal with deviation, and after the original signal is converted by the old formula, the finally displayed moisture value is normal, and the actual value is gradually deviated. Because this drift is gradual and nonlinear, the resulting data bias tends to still fall within the "normal" range preset by the system (e.g., 20% -30%). For example, the actual moisture has fallen to the drought edge of 20%, the system may still show 25% and no alarm is triggered. The abnormality determination function of the system is thus silently disabled. This results in a typical "data normalization, plant abnormality" dilemma. When a manager makes a tour, the fuzzy symptoms of chronic water deficiency stress such as slow growth, leaf color loss and the like of crops can be observed, but when the system is consulted, all data show normal and no alarm. Because of the hard data of the more trusted system, the manager usually eliminates simple irrigation problems, which in turn are suspected to be more complex reasons such as plant diseases and insect pests, unbalanced nutrition and the like, so that time and labor consuming measures such as sample delivery, blind fertilization and pesticide application and the like are taken, and the simplest adjustment of irrigation is delayed. As a result, the physiological processes of photosynthesis, nutrient absorption, etc. continue to be impaired when crops are under undetected chronic water stress for a long period of time in the critical growth phase. This not only reduces crop resistance, but also eventually leads to reduced yields and poor fruit quality (e.g., low sugar, poor mouthfeel, poor storage) directly affecting economic benefits. Technical tools originally aimed at guaranteeing production are caused by misleading decisions and damaging production due to hidden faults in the tools. This dilemma highlights the importance of achieving long-term on-line self-diagnostics and adaptive calibration of sensors in a dynamically changing agricultural environment. Disclosure of Invention The application provides a sensor-based intelligent detection method and system for vegetable and fruit planting environment parameters, and aims to solve the technical problems that in long-term operation of an existing intelligent detection system, data deviation is caused by sensor performance drift, management decisions are misled, and crop growth and quality are affected. On one hand, the intelligent detection method for the vegetable and fruit planting environment parameters based on the sensor comprises the following steps: Applying control