Intelligent agricultural resource optimization scheduling system based on cloud computing

Abstract

The invention belongs to the technical field of resource management, and relates to a cloud computing-based intelligent agricultural resource optimization scheduling system, which comprises a biological signal acquisition module used for acquiring biological signals of crops in a target area; the environment data acquisition module is used for acquiring environment parameters outside crops; the dynamic demand analysis module is used for fusing the biological signals and the environmental parameters, generating dynamic demand indexes and judging whether a resource cooperation request is triggered or not; the block chain scheduling management module is used for receiving the resource cooperation request and calling surplus resources of the target farm to execute operation; the contribution point management module is used for generating verifiable contribution points and dynamically adjusting the exchange weight; and the intelligent verification module is used for comparing the deviation between the chain record and the local sensor data, triggering an automatic verification mechanism and correcting the account book record. The method can solve the problems that the contribution degree of participants lacks a credible quantitative standard, the benefit distribution contradiction inhibits the resource sharing willingness between farms.

Description

Technical Field

[0001] The present invention belongs to the technical field of resource management and relates to a cloud computing-based smart agricultural resource optimization and scheduling system. Background Art

[0002] Traditional agricultural resource scheduling suffers from inefficient cross-farm collaboration and a lack of dynamic demand response mechanisms. Information barriers between agricultural resource suppliers and demanders lead to regional imbalances in water and fertilizer supply and demand, insufficient real-time correlation between crop growth and external environmental changes, and resource allocation decisions that rely heavily on empirical predictions, making it difficult to accurately match crops' actual physiological needs. Traditional monitoring methods often rely on manual inspections or isolated environmental sensing devices, collecting data in a single dimension and lacking biological depth, making them unable to effectively capture the hidden needs of crops.

[0003] Existing solutions mostly focus on resource allocation within a single farm, triggering irrigation or fertilization operations based on pre-set thresholds. These solutions lack the ability to share resources and conduct dynamic negotiation among multiple entities. While some electronic management systems can visualize basic data, their resource allocation rules rely on static algorithmic models and are unable to adapt to environmental fluctuations and the differentiated needs of crop growth stages. Existing technologies lack effective market-based incentives for the coordinated scheduling of agricultural machinery and equipment, making cross-regional collaboration difficult to implement due to a lack of trust and disputes over the distribution of benefits.

[0004] In response to the above problems, this traditional approach has significant drawbacks: resource scheduling decisions are significantly delayed, emergency response is not timely enough, and the optimal intervention period is easily missed; there is a lack of credible quantitative standards for the contribution of participating entities, and conflicts in interest distribution inhibit the willingness of farms to share resources, ultimately affecting the achievement of large-scale resource optimization goals. Summary of the Invention

[0005] In order to solve the above problems, the present invention provides a cloud computing-based smart agricultural resource optimization and scheduling system.

[0006] A cloud computing-based smart agricultural resource optimization and scheduling system, comprising: A biological signal acquisition module is used to obtain biological signals of crops in the target area; Environmental data acquisition module, used to obtain environmental parameters outside the crop; Dynamic demand analysis module, which is used to integrate biological signals and environmental parameters to generate dynamic demand indicators and determine whether to trigger resource collaboration requests; The blockchain scheduling management module is used to receive resource collaboration requests and call on the target farm's surplus resources to perform operations; Contribution points management module, which is used to generate verifiable contribution points based on resource consumption, duration, and urgency, and dynamically adjust the redemption weight; The intelligent verification module is used to compare the deviation between the on-chain records and local sensor data, trigger the automatic verification mechanism and correct the ledger records.

[0007] Preferably, obtaining biological signals of crops in a target area includes the following steps: S11. Deployment of implantable sensors and acquisition of biosignals; The implantable sensor includes probes A, B, and C. Probe A measures the water flow velocity within the stem, probe B collects temperature changes of leaf epidermal cells, and probe C records the potential difference fluctuations at the root-soil interface. S12, multimodal biological signal fusion processing; The data collected by probes A, B, and C are fused and time-stamp aligned to serve as the input source for the dynamic demand analysis module. S13, encrypted transmission and edge storage; The data collected by probes A, B, and C are packaged into encrypted data frames. The data frames are end-to-end encrypted using the lightweight TEA encryption algorithm before transmission, and the encryption keys are dynamically allocated by the blockchain scheduling management module.

[0008] Preferably, obtaining the external environmental parameters of the crop includes the following steps: S21, distributed multi-sensor array deployment; Deploy sensor nodes, each containing a six-degree-of-freedom weather station and a soil profile probe. The main body of the weather station is integrated with a micro-lidar, and an omnidirectional wind pressure sensor is deployed on the top of the weather station. The six-degree-of-freedom weather station is an integrated module with the ability to detect three-dimensional wind speed, wind direction, temperature, humidity, air pressure, and light intensity. S22, space-time data alignment processing; Start the multi-parameter data fusion algorithm to satisfy the following formula:

[0009] in, represents the data fusion parameter, Indicates the The calibration weights of the sensors, Indicates the The raw measurement values ​​of the sensors, Indicates the clock deviation compensation of each node, Indicates the timestamp corresponding to the current data fusion processing.

[0010] Preferably, the dynamic demand analysis module includes a water demand coefficient calculation unit, a transpiration efficiency calculation unit, and a fertility gap assessment unit; The water demand coefficient calculation unit is used to normalize the stem flow rate to obtain the water demand coefficient; The transpiration efficiency calculation unit is used to calculate the transpiration efficiency index according to the difference between the leaf surface temperature and the ambient temperature; The fertility gap assessment unit is used to combine the root potential with a preset threshold value to generate a fertility gap level.

