Agricultural production and sales whole chain dynamic tracking and management method based on space-time big data

Abstract

This invention relates to the field of agricultural technology extension services, and discloses a method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data. The method includes: acquiring agricultural product location data sequences, temperature and humidity data, and vibration acceleration data; calculating agricultural product quality activity indicators to determine the intervention benchmark for agricultural technology extension services; calculating the sum of radial displacement velocities of agricultural products en route towards the target sales node to determine the flow convergence degree and identify the intensity of competition for extension resources; reducing the extension weight of the target sales node and generating scheduling instructions containing path offset data; extracting the inertial parameters before signal limitations, calculating the quality activity indicators during signal interruption, and correcting statistical deviations; and calculating the harvest completion rate at the production site based on the harvest query feedback delay. This invention establishes a spatiotemporal trajectory and quality loss mapping mechanism to address the risk of extension resource crowding and maintain the consistency of control logic.

Description

Technical Field

[0001] This invention relates to a method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data, belonging to the field of agricultural technology extension service technology. Background Technology

[0002] Current agricultural extension services involve precise alignment of production and sales decisions with phenological windows. Existing technologies utilize harvesting records from production bases, data from transfer nodes, and the status of sales terminals for management. The post-harvest control process is defined as unidirectional monitoring of physical displacement, which is coordinated with routine logistics scheduling by tracking changes in the physical location of the carrier.

[0003] As the scale of circulation increases, simply recording location changes reveals inherent logical flaws. Agricultural products experience physiological activity decline and market supply and demand fluctuations during displacement. Because growth status data at the production end and trajectory data at the circulation end are disconnected, maintaining the effectiveness of promotion decisions typically requires investment in manual sampling or deploying sensor monitoring networks, which introduces delays in information acquisition during execution. Simply shortening the logistics cycle or increasing sampling frequency cannot establish a mapping relationship between trajectory characteristics and product quality. Existing management logic ignores the cumulative impact of trajectory deviation, time stagnation, and mechanical stress on the physiological integrity of products. When faced with complex changes in operating conditions, simple physical location parameters cannot support the dynamic adjustment of promotional weights at the sales end, causing the allocation logic of promotional resources to lag behind the actual situation of agricultural products. For example, Chinese invention patent CN118941123A discloses a digital agriculture full-chain management traceability system and method based on blockchain and AI algorithms. By acquiring past growth data and combining it with the blockchain consensus mechanism, it can realize the pre-tracing of growth status and actual measurement comparison. However, this type of static parameter backtracking control logic is limited to monitoring physiological indicators at the production end, such as height and number of leaves. It ignores the real-time mapping relationship between the physical movement characteristics of the carrier in the circulation link and the decline of the physiological activity of agricultural products. In complex long-distance transportation and large-scale harvesting windows, due to the lack of quantitative judgment of physical stress, trajectory deviation and geographic spatial convergence density in transit, the system is difficult to perform predictive promotion weight adjustment before the overall quality of agricultural products deteriorates, and cannot break the perception barrier between the biological attributes and market promotion attributes of agricultural products.

[0004] Therefore, how to directly transform the physical operation characteristics of the carrier into production and sales risk hedging decisions, so as to break down the perception barrier between the biological attributes and market attributes of agricultural products, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data, the method comprising the following steps: Step S101: Obtain the location data sequence of agricultural products during the circulation cycle, and extract the temperature and humidity data and vibration acceleration data of agricultural products in the transportation path to characterize the environmental state. Step S102: Based on the location data sequence, temperature and humidity data and vibration acceleration data, calculate the agricultural product quality activity index during the circulation process, and generate an intervention benchmark signal for agricultural technology extension services when the agricultural product quality activity index reaches the preset quality deterioration threshold. Step S103: Monitor the location feedback trajectory of agricultural products entering the target sales node within a preset radius, calculate the sum of the radial displacement velocities of all agricultural products moving toward the target sales node relative to the target sales node per unit time, determine the agricultural product circulation convergence degree of the target sales node, and use this agricultural product circulation convergence degree to characterize the resource competition intensity of the agricultural product promotion position corresponding to the target sales node. Step S104: When the convergence of agricultural product circulation exceeds the preset promotion resource carrying capacity threshold, the promotion weight parameter for the target sales node in the agricultural technology promotion service platform is reduced, and the promotion guidance weight of adjacent redundant nodes is increased simultaneously, so as to generate a dynamic scheduling instruction containing path offset data that guides agricultural products to transfer to redundant nodes. Step S105: When agricultural products enter the area where satellite positioning signals are restricted, extract the inertial parameters of the operation before entering the area where satellite positioning signals are restricted, calculate the quality activity index of agricultural products during the signal interruption phase by combining the local time statistics unit, and correct the supply quantity statistical deviation between the place of origin and the place of destination based on the determined stagnation time at the moment the signal is restored. Step S106: Monitor the displacement change rate of agricultural products at the origin node of the production area. When the displacement change rate is lower than the preset displacement fluctuation threshold, send a harvesting progress query request to the production control terminal of the production area. Calculate the harvesting completion rate of the production area based on the time delay change characteristics of the feedback response. Input the harvesting completion rate of the production area as a supply-side variable into the dynamic scheduling command.