[0011] Preferably, combining the root potential with a preset threshold to generate a fertility gap level includes the following steps: The real-time root potential is compared with the threshold value by the difference percentage quantification method, satisfying the following formula:

[0012] in, Indicates the implementation difference, Indicates the real-time measured root potential value. Indicates the current nutrient absorption threshold, represents the basal potential threshold; Combined with historical data, the laboratory divides fertility gap into four levels: normal range, mild deficiency, moderate deficiency, and severe deficiency; the first threshold of fertility gap is pre-set , the first threshold , the first threshold The hierarchical logic is constructed using a multi-interval partitioning strategy and a progressive response mechanism to avoid resource waste caused by a single threshold trigger. like It indicates that the fertility gap level is within the normal range. Indicates that the fertility gap level is in a mild deficiency. Indicates that the fertility gap level is moderately deficient. It indicates that the fertility gap level is in severe deficiency.

[0013] Preferably, the dynamic demand analysis module further includes a dynamic indicator generation unit and a trigger condition detection unit; The dynamic index generating unit performs weighted summation on the water demand coefficient, the transpiration efficiency index, and the fertility gap level to generate a dynamic demand index; The trigger condition detection unit inputs the dynamic demand indicator into the trigger condition detection unit, and pre-sets a threshold for determining the trigger to determine whether to trigger a resource collaboration request.

[0014] Preferably, the blockchain scheduling management module includes a smart contract generation unit and a resource matching and screening unit; A smart contract generation unit, used to write the geographic location, resource type, and verification hash value of the resource request into the smart contract; Resource matching screening unit, used to generate a matching priority list based on farm location, resource surplus, and contribution point history.

[0015] Preferably, the contribution points management module includes a basic points calculation unit, an emergency weighted correction unit, and a points redemption interface; The basic score calculation unit is used to generate basic points according to the product of the resource type base value and the usage time; The emergency weighted correction unit is used to adjust the integral weight according to the amplitude and time attenuation coefficient of the dynamic demand index exceeding the threshold; The points redemption interface is used to convert contribution points into agricultural machinery leasing rights or agricultural input discount certificates.

[0016] Preferably, adjusting the integral weight according to the magnitude of the dynamic demand indicator exceeding the threshold and the time attenuation coefficient includes the following steps: The basic integral value enters the emergency weighted correction unit for dynamic adjustment. The unit has a built-in double-layer adjustment mechanism of emergency adjustment coefficient and time decay coefficient. The emergency adjustment coefficient is calculated based on the percentage of the dynamic demand index exceeding the preset threshold, and satisfies the following formula:

[0017] in, It represents the urgency adjustment coefficient, which reflects the dynamic multiplier of the suddenness of resource demand and is positively correlated with the deviation value of the demand index; Indicates the current dynamic demand index value; Indicates the threshold value of the current dynamic demand indicator; The time decay coefficient is calculated based on the interval between the resource request initiation time and the current time, using an exponential decay mode, and satisfies the following formula:

[0018] in, Represents the time decay coefficient, which is a compensation factor used to measure the timeliness of response. The longer the delay, the smaller the correction force. Indicates the decay rate set by the system, Indicates the time difference after the request is initiated; The final corrected integral value satisfies the following formula:

[0019] in, Indicates the basic integral value, Indicates the final corrected integral value.

[0020] Preferably, the intelligent verification module includes a verification trigger unit, a data comparison unit, and a dispute handling unit; The verification trigger unit is used to detect whether the difference in hash value between the resource call record and the local sensor data exceeds a threshold; The data comparison unit extracts the equipment logs and sensor storage data of the target farm; The dispute handling unit freezes the dispute points and initiates a manual review process, and the verified results are updated to the blockchain ledger.

[0021] In summary, the present invention has the following beneficial technical effects: 1. Through the fusion analysis of multi-dimensional biological signals and dynamic environmental parameters, we can break through the technical bottlenecks of traditional agricultural monitoring, which relies heavily on experience and lacks the ability to identify implicit needs. We can build a quantitative assessment system for the water and fertilizer needs of crops throughout their life cycle, achieve precise triggering of key operations such as irrigation and fertilization, and optimally match them with resource types, effectively reducing resource abuse and production waste. 2. Build a distributed dispatching network based on blockchain technology, breaking down information silos for cross-party resource sharing. Automatically execute resource collaboration agreements and dynamically calculate points through smart contracts, eliminating the time delays and trust friction associated with manual negotiations. Rapidly dispatch remote agricultural machinery resources in unexpected demand scenarios, significantly shortening disaster response cycles and improving collaborative network stability. 3. The innovative design of the verifiable contribution points system and multiple data verification mechanism binds resource consumption to crop physiological indicators through hashing, and uses on-chain and off-chain cross-validation to ensure the fairness and security of point distribution, stimulate the enthusiasm of large-scale farms to participate in resource sharing, and form a sustainable agricultural resource value exchange ecosystem. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. The drawings are used to provide a further understanding of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 This paper discloses a framework diagram of a smart agricultural resource optimization and scheduling system based on cloud computing.

[0024] Figure 2 This paper discloses a structural diagram of a smart agricultural resource optimization and scheduling system based on cloud computing. DETAILED DESCRIPTION

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] The following is combined with Figure 1-Figure 2 The preferred embodiments of the present invention are described in detail.

[0027] See attached Figure 1-Figure 2 The present invention proposes a cloud computing-based smart agricultural resource optimization and scheduling system, which includes the following modules: A biological signal acquisition module is used to obtain biological signals of crops in the target area; Environmental data acquisition module, used to obtain environmental parameters outside the crop; Dynamic demand analysis module, which is used to integrate biological signals and environmental parameters to generate dynamic demand indicators and determine whether to trigger resource collaboration requests; The blockchain scheduling management module is used to receive resource collaboration requests and call on the target farm's surplus resources to perform operations; Contribution points management module, which is used to generate verifiable contribution points based on resource consumption, duration, and urgency, and dynamically adjust the redemption weight; The intelligent verification module is used to compare the deviation between the on-chain records and local sensor data, trigger the automatic verification mechanism and correct the ledger records.

[0028] Among them, the biological signal acquisition module and the environmental data acquisition module are both connected to the dynamic demand analysis module; the dynamic demand analysis module is connected to the blockchain scheduling management module; the blockchain scheduling management module is respectively connected to the contribution points management module and the intelligent verification module.

[0029] In one embodiment of the present invention, obtaining biological signals of crops in a target area includes the following steps: S11. Deployment of implantable sensors and acquisition of biosignals; Specifically, a minimally invasive channel with a diameter of 0.5 mm is opened at the junction of the crop root and stem, and a three-probe implantable sensor is embedded in the junction of the crop's xylem and phloem. The three-probe implantable sensor includes probe A, probe B, and probe C. The probe surfaces are coated with a biocompatible nanocoating. Probe A integrates a microfluidic thermal unit, probe B is equipped with a high-precision infrared temperature measurement unit, and probe C is equipped with an anti-polarization microelectrode. After the three-probe implantable sensor is turned on, probe A continuously measures the water flow velocity within the stem using the principle of thermal diffusion, probe B collects temperature changes of leaf epidermal cells at a frequency of 10 times per second, and probe C records the potential difference fluctuations at the root-soil interface. The data collected by probes A, B, and C in the three-probe implantable sensor are packaged into encrypted data frames every 5 minutes.

[0030] Among them, the three-probe implantable sensor is a miniature biological signal monitoring device manufactured using MEMS technology. It contains three types of sensing units: thermal, infrared, and electrochemical. It achieves non-destructive coupling with plant tissues through a flexible substrate. The water flow velocity in the stem is measured using the stem flow rate based on the principle of thermal diffusion. The water flow dynamics parameters are calculated by detecting the upstream and downstream temperature gradient changes after emitting pulses from a heat source. The potential difference fluctuation is the root potential. The transmembrane potential difference generated during the ion exchange process of the plant roots reflects the crop's absorption activity of soil nutrients and the metabolic intensity of the root system.

[0031] For example, in an experiment during the tillering period of wheat, probe A in the three-probe implantable sensor monitored that the stem flow rate reached a peak of 1.2 mm / s at 12 noon. At the same time, probe C detected that the root potential dropped by 120 mV within 3 hours after fertilization, which is consistent with the biological laws of enhanced crop transpiration and increased nitrogen absorption rate.

[0032] S12, multimodal biological signal fusion processing; Specifically, a sliding window differential filter is used on the stem flow rate data to eliminate environmental vibration interference. A compensation algorithm is used to correct the measurement deviation caused by wind speed for the leaf temperature data. A Kalman filter is used to remove the soil conductivity fluctuation noise from the root potential data. The three types of processed biological signals are aligned with their timestamps and serve as the input source for the dynamic demand analysis module. Among them, the sliding window differential filter is a dynamic denoising method based on a time window, which compares the data change rate of adjacent time periods to identify abnormal fluctuations; the compensation algorithm introduces the readings of the ambient wind speed sensor to dynamically correct the leaf surface temperature measurement value, eliminating the influence of the wind cooling effect on infrared temperature measurement; the Kalman filter estimates the true value of the root potential through a state-space model, minimizing the measurement error of transient soil ion concentration.

[0033] S13, encrypted transmission and edge storage; Specifically, the data collected by probes A, B, and C are packaged into encrypted data frames every 5 minutes. The data frames are end-to-end encrypted using the lightweight TEA encryption algorithm before transmission, and the encryption key is dynamically allocated by the blockchain scheduling management module. After receiving the data, the edge node constructs a structured storage entry containing the data hash value, timestamp, and device ID, and uploads a copy of the data to the IPFS distributed storage network to generate an unalterable fingerprint index. Among them, the TEA encryption algorithm, a symmetric key block cipher scheme, uses a 128-bit key and a 64-bit data block structure, taking into account the computing power of agricultural IoT devices and the data security requirements; the fingerprint index, a unique identifier based on content addressing, ensures the integrity verification capability of biological signal data through the SHA-256 hash algorithm.

[0034] The beneficial effects of the biosignal acquisition module: Through multi-dimensional biosignal collaborative sensing technology, it breaks through the limitations of traditional agricultural monitoring that relies solely on environmental parameters, achieving comprehensive analysis from the internal physiological mechanisms of plants to the effects of the external environment. The minimally invasive design and anti-interference processing of the implantable sensor ensure the feasibility of long-term monitoring, and the pre-processing and encryption architecture of the edge computing node ensures data credibility. Accurately capture the coupling relationship between crop water transport efficiency, nutrient absorption dynamics, and photosynthesis intensity, providing physiological-level data support for subsequent intelligent decision-making, and significantly improving the accuracy of drought warnings and the targeting of fertilization and irrigation.

[0035] In one embodiment of the present invention, obtaining environmental parameters outside the crop includes the following steps: S21, distributed multi-sensor array deployment; Specifically, sensor nodes are arranged in a 30m×30m grid in the farmland environment. Each node contains a six-degree-of-freedom weather station and a soil profile probe. The weather station body is integrated with a micro-lidar, and an omnidirectional wind pressure sensor is deployed on the top of the weather station. The probe depth is layered and set to 0-15 cm shallow measurement points, 15-40 cm mid-layer measurement points, and 40-80 cm deep measurement points. The shallow measurement points are equipped with dielectric constant soil moisture sensors, the mid-layer measurement points are equipped with nitrate ion selective electrodes, and the deep measurement points are equipped with redox potential composite probes. The six-degree-of-freedom weather station is an integrated module capable of detecting three-dimensional wind speed, wind direction, temperature, humidity, air pressure, and light intensity. The internal gyroscope corrects measurement errors caused by physical displacement in real time. The dielectric constant soil moisture sensor is a measuring device based on the frequency domain reflectometry method. It calculates the moisture content by detecting the propagation delay of electromagnetic waves in the soil medium. The nitrate ion selective electrode directly detects the changes in nitrate ion concentration in the middle layer of the soil and quantifies the effective nitrogen fertilizer content available for crop absorption in real time through electrochemical methods. The redox potential composite probe is a dual-channel detection unit that integrates molybdenum electrodes and platinum electrodes, which synchronously measures the electron exchange capacity of oxidized and reduced substances in the soil.