[0006] Preferably, the calculation of agricultural product quality activity index in step S102 includes: obtaining the average temperature and humidity fluctuation values ​​and the cumulative road bump frequency that characterize the transportation environment; establishing a mapping relationship between agricultural product quality activity index and transportation mileage, environmental conditions and cumulative storage time; and calculating the quality loss rate of agricultural products in different logistics periods by calculating the movement distance of agricultural products in each segment of the production and sales chain and combining the mapping relationship.

[0007] Preferably, determining the convergence degree of agricultural product circulation at the target sales node in step S103 includes: establishing a geofence space centered on the target sales node, obtaining the GPS coordinates and timestamp data of all agricultural products in transit within the geofence space; converting the coordinates and timestamp data into radial movement rates relative to the center, and summing the radial movement rates to obtain the inflow flux representing the state of the promotion position being squeezed.

[0008] Preferably, the intensity of resource competition is positively correlated with the inflow flux; the method also includes: dynamically adjusting the promotion resource carrying capacity threshold based on the historical average daily throughput of the target sales node under different seasonal phenological periods, so that the trigger sensitivity of the dynamic scheduling command is aligned with the maturity window of agricultural products.

[0009] Preferably, the reduction of the promotion weight parameter of the target sales node in step S104 includes: identifying the market value level of agricultural products through agricultural product quality activity indicators; when the agricultural product circulation convergence exceeds the limit, prioritizing the entry of agricultural products with agricultural product quality activity indicators at the quality deterioration threshold into the warehousing channel of the target sales node, and suspending the display rights of agricultural products with agricultural product quality activity indicators higher than the preset safety value at the target sales node.

[0010] Preferably, step S105, simulating the quality loss process of agricultural products during the signal interruption phase, includes: establishing a quality loss function relationship describing the relationship between the signal interruption duration and the quality loss variable; the quality loss function relationship is expressed by the following formula: ,in, This represents the quality loss variable during the signal interruption phase. This is a motion state correction factor determined based on the average moving speed before entering an area with limited satellite positioning signals. and These are the signal interruption time and the recovery time, respectively. This represents the metabolic loss coefficient as it evolves over time.

[0011] Preferably, step S106, which calculates the harvest completion rate at the production site based on the time delay variation characteristics of the feedback response, includes: obtaining the feedback response time of the production control terminal at the production site to the harvest progress query request, and identifying the changing trend of the feedback response time; when the feedback response time shows a linear increasing trend, it is determined that the harvest completion rate at the production site is at the end of the harvesting period, and the transportation capacity allocation ratio for the starting node at the production site is reduced.

[0012] Preferably, the method further includes: constructing a resource allocation equilibrium logic diagram based on the convergence degree of agricultural product circulation of all target sales nodes within the full chain coverage; when multiple target sales nodes within a local administrative region are simultaneously at the upper limit of resource competition intensity, adjusting the transportation weight to forcibly delay the delivery time interval of the originating node in the production area, so as to reduce the convergence degree of agricultural product circulation from the source of supply.

[0013] Preferably, the dynamic scheduling instructions are sent to the agricultural product transport terminals through the back-end server of the agricultural technology extension service platform; the dynamic scheduling instructions also include: alternative sales node location data and premium suggestion data based on agricultural product quality activity indicators, so as to maintain the consistency of the control logic of agricultural technology extension services in the context of unstable communication.