[0036] For example, in a corn field test in the black soil area, the nitrate concentration detected at the middle measuring point dropped to 120 mg / kg on the 7th day after fertilization, and the deep redox potential rose to 380 mV during the same period, which is consistent with the growth stage characteristics of corn roots extending downward and accelerating nutrient absorption.

[0037] S22, space-time data alignment processing; Specifically, start the multi-parameter data fusion algorithm to satisfy the following formula:

[0038] in, represents the data fusion parameter, Indicates the The calibration weights of the sensors, Indicates the The raw measurement values ​​of the sensors, Indicates the clock deviation compensation of each node, Indicates the timestamp corresponding to the current data fusion processing; After the calculation, the discrete data is reconstructed into a three-dimensional field distribution map with an accuracy of 0.1 meters. The soil profile data is superimposed with the near-ground meteorological data to generate a microenvironment dynamic model, eliminating the noise peaks caused by sudden electromagnetic interference; Among them, the multi-parameter data fusion algorithm dynamically allocates calibration weights according to the historical error rate of the sensor to eliminate the influence of individual differences in equipment; the three-dimensional field distribution map, based on the continuous spatial parameter model constructed by the Kriging interpolation method, calculates the spatial correlation between each measuring point.

[0039] The beneficial effects of the environmental data acquisition module: the use of geospatial coupling and multi-source heterogeneous data fusion technology to achieve full-area three-dimensional perception of the farmland microenvironment, the distributed sensor array breaks through the blind spot limitations of traditional point monitoring, the multi-level soil profile probe reveals the laws of nutrient migration, and the dynamic weighted fusion algorithm filters out equipment errors and electromagnetic interference.

[0040] In one embodiment of the present invention, the dynamic demand analysis module includes a water demand coefficient calculation unit, a transpiration efficiency calculation unit, a fertility gap assessment unit, a dynamic index generation unit, and a trigger condition detection unit; The water demand coefficient calculation unit is used to normalize the stem flow rate to obtain the water demand coefficient; Specifically, it receives data on stem flow rate, leaf temperature, and root potential from the biological signal acquisition module and data on temperature, humidity, and light intensity from the environmental data acquisition module, and eliminates the comparability barriers of parameters of different dimensions through feature scaling; Among them, the stem flow rate is converted into a water demand coefficient in the range of 0-1 through the range method; the range method is a method of linear mapping based on the maximum and minimum values ​​of historical data.

[0041] The normalized formula for stem flow rate satisfies the following formula:

[0042] in, represents the water demand coefficient, represents the original stem flow rate value, Indicates the minimum stem flow rate of the crop during its current growth period. Indicates the theoretical threshold value of the corresponding growth period.

[0043] The transpiration efficiency calculation unit is used to calculate the transpiration efficiency index according to the difference between the leaf surface temperature and the ambient temperature; Specifically, the difference between leaf surface and ambient temperature is linearly mapped to generate the transpiration efficiency index, and the normalized leaf surface-ambient temperature difference is integrated with the light intensity to establish a dynamic weight distribution model. , each increase in light intensity , at least one of the conditions is met, triggering an iteration of the weight parameters of the dynamic weight allocation model; Among them, the dynamic weight allocation model is an algorithm that dynamically adjusts the calculation weights according to environmental parameters; Calculate the transpiration efficiency index. The light intensity is calculated using a piecewise function. When the sun is directly shining, the light intensity saturation correction coefficient is enabled to meet the following formula:

[0044] in, is the transpiration efficiency index; is the leaf-ambient temperature difference; represents the light response regulator; Indicates the current light intensity; It is the optimal light intensity threshold for crop photosynthesis and the reference value of light intensity when the photosynthetic efficiency of different crops reaches its peak.

[0045] Among them, the light saturation correction coefficient is a nonlinear compensation parameter introduced to reduce data distortion under strong light conditions; the light intensity uses sliding time window filtering to remove sudden noise caused by short-term cloud cover; the sliding time window filtering is a dynamic filtering method that retains a fixed-length data window and replaces instantaneous outliers with the mean of the data within the window.

[0046] The fertility gap assessment unit is used to combine the root potential with a preset threshold to generate a fertility gap level; Specifically, two sets of key thresholds are pre-set based on crop type and growth stage: the nutrient absorption threshold and the basal potential threshold. The nutrient absorption threshold is the minimum potential value required for normal root metabolism during the current growth period, which is calibrated based on agronomic experiments; the basal potential threshold is the extreme potential value when the root system is dormant or completely lacks nutrients, which is obtained through long-term observation data statistics. The real-time root potential is compared with the threshold value by the difference percentage quantification method, satisfying the following formula:

[0047] in, Indicates the implementation difference, Indicates the real-time measured root potential value. Indicates the current nutrient absorption threshold, represents the basal potential threshold; The value range is constrained to be in the range of 0-1. When directly judging , no fertilizer intervention is required.

[0048] Among them, the difference percentage quantification method converts the difference between the measured value and the theoretical threshold into a percentage scale, which intuitively reflects the relative severity of the nutrient gap.

[0049] Combined with historical data, the laboratory divides fertility gap into four levels: normal range, mild deficiency, moderate deficiency, and severe deficiency; the first threshold of fertility gap is pre-set , the first threshold , the first threshold The hierarchical logic is constructed using a multi-interval partitioning strategy and a progressive response mechanism to avoid resource waste caused by a single threshold trigger. like It indicates that the fertility gap level is within the normal range. Indicates that the fertility gap level is in a mild deficiency. Indicates that the fertility gap level is moderately deficient. It indicates that the fertility gap level is in severe deficiency.

[0050] Exemplary, pre-set first threshold value of fertility gap coefficient , the first threshold , the first threshold ; if , indicating that the fertility gap level is within the normal range, the root potential of the crop is in a sufficient state, and the nutrient supply meets the metabolic demand; if , indicating that the fertility gap level is at a mild level of deficiency, the root absorption efficiency of crops has decreased, and subsequent trends need to be monitored but no immediate response is required; if , indicating that the fertility gap level is at a moderate level, the crop's tissue metabolic rate is inhibited, and a low-priority fertilization recommendation is triggered; if , indicating that the fertility gap level is in a severe deficiency, the root cell membrane potential of the crop is unbalanced, and an emergency fertilization instruction is immediately generated.

[0051] The preset threshold is dynamically corrected in combination with environmental parameters. The dynamic threshold correction integrates environmental parameters and biological characteristics to make the preset threshold setting consistent with actual field conditions.

[0052] For example, when the soil moisture content is lower than 40%, Automatically increase by 5% to avoid false fertilization errors caused by water stress; adjust according to the difference between leaf temperature and soil temperature , the temperature difference increases by 1 , A decrease of 2 millivolts reflects an increase in the activity of crop root metabolism.

[0053] The dynamic index generating unit performs weighted summation on the water demand coefficient, the transpiration efficiency index, and the fertility gap level to generate a dynamic demand index; Water demand coefficient , transpiration efficiency index , implement the difference G input dynamic index generation unit, use the adaptive weight distribution algorithm for fusion, and satisfy the following formula,

[0054] in, It is a dynamic demand indicator; is the baseline transpiration efficiency value; 、 、 Represents a weight coefficient that dynamically adjusts based on the crop's growth stage. For example, the default values ​​for the vegetative growth phase are 0.5, 0.3, and 0.2, while those for the reproductive phase are adjusted to 0.4, 0.4, and 0.2. The adaptive weight allocation algorithm is a fusion mechanism that automatically adjusts the importance of each parameter based on the crop's phenological phase, ensuring targeted demand analysis at different growth stages. The dynamic demand index is a normalized, comprehensive decision-making parameter with a value range of 0-1, with higher values ​​indicating greater urgent demand for the crop.

[0055] Exemplary, during the tasseling period of corn, multiple laboratory tests were performed to determine , , , , using the reproductive growth period weight ( 、 、 2) Calculation shows that,

[0056] The trigger condition detection unit, dynamic demand index An input trigger condition detection unit pre-sets a trigger threshold to determine whether to trigger a resource collaboration request; Specifically, two consecutive cycle thresholds are pre-set, which are 0.5 and 0.7; when When the value exceeds 0.5 for two consecutive cycles, the inspection warning is activated and an inspection suggestion is sent to the management personnel; When the value exceeds 0.7 for three consecutive cycles or a single value exceeds 0.85, the next blockchain scheduling management module will be directly triggered to initiate an emergency response; Among them, the threshold of continuous cycles, the protection mechanism to avoid false triggering due to instantaneous interference, verifies the authenticity of the demand through time continuity; initiates emergency response, and the cross-module collaborative working mechanism converts the analysis results into executable production instructions.

[0057] For example, during the cotton flowering and boll period, the crops are For three consecutive hours, it reached 0.72, 0.75, and 0.78, directly triggering the blockchain scheduling management module to initiate an emergency response: automatically triggering emergency irrigation instructions and executing drought relief measures ahead of manual decision-making.

[0058] The beneficial effects of this verification case are: it can accurately identify hidden water stress and nutrient imbalance of crops, realize early warning and graded control of water and fertilizer requirements, significantly improve water and fertilizer utilization efficiency and reduce the incidence of physiological diseases.

[0059] In one embodiment of the present invention, the blockchain scheduling management module includes a smart contract generation unit and a resource matching and screening unit; A smart contract generation unit, used to write the geographic location, resource type, and verification hash value of the resource request into the smart contract; Specifically, a pre-set irrigation, fertilization, or early warning contract template is loaded and filled with parameters such as dynamic demand indicator values, farm geographic coordinates, and execution time limits to generate a programmable contract script. The weight formula embedded in the contract script dynamically adjusts the irrigation or fertilization amount. Among them, smart contract generation is a technology that automatically creates blockchain executable code through parameterized templates, converting dynamic demand indicators into verifiable trigger instructions on the chain.

[0060] Resource matching screening unit, used to generate a matching priority list based on farm location, resource surplus, and contribution points history; Specifically, after the contract takes effect, the resource matching and screening unit calls the IoT device registry to match the irrigation equipment and fertilizer equipment numbers of the target plot. The resource matching and screening unit generates a scheduling queue based on the real-time status of the equipment and the task priority. For example, the near-end device in the standby state is given priority, satisfying the following formula:

[0061] in, Indicates resource matching degree, Indicates the matching coefficient between resource type and demand, Indicates the historical usage attenuation factor of the device. Indicates the spatial distance loss coefficient between farmland and equipment, 、 、 Represents the dynamic weight parameter of the output.

[0062] For example, in a corn field sprinkler irrigation task, the resource matching screening unit screens mobile irrigation vehicles that are 2.5 kilometers away from the target farmland. reached 0.92, The final calculated resource matching degree exceeds the threshold of 0.8, triggering the device scheduling instruction.