[0014] Preferably, the temperature and humidity data in step S101 is obtained by reusing base station signaling of the mobile communication network; the method maps the positioning data sequence to a multi-dimensional grid in the geographic information system, identifies the temperature offset value in each grid, and uses the temperature offset value as a correction parameter for the agricultural product quality loss calculation model to improve the calculation accuracy of agricultural product quality activity indicators.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. In the dynamic tracking of the entire agricultural production and sales chain, the method collects the spatiotemporal trajectory vector of agricultural product carriers, extracts trajectory deviation signals and time stagnation features from them, and combines them with phenological attenuation rules to transform the physical displacement state of the carriers into the evolution trend of agricultural product commodity attributes in real time. This determines the production-sales consistency deviation, and dynamically adjusts the promotion weight at the sales end based on the deviation. The system establishes a logical feedback between supply capacity prediction and market demand window at the stage when agricultural products are in transit. This mechanism eliminates the perception barrier between the growth status at the production end and the dynamic information at the circulation end, ensuring that promotion instructions are supported by real-time physical reality and avoiding resource allocation misalignment caused by the lag in quality information in traditional promotion services.

[0016] 2. The method extracts the instantaneous pulse and high-frequency fluctuation characteristics of the acceleration component in the spatiotemporal trajectory vector, calculates the cumulative stress load during the circulation process, generates a latent damage factor to correct the remaining value of the commodity, and achieves a deep perception of the internal physiological integrity of agricultural products by utilizing existing positioning data. When a high risk of latent damage is detected, the promotion weight of high-premium markets is automatically reduced, and a rapid clearing mode of nearby sales nodes is triggered. The system can capture the critical point of quality damage before the overall deterioration of agricultural products occurs. This not only protects the reputation of high-end promotion channels, but also gives agricultural extension organizations more time to make decisions under complex transportation conditions.

[0017] 3. The method determines the spatiotemporal convergence density of sales nodes by monitoring the spatiotemporal trajectory vectors of multiple agricultural product carriers destined to arrive at the same node. This density is then used to identify the competitive intensity of promotional resources at that node. When the convergence density exceeds a preset carrying capacity threshold, the system proactively reduces the promotional service weight of that node and simultaneously increases the guidance weight of nearby redundant nodes. This logic, through the identification of group trajectory characteristics within the geographic space, enables proactive perception of the risk of promotional resources being squeezed out. It avoids node overload and unsold goods caused by the instantaneous convergence of carriers from multiple production areas during peak harvest seasons, achieving a spontaneous and balanced allocation of promotional resources across the entire network. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the dynamic tracking and control process of the entire agricultural production and sales chain based on spatiotemporal big data, as described in this invention. Figure 2 This is a module structure diagram of the agricultural production and sales chain dynamic tracking and control system of the present invention. Detailed Implementation

[0019] The following embodiments are intended to explain the present invention, and not to limit the scope of protection of the present invention.

[0020] This invention provides a method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data. It utilizes a topologically linked space comprised of production, logistics, and sales ends. By collecting location data sequences of agricultural product carriers in physical space, environmental temperature and humidity data, and vibration acceleration data, and employing a phenological attenuation model, the spatial deviation, temporal stagnation, and physical stress characteristics of the carriers along the transportation path are converted into agricultural product quality activity indicators. Furthermore, the promotion weight parameters in the agricultural technology extension service platform are dynamically adjusted based on the convergence degree of agricultural product circulation at sales nodes. A feedback mechanism is established between the carrier's kinetic energy dissipation characteristics and the remaining value of the commodity, enabling spontaneous balanced allocation of agricultural extension resources across the entire network. This method performs a full-chain spatiotemporal logical topological mapping, pre-defining a spatiotemporal weight matrix composed of production bases, logistics nodes, and sales areas. This is used to record the coordinates of each geographical region, so that the supply and demand heat weights of a specific geographical region are matched with the circulation friction coefficient; the system utilizes Beidou positioning terminals deployed on agricultural product carriers, according to... to By setting a preset sampling period, the latitude and longitude coordinates and timestamps of agricultural products during their circulation cycle are continuously acquired to form a sequence of agricultural product location data. The temperature offset of the geographic grid where the agricultural product carrier is located is obtained through base station signaling of the mobile communication network, and humidity data collected by sensors and vibration acceleration data output by a triaxial accelerometer are retrieved. .

[0021] Based on the fact that agricultural products experience a decline in physiological activity during physical displacement, and that changes in physical location cannot directly reflect the remaining value of the commodity, the system adopts an agricultural product quality activity index. Characterize the access benchmark for promotion services and extract location data sequences. mileage in and the average temperature fluctuation in the temperature offset value The system calculates the distance agricultural products travel in each segment of the production and sales chain, combined with the cumulative storage time. Establish a mapping relationship, including calculating the activity index of agricultural product quality. The logical procedure is as follows: obtain the cumulative frequency of road surface bumps. Used to correct the rate of quality loss, when the quality activity index of agricultural products... Reduced to the preset quality degradation threshold At that time, the system generates an intervention reference signal for agricultural technology extension services; the biological time counting unit is implemented through the processor's timing logic, which follows... The sampling frequency is used to read the ambient temperature data output by the sensor, and a preset sampling frequency is used to read the ambient temperature data. The temperature effect equation converts each physical second into an equivalent physiological duration reflecting the physiological oxidative activity of agricultural products. The processor performs an accumulation operation on the equivalent physiological duration to output the cumulative storage duration. This converts the physical transport time into a parameter reflecting the degree of bioenergy decay; before the tracking logic is initiated, the processor calls a pre-stored variety physiological characteristic matrix, which contains the variety's physiological characteristics... to Temperature range baseline respiratory rate The system utilizes a high and low temperature alternating damp heat test chamber to perform equivalent physiological duration calibration on agricultural product samples, and measures the carbon dioxide release at different temperature points to fit and derive the metabolic loss coefficient. Temperature sensitivity constant, and uses this temperature sensitivity constant as an inherent attribute input to the agricultural product quality loss calculation model, as well as the quality loss threshold. Select the variety whose fruit firmness decreased to the initial value. Normalized active sites were determined, and fruit firmness was measured using a texture analyzer puncture test. The sampling frequency was set to [value missing]. .

[0022] During peak harvest season, agricultural products from multiple production areas converge on specific sales nodes within the same time window, leading to a risk of crowding out promotional slots at those nodes. The system then configures a flow convergence degree. The monitoring procedures establish a radius of [missing information] centered on the target sales node. The geofenced space, in which The value is the agricultural product under this transportation condition. Average travel distance per hour; obtain the real-time coordinate sequence of all agricultural product carriers in transit within the geofenced space, and convert it into radial movement rate relative to the center. The sum of radial displacement velocities of all agricultural products in transit relative to the target sales node is calculated per unit time to obtain the inflow flux characterizing the state of crowding out promotion positions. The processor will use the spatiotemporal weight matrix Each element in the data is mapped to an independent raster cell in the GIS system, and the system uses the location data sequence received from each cell. Spatial density calculation flow convergence And when the flow convergence degree was monitored Exceeding the promotion resource capacity threshold At that time, the promotion guidance weights of adjacent redundant nodes are automatically allocated according to the inverse square ratio of the distance to neighboring grids. This establishes a linear feedback loop between dynamic traffic load and promotion weight offset at the physical topology level.

[0023] To achieve a balanced distribution of promotional resources across geographic space and reduce waiting costs in the delivery process, the system executes promotional weight parameters. The dynamic correction process, in monitoring the degree of flow convergence Exceeding the preset threshold for promotional resources At that time, the backend server of the agricultural technology extension service platform reduced the promotion weight parameters for the target sales nodes. Simultaneously, the promotion guidance weight of adjacent redundant nodes is increased proportionally. During this process, the system identifies the market value level of agricultural products and prioritizes ensuring the quality activity indicators of agricultural products. At the quality deterioration threshold Batch after batch enters the warehousing channel of the sales node, generating dynamic scheduling instructions containing path offset data guiding the transfer of agricultural products to redundant nodes, and issuing them to the carrier terminal. By adjusting the dispatch weights, the delivery time interval is optimized, reducing the carrier convergence pressure from the supply source. To address the obstacle of monitoring interruption caused by carriers entering signal-restricted areas with complex geographical environments, the system introduces a spatiotemporal trajectory shadow reconstruction mechanism. When agricultural products enter areas with restricted satellite positioning signals, the storage unit extracts the inertial parameters of operation before entering the blind zone, including instantaneous rate gradient and motion direction stability indicators. The system activates the local time statistics unit, using the biological time counting unit to simulate the energy state evolution of agricultural products during the signal interruption phase. By calculating the elastic coefficient of displacement and time of the carrier in the signal interruption interval, it determines whether there is unplanned stagnation. At the moment of signal recovery... The system calculates the quality loss variable during the signal interruption phase based on the following quality loss function. : ,in, The moment the signal is interrupted. This is a motion state correction factor determined based on the average moving speed before entering an area with limited satellite positioning signals. This represents the metabolic loss coefficient as it evolves over time.