[0063] The generated smart contract must be collaboratively verified by a preset node group using an improved PBFT consensus mechanism. The verification node group includes the owner node of the farmer, the verification nodes of adjacent plots, and the regulatory agency node. Each node independently verifies the legitimacy of the contract's demand indicator hash value and geographic coordinates. Once consensus is passed, the smart contract is encapsulated into a transaction block with a timestamp and broadcast to all nodes in the agricultural blockchain network. Among them, the improved PBFT consensus optimizes the Byzantine fault tolerance algorithm and selects high-reliability nodes to participate in verification through a dynamic credit scoring mechanism; The demand indicator hash value is an irreversible encrypted summary of the dynamic demand analysis module output data using the SM3 hash algorithm, which is used for on-chain data integrity verification; The geographic coordinate legitimacy verification is to verify whether the latitude and longitude of the target plot are within the registered farm boundary through GIS map fence technology.

[0064] For example, an irrigation truck initiates a precision irrigation program based on a smart contract, and a pressure sensor collects water flow feedback in real time. This feedback data is encrypted by the TEE trusted execution environment and then generated into an on-chain data packet containing execution time, resource consumption, and operation digital fingerprints, which are then written into the blockchain immutable ledger in the blockchain scheduling management module. The TEE (Trusted Execution Environment) is an isolated computing area based on hardware encryption, ensuring that the integrity of the executed data is verified in an encrypted state. The immutable ledger uses a block hash chain structure, where each new block contains the hash value of the previous block, forming a tamper-proof data chain.

[0065] In one embodiment of the present invention, the contribution points management module includes a basic points calculation unit, an emergency weighted correction unit, and a points redemption interface; The basic score calculation unit is used to generate basic points according to the product of the resource type base value and the usage time; Specifically, the contribution points management module activates the basic points calculation unit, generating initial points based on the equipment's participation time, resource consumption, and task urgency. Dynamic adjustment factors are introduced into the points calculation formula, such as a 20% increase in the weight of drought emergency tasks. Points data is written to the blockchain ledger after multi-node verification and synchronized to the cooperative farm's points pool via a cross-chain protocol, allowing points to be redeemed for irrigation quotas or agricultural machinery rental services. Points are automatically settled based on the execution results of the smart contract. The scoring rules satisfy the following formula:

[0066] in, is the basic integral value, is the resource type base value weight, It is the actual usage time of the resources called; the result of the points settlement is updated to the farm account through the side chain channel.

[0067] For example, when the system calls a farm's irrigation equipment to provide water resources assistance, if the current water resource base value Set to 10 points / minute, the actual resource usage time Generate points for 30 minutes ; Among them, the resource type base value represents the quantitative standard set according to the contribution value of different agricultural resources, including the water resource coefficient (the value of water consumption per unit time), the fertilizer coefficient (the conversion coefficient of the associated nitrogen, phosphorus and potassium content), and the equipment sharing coefficient (dynamically weighted according to the equipment depreciation rate).

[0068] The emergency weighted correction unit is used to adjust the integral weight according to the amplitude and time attenuation coefficient of the dynamic demand index exceeding the threshold; Specifically, the basic integral value enters the emergency weighted correction unit for dynamic adjustment. The unit has a built-in double-layer adjustment mechanism of emergency adjustment coefficient and time attenuation coefficient. The emergency adjustment coefficient is calculated based on the percentage of the dynamic demand index exceeding the preset threshold, and satisfies the following formula:

[0069] in, It represents the urgency adjustment coefficient, which reflects the dynamic multiplier of the suddenness of resource demand and is positively correlated with the deviation value of the demand index; Indicates the current dynamic demand index value; Indicates the threshold value of the current dynamic demand indicator; The time decay coefficient is calculated based on the interval between the resource request initiation time and the current time, using an exponential decay mode, and satisfies the following formula:

[0070] in, Represents the time decay coefficient, which is a compensation factor used to measure the timeliness of response. The longer the delay, the smaller the correction force. Indicates the decay rate set by the system, Indicates the time difference after the request is initiated; The final corrected integral value satisfies the following formula:

[0071] For example, if the dynamic demand index in a resource request exceeds the threshold by 25%, the system-set decay rate , the scheduling is completed 2 hours after initiation; Calculation shows that, , After the final revision, the original basic score of 300 points .

[0072] The points redemption interface is used to convert contribution points into agricultural machinery leasing rights or agricultural input discount certificates; Specifically, the final revised contribution points are written into the blockchain account through the smart contract, activating the points redemption interface. The redemption process uses a two-factor authentication mechanism: 1. Convert points into agricultural machinery usage time or agricultural input discount coupons according to a preset ratio. For example, 500 points can be exchanged for 1 hour of harvester usage or a 5% discount on fertilizer purchases. 2. The exchange result is written into the blockchain smart contract, which automatically generates an encrypted digital token with a timestamp. Only the private key held by the target farm can unlock and call it. The encrypted digital token contains an encrypted data structure of the device identifier, service type, and validity period. It is also encrypted using the elliptic curve digital signature algorithm to ensure that only the target device can parse and execute it. 3. When calling equipment from other farms, the system automatically deducts the corresponding points and updates the account balances of both parties on the chain.

[0073] For example, after Farm A accumulates 1,200 contribution points, it applies to redeem the drone spraying service of the neighboring Farm B (unit price 200 points / hour). The system automatically generates a 6-hour service token with a validity period of 48 hours. Farm B's equipment reads the token and performs the service. At the same time, the blockchain deducts 1,200 points from Farm A and adds 1,200 points to Farm B.