[0024] Motion state correction factor The values ​​are determined based on the vibration energy distribution before entering the satellite positioning signal-restricted area; the processor calculates the values ​​before the signal disappears. Vibration acceleration data The sequence is processed by Fast Fourier Transform to obtain acceleration pulses. to Frequency domain energy accumulation value When the cumulative energy value Exceeding the preset stress damage threshold When the carrier is determined to be operating on an unpaved road surface, the motion state correction factor is applied. From the benchmark value Adjusted to to The region represents the area where mechanical stress induces an increase in the rate of endogenous ethylene synthesis, utilizing the aforementioned motion state correction factor. The system compensates for the quality loss function to predict the quality evolution trend within the signal blind zone. Since the delay in acquiring harvest progress data at the production end affects the decision-making timeliness at the sales end, the system executes a starting displacement change rate monitoring procedure to monitor the displacement change rate of agricultural products at the origin node. When the displacement change rate is continuously lower than the preset displacement fluctuation threshold and the cumulative time reaches the preset detection window, the promotion server sends a harvest progress query request to the production control terminal at the production site. The system analyzes the feedback response delay characteristics of the production control terminal to the request. When the feedback response time shows a linear increasing trend, it determines that the harvest completion rate at the production site is in the late harvest stage and uses the harvest completion rate at the production site as a supply-side variable input into the dynamic scheduling command. The processor extracts the feedback response delay. Based on the evolution characteristics of the preset monitoring cycle, reflecting the computational resource competition and logical queuing waiting time generated by the production control terminal at the production site during the peak period of material warehousing, the system calculates continuously... Feedback response delay within each detection window Second derivative ,in Set as to When the second derivative The feedback response delay is greater than zero. Crossing the reference delay When the boundary is doubled, the production site operation is determined to have entered the equipment removal and cleanup stage, and the production site harvesting completion rate is determined. According to the formula calculate, To indicate the completion rate of harvesting at the production site. This is a correction factor based on the variety's harvesting cycle. For the current moment, The start time of harvesting. For actual feedback delay, This is the initial reference delay.

[0025] Example 1: In the scenario of transporting fresh peaches during peak harvest season, the cold chain carrier cluster originating from the production base enters a 45-minute area with restricted satellite positioning signals on its journey to the target sales node. Due to high-frequency turbulence caused by geological structures, the system, based on the spatiotemporal trajectory shadow reconstruction mechanism described in the aforementioned specific implementation method, extracts the agricultural product positioning data sequence in real time before entering the area with restricted satellite positioning signals. The final instantaneous velocity gradient value in the data is combined with the vibration acceleration data collected in real time by the triaxial accelerometer and stored in the local storage unit. By determining the acceleration pulse in to The energy distribution characteristics in the frequency domain are used to determine the motion state correction factor at this time. During this signal interruption period, the system continuously calculates the metabolic loss coefficient using the local biological time counting unit. The temperature offset value is nonlinearly superimposed with the physical stress generated by high-frequency vibration. When the carrier is at time... When communication is restored, the processor executes the quality loss function. The operation, where, The moment the signal is interrupted. This is a motion state correction factor determined based on the average moving speed before entering an area with limited satellite positioning signals. The metabolic loss coefficient evolves over time, and the calculated result represents the agricultural product quality activity index for this batch of agricultural products. Within 45 minutes, the value decreased from an initial 0.92 to 0.78, which is close to the preset quality degradation threshold. That is, 0.75.

[0026] Simultaneously, by monitoring the geofenced space surrounding the target sales node, the real-time coordinate sequence of all agricultural product carriers in transit within the space is obtained, and the resulting flow convergence degree is calculated. The value reached 12.5, exceeding the preset threshold for promotional resource capacity. Due to the activity indicators of agricultural product quality The decline trend and the turnover pressure of the target node are at a high level in sync, and the system automatically triggers the promotion weight parameter. The dynamic revision process no longer applies the promotion and guidance logic to the original primary wholesale markets; instead, the processor bases its decisions on quality loss variables. The resulting surplus value of goods will be used as the promotion weight parameter for this batch. Adjust the settings and search for locations within 15 kilometers of the current coordinates with high flow convergence. For secondary pre-position warehouse nodes with a score below 5.0, their corresponding promotion guidance weight will be reduced. The system generates a 30% improvement in dynamic scheduling instructions, which include path offset data to bypass congested sections. This guides the carrier to enter a secondary node with distribution capabilities within 20 minutes, enabling the batch of agricultural products to be stored and labeled before morphological deterioration. The physical stress characteristics of the carrier in the blind zone are directly converted into a corrective weight for the competitive priority of promotion resources, proving the causal coupling between the physical characteristics of the carrier and the market circulation heat in the entire chain's spatiotemporal logical space.

[0027] Example 2: This experiment utilizes a cold chain transportation simulation platform integrating a constant temperature and humidity control unit and a triaxial vibration unit. The experimental data comes from real-time signals collected by the sensor array of the experimental platform. The range of the triaxial accelerometer is... The resolution is The sampling frequency is set to The temperature sensor accuracy is The sampling period is set to This value is based on a trade-off between the power consumption load of the data processing chip and the rate of change of the thermosensitive characteristics of agricultural products. It is used to maintain the energy efficiency ratio while capturing instantaneous temperature fluctuations. To simulate road environment disturbances, an amplitude of [value missing] was injected into the vibration signal during the experiment. Gaussian white noise; the test scheme sets up three working condition groups with gradient intensities. Test group one simulates a plain highway section, and the ambient temperature fluctuation range is set to be... And the rate of displacement change remained stable; Experiment 2 simulated mountain roads, with the ambient temperature fluctuation range set at [range missing]. The system also superimposed high-frequency acceleration pulses; as a control group, experimental group three adopted a judgment logic that did not include vibration stress correction terms, and the system collected the original positioning data sequence of agricultural products in each group within a 24-hour circulation cycle. In conjunction with environmental parameters, calculate the agricultural product quality activity indicators according to the aforementioned specific implementation procedures. .

[0028] Table 1: Comparison of data between the sample group and the control group of this invention Based on the data analysis in Table 1, after taking into account the correction for vibration stress, the calculated agricultural product quality activity index for experimental group two was... The value is 0.68, which is lower than the preset quality loss threshold. That is, 0.75, to generate a benchmark signal for promotion service intervention, and to assign promotion weight parameters to the target sales node. The value was reduced to 0.45; test group three lacked the cumulative road bump frequency. The correction logic yielded the agricultural product quality activity index. A value of 0.82 leads to the system maintaining a high promotion weight, resulting in a mismatch between promotional resources and the actual value of the goods; in the group exceeding the upper limit, as environmental temperature fluctuations and mechanical stress intensity exceed the preset range, the agricultural product quality activity index... The non-linear decline indicates that the quality degradation is irreversible.

[0029] Example 3: This example combines Figures 1 to 2 This paper explains the method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data, such as... Figure 1 As shown, starting from step S101, the process acquires the location data sequence of agricultural products within their circulation cycle and extracts temperature and humidity data, as well as vibration acceleration data, representing the environmental conditions of the agricultural products along the transportation path. The process then proceeds to step S102, where the agricultural product quality activity index is calculated based on the aforementioned data. When this index reaches a preset quality deterioration threshold, an intervention benchmark signal for agricultural technology extension services is generated. The subsequent step S103 monitors the location trajectory within a preset radius of the target sales node and calculates the sum of radial displacement velocities to determine the circulation convergence degree, thus characterizing the resource competition intensity of the target sales node's promotional location. Based on the above determination step S10... 4. When the flow convergence exceeds the promotion resource carrying capacity threshold, the promotion weight of the target node is reduced and the weight of adjacent redundant nodes is increased, thereby generating a dynamic scheduling instruction containing path offset data; at the same time, it includes step S105 for specific working conditions, extracting the running inertia parameters and calculating the quality activity index of the interruption stage when entering the signal-restricted area, and correcting the production and sales supply statistical deviation according to the determined stagnation time at the moment the signal is restored. In addition, step S106 constructs a supply-side feedback mechanism, which monitors the displacement change rate of the starting node and sends a harvesting progress query request when the fluctuation is low, calculates the harvesting completion rate of the production area according to the feedback delay characteristics and inputs it into the dynamic scheduling instruction.