[0074] This demonstration demonstrates the following benefits: When faced with sudden droughts, neighboring farms are more willing to respond quickly to high-point irrigation collaboration requests. During seasonal fertilization, dynamic adjustments to points attract more farms to participate in fertility support. This improves agricultural resource mismatch, enhances the stability of the collaborative network, and improves emergency response efficiency. The point redemption mechanism also effectively reduces the manual negotiation costs associated with traditional resource leasing.

[0075] In one embodiment of the present invention, the intelligent verification module includes a verification trigger unit, a data comparison unit, and a dispute handling unit; The intelligent verification module establishes a cross-verification mechanism for on-chain and off-chain data, using asymmetric encryption technology to ensure the non-repudiation of the verification process. By comparing the device operation log, sensor raw data, and blockchain ledger records, it identifies abnormal data flows that are forged or tampered with, and reconstructs the trusted data chain based on the majority consensus principle. The closed-loop feedback loop of verification triggering, arbitration, and correction ensures the ultimate consistency of system operation. The verification trigger unit is used to detect whether the difference between the hash value of the resource call record and the local sensor data exceeds 5%; Specifically, when the hash value of the crop stalk flow rate recorded in the blockchain account book is less than 95% similar to the hash value of the original data stored by the local sensor, the verification trigger unit sends a verification request to the blockchain scheduling management module. Hash similarity represents an indicator of the similarity of the feature code generated by the hash algorithm after two data sets are processed. It is used to quantify the consistency between the on-chain record and the locally stored data and satisfies the following formula:

[0076] The "number of matching bytes" is the number of bytes with the same value at the same position in the two hash value binary sequences, and the "total number of bytes" is the total length of the hash value.

[0077] For example, the total bytes of the SHA-256 hash value are 32. If the comparison finds that 3 bytes are inconsistent, the similarity is (29 / 32)×100%≈90.6%, triggering the verification mechanism.

[0078] The data comparison unit extracts the equipment logs and sensor storage data of the target farm; Specifically, the data comparison unit retrieves three independent data sources stored on the target farm: raw sensor data files, blockchain ledger records, and equipment runtime logs. The acquisition frequency bands of the three data are strictly synchronized, and slice alignment is performed using a 15-second time window. Among them, the device operation timing log is a time-stamped machine operation record generated by the IoT execution module, including the water pump start and stop time, pressure value, and flow data.

[0079] The dispute handling unit freezes the dispute points and initiates a manual review process, and the review results are updated to the blockchain ledger; Specifically, the dispute resolution unit arranges the sliced ​​data in chronological order and compares the specific values ​​of blockchain record A, sensor data B, and device log C. If the on-chain record A matches the device log C but the sensor data B does not, the consensus result of A and C will be prioritized, and the points related to data B for that time period will be automatically frozen. This is the "three-source, two-identity verification"; When the verification confirms that there is a deviation in the blockchain record, the system generates a new smart contract patch file, which is verified and signed by the farmer's private key. The combined public key combination preset by the system forms a hybrid signature, which is "double-signature authorization". "Dual-signature authorization" is a data modification agreement signed by the farm manager's private key and the system supervisory public key. The corrected ledger record is stored in the form of a newly added blockchain node, and the hash index of the original erroneous data is retained for audit backtracking.

[0080] For example, a farm's blockchain ledger records an irrigation operation at 9:00 AM on May 12, 2024. However, the local soil moisture sensor shows no change in water level during that period. The system automatically retrieves the device log and discovers that the irrigation pump did not actually start, thus determining an error in the blockchain record. Three-source, two-identity verification confirms the operation signal was erroneously triggered, freezes the points for the corresponding period, and updates the ledger. The corrected points record is automatically synchronized to the contribution points management module, preventing subsequent resource redemption disputes.

[0081] The beneficial effects of this verification example: multi-source data verification improves the accuracy of anomaly detection and effectively identifies scenarios such as sensor failure and data tampering; the layered arbitration mechanism reduces the risk of misjudgment and avoids single-point data errors causing large-scale system jitter; on-chain corrections retain complete operation traces and achieve transparent traceability of agricultural resource scheduling; the dual signature mechanism balances credibility and execution efficiency, maintaining real-time system response while ensuring data authenticity.

[0082] The interface definitions and communication protocols between modules clarify the interface specifications between modules, including: interface name, parameter type, return value, etc., and formulate corresponding communication protocols to ensure that data and information can be transmitted correctly and efficiently between modules.

[0083] The above embodiments are only used to illustrate the technical solutions of the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A cloud computing-based smart agricultural resource optimization and scheduling system, characterized in that: include: A biological signal acquisition module is used to obtain biological signals of crops in the target area; Environmental data acquisition module, used to obtain environmental parameters outside the crop; Dynamic demand analysis module, which is used to integrate biological signals and environmental parameters to generate dynamic demand indicators and determine whether to trigger resource collaboration requests; The blockchain scheduling management module is used to receive resource collaboration requests and call on the target farm's surplus resources to perform operations; Contribution points management module, which is used to generate verifiable contribution points based on resource consumption, duration, and urgency, and dynamically adjust the redemption weight; The intelligent verification module is used to compare the deviation between the on-chain records and local sensor data, trigger the automatic verification mechanism and correct the ledger records.

2. A cloud computing-based smart agricultural resource optimization and scheduling system according to claim 1, characterized in that: Acquiring biological signals of crops in the target area includes the following steps: S11. Deployment of implantable sensors and acquisition of biosignals; The implantable sensor includes probes A, B, and C. Probe A measures the water flow velocity within the stem, probe B collects temperature changes of leaf epidermal cells, and probe C records the potential difference fluctuations at the root-soil interface. S12, multimodal biological signal fusion processing; The data collected by probes A, B, and C are fused and time-stamp aligned to serve as the input source for the dynamic demand analysis module. S13, encrypted transmission and edge storage; The data collected by probes A, B, and C are packaged into encrypted data frames. The data frames are end-to-end encrypted using the lightweight TEA encryption algorithm before transmission, and the encryption keys are dynamically allocated by the blockchain scheduling management module.