[0030] like Figure 2As shown, the architecture consists of three parallel functional systems and their subordinate modules. The multi-dimensional perception and data acquisition system on the left is responsible for basic information collection, specifically including four modules: location data sequence acquisition, temperature, humidity and vibration data extraction, harvest completion inversion at the production site, and operation inertial parameter extraction. The core indicator calculation and modeling system in the middle is responsible for data processing and feature extraction, covering four computational units: agricultural product quality activity index calculation, circulation convergence determination, resource competition intensity identification, and signal interruption stage deviation correction. The dynamic decision-making and scheduling control system on the right is responsible for strategy generation, including four execution modules: promotion service intervention benchmark generation, promotion weight parameter reduction, redundant node promotion weight increase, and dynamic scheduling command issuance. The arrows indicate the direction of the data flow from the perception system through the calculation system to finally transform into decision-making and control commands in a unidirectional closed-loop logic.

[0031] Example 4: In a specialized agricultural technology extension service scenario for the Hongyan strawberry variety, the system implements a quality deterioration threshold. The parameter calibration procedure was initiated, and the initial state definition procedure was started. The target species were defined as Red Face strawberries with a maturity of 85% to 90% and an initial moisture content of 90.5%. The enabling environment was a mobile promotion terminal with BeiDou positioning and 5G communication capabilities. Under controlled conditions, the system performed gradient temperature stress testing on this batch of strawberries. Pressure sensors were used to measure the evolution of fruit firmness over time. It was determined that when the fruit firmness decreased to 70% of its initial value, the strawberry variety entered a quality deterioration period. At this point, the calculated agricultural product quality activity index was... The value was recorded as 0.65, and this was locked as the quality deterioration threshold for this variety in the current production season. Regarding the metabolic loss coefficient To determine this, the system pre-records respiration intensity curves under different temperature gradients in the storage unit, and establishes the average temperature fluctuation value based on the Arrhenius formula. The linear compensation relationship between the metabolic rate and the actual temperature is calculated by inputting measured temperature data and retrieving the respiratory intensity curve, and the corresponding metabolic loss coefficient is output. .

[0032] Promotional resource capacity threshold for target sales nodes The system executes a calibration procedure based on physical processing throughput to obtain the number of cold chain inbound berths in the target storage center. Average turnover rate per berth Set their product as the maximum inflow limit for the sales node within a unit of time, and the promotion resource carrying capacity threshold. To monitor the harvesting progress at the production site, the system periodically sends harvesting progress query request pulses to the mobile extension terminals at the production site and initiates a process judgment quantification procedure to analyze the feedback response delay characteristics. The system collects delay data within 10 consecutive detection windows to construct a delay change rate function. ,in, The rate of change of time delay, To provide feedback on response delay, For sampling time, when observed Furthermore, when the latency value exhibits a linear growth trend, based on the latency change rate... Calculation of slope characteristics to determine harvest completion rate at the production site Specifically, the degree of harvest completion at the production site is determined using the following linear mapping formula. : ,in, To indicate the completion rate of harvesting at the production site. This is a correction factor based on the variety's harvesting cycle. To calculate the cumulative harvesting time, To provide feedback on response delay, This serves as the baseline time delay for the initial harvesting stage; when the calculation yields... When the value exceeds 0.95, the system determines that the production area is in the late stage of harvest and triggers the promotion weight parameter. The dynamic revision process will promote the allocation of resources to the production areas to improve harvesting completion rates. Switch to emerging production areas with a value below 0.30.