3. A cloud computing-based smart agricultural resource optimization and scheduling system according to claim 2, characterized in that: Obtaining the external environmental parameters of the crop includes the following steps: S21, distributed multi-sensor array deployment; Deploy sensor nodes, each containing a six-degree-of-freedom weather station and a soil profile probe. The main body of the weather station is integrated with a micro-lidar, and an omnidirectional wind pressure sensor is deployed on the top of the weather station. The six-degree-of-freedom weather station is an integrated module with the ability to detect three-dimensional wind speed, wind direction, temperature, humidity, air pressure, and light intensity. S22, space-time data alignment processing; Start the multi-parameter data fusion algorithm to satisfy the following formula: in, represents the data fusion parameter, Indicates the The calibration weights of the sensors, Indicates the The raw measurement values ​​of the sensors, Indicates the clock deviation compensation of each node, Indicates the timestamp corresponding to the current data fusion processing.

4. A cloud computing-based smart agricultural resource optimization and scheduling system according to claim 3, characterized in that: Dynamic demand analysis module, including water demand coefficient calculation unit, transpiration efficiency calculation unit, and fertility gap assessment unit; The water demand coefficient calculation unit is used to normalize the stem flow rate to obtain the water demand coefficient; The transpiration efficiency calculation unit is used to calculate the transpiration efficiency index according to the difference between the leaf surface temperature and the ambient temperature; The fertility gap assessment unit is used to combine the root potential with a preset threshold value to generate a fertility gap level.

5. The cloud computing-based smart agricultural resource optimization and scheduling system according to claim 4 is characterized in that: Combining the root potential with the preset threshold value to generate the fertility gap level includes the following steps: The real-time root potential is compared with the threshold value by the difference percentage quantification method, satisfying the following formula: in, Indicates the implementation difference, Indicates the real-time measured root potential value. Indicates the current nutrient absorption threshold, represents the basal potential threshold; Combined with historical data, the laboratory divides fertility gap into four levels: normal range, mild deficiency, moderate deficiency, and severe deficiency; the first threshold of fertility gap is pre-set , the first threshold , the first threshold The hierarchical logic is constructed using a multi-interval partitioning strategy and a progressive response mechanism to avoid resource waste caused by a single threshold trigger. like It indicates that the fertility gap level is within the normal range. Indicates that the fertility gap level is in a mild deficiency. Indicates that the fertility gap level is moderately deficient. It indicates that the fertility gap level is in severe deficiency.

6. A cloud computing-based smart agricultural resource optimization and scheduling system according to claim 4, characterized in that: The dynamic demand analysis module also includes a dynamic indicator generation unit and a trigger condition detection unit; The dynamic index generating unit performs weighted summation on the water demand coefficient, the transpiration efficiency index, and the fertility gap level to generate a dynamic demand index; The trigger condition detection unit inputs the dynamic demand indicator into the trigger condition detection unit, and pre-sets a threshold for determining the trigger to determine whether to trigger a resource collaboration request.

7. The cloud computing-based smart agricultural resource optimization and scheduling system according to claim 6, characterized in that: Blockchain scheduling management module, including smart contract generation unit and resource matching and screening unit; A smart contract generation unit, used to write the geographic location, resource type, and verification hash value of the resource request into the smart contract; Resource matching screening unit, used to generate a matching priority list based on farm location, resource surplus, and contribution point history.

8. The cloud computing-based smart agricultural resource optimization and scheduling system according to claim 7, characterized in that: Contribution points management module, including basic points calculation unit, emergency weighted correction unit, and points redemption interface; The basic score calculation unit is used to generate basic points according to the product of the resource type base value and the usage time; The emergency weighted correction unit is used to adjust the integral weight according to the amplitude and time attenuation coefficient of the dynamic demand index exceeding the threshold; The points redemption interface is used to convert contribution points into agricultural machinery leasing rights or agricultural input discount certificates.

9. The cloud computing-based smart agricultural resource optimization and scheduling system according to claim 8, characterized in that: Adjusting the integral weight according to the magnitude of the dynamic demand indicator exceeding the threshold and the time attenuation coefficient includes the following steps: The basic integral value enters the emergency weighted correction unit for dynamic adjustment. The unit has a built-in double-layer adjustment mechanism of emergency adjustment coefficient and time decay coefficient. The emergency adjustment coefficient is calculated based on the percentage of the dynamic demand index exceeding the preset threshold, and satisfies the following formula: in, It represents the urgency adjustment coefficient, which reflects the dynamic multiplier of the suddenness of resource demand and is positively correlated with the deviation value of the demand index; Indicates the current dynamic demand index value; Indicates the threshold value of the current dynamic demand indicator; The time decay coefficient is calculated based on the interval between the resource request initiation time and the current time, using an exponential decay mode, and satisfies the following formula: in, Represents the time decay coefficient, which is a compensation factor used to measure the timeliness of response. The longer the delay, the smaller the correction force. Indicates the decay rate set by the system, Indicates the time difference after the request is initiated; The final corrected integral value satisfies the following formula: in, Indicates the basic integral value, Indicates the final corrected integral value.

10. The cloud computing-based smart agricultural resource optimization and scheduling system according to claim 1, characterized in that: The intelligent verification module includes a verification trigger unit, a data comparison unit, and a dispute handling unit; The verification trigger unit is used to detect whether the difference in hash value between the resource call record and the local sensor data exceeds a threshold; The data comparison unit extracts the equipment logs and sensor storage data of the target farm; The dispute handling unit freezes the dispute points and initiates a manual review process, and the verified results are updated to the blockchain ledger.

Images