[0033] Example 5: In the initial calibration scenario of deploying a cross-regional citrus production and sales chain, the system performs circulation parameter calibration for a specific geographical area, selecting past data from that geographical area. Using average cold chain logistics time data for each production season and road surface smoothness statistics as raw inputs, a discretization mapping algorithm is used to transform geographical coordinates into spatiotemporal elements including terrain slope and transshipment frequency, thereby determining the spatiotemporal weight matrix. The initial circulation baseline value; before initiating the tracking logic, the system retrieves the historical physiological characteristic dataset of agricultural products corresponding to the production area, and determines the standard ambient temperature by performing steady-state response analysis. The static loss slope is used as the metabolic loss coefficient. The baseline offset is used to eliminate the influence of different geographical environments on the promotion weight parameters. Correcting the effects of deviations in the action.

[0034] When the system encounters signal drift caused by batch replacement of sensor hardware, it initiates the zero-point residual calibration program of the local time statistics unit and the triaxial accelerometer, which continuously collects data while stationary. Vibration acceleration data The sequence is analyzed and its statistical mean is calculated to obtain a systematic deviation correction value at the hardware level. This correction value is then fed back to the computational loop to calibrate the real-time acquired instantaneous pulse intensity, ensuring accurate calculation of the motion state correction factor. Remove pseudo-random noise caused by hardware aging; synchronously execute processor to promote resource carrying capacity threshold. The dynamic verification logic obtains the number of cold chain berths. And the berth turnover rate under simulated full-load inbound scenarios. Based on the fluctuation characteristics, the fault tolerance range of the warehousing priority determination logic is recalibrated to ensure that the system maintains the activity index of agricultural product quality when environmental conditions undergo physical deviations. Consistency with actual physiological state.

[0035] Example 6: In an agricultural technology extension service deployment scenario targeting kiwifruit varieties, the system executes an offline construction program for the respiration intensity benchmark and loss coefficient matrix. It selects kiwifruit samples in the physiological after-ripening stage and records the initial soluble solids content. The samples are then placed in a temperature-controlled environment with a precision of [insert temperature here]. Furthermore, within the sealed test chamber integrating the carbon dioxide sensor, and in accordance with... Adjust the ambient temperature step size to cover to Within the logistics operating range, the system continuously collects the rate of change of carbon dioxide concentration under various temperature gradients and fits the data to derive the loss rate constant at the corresponding temperature. The processor establishes the average temperature fluctuation value based on the Arrhenius equation. An exponential function model relating metabolic rate and metabolic rate is used, and the resulting parameter mapping table is stored in a local knowledge base as a metabolic loss coefficient. The retrieval benchmark; regarding the risk of mechanical damage caused by transportation stress, the system uses a texture analyzer to analyze the impact of different cumulative road bump frequencies. The treated samples underwent pressure puncture tests to measure the evolution of fruit deformation under pressure and the puncture torque of the peel, thereby determining the quality degradation threshold of this variety at the current moisture content. It is 0.68.

[0036] When dealing with decision-making deviations caused by changes in unpredictable logistics routes, the system activates the spatiotemporal weight matrix. The dynamic adaptive update procedure acquires the location data sequence of the carriers carrying agricultural products in the newly added path. Extract the radial movement rate of the carrier in each road segment. The actual travel time is calculated as a percentage of the navigation prediction time, and the flow rate per unit distance due to traffic congestion is also calculated. The system calculates the flow parameter correction for a specific geographic raster by analyzing the fluctuation characteristic values. This correction is then fed back to the backend computing unit of the agricultural technology extension service platform to update the spatiotemporal weight matrix. The distribution of circulation weights within the system enables it to maintain the activity indicators of agricultural product quality when faced with path deviations or temporarily added transit nodes. The consistency of predictions makes the promotion weight parameters The allocation instructions are always based on the latest physical flow heat.

[0037] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data, characterized in that, The method includes the following steps: Step S101: Obtain the location data sequence of agricultural products during the circulation cycle, and extract the temperature and humidity data and vibration acceleration data of agricultural products in the transportation path to characterize the environmental state. Step S102: Based on the location data sequence, temperature and humidity data and vibration acceleration data, calculate the agricultural product quality activity index during the circulation process, and generate an intervention benchmark signal for agricultural technology extension services when the agricultural product quality activity index reaches the preset quality deterioration threshold. Step S103: Monitor the location feedback trajectory of agricultural products entering the target sales node within a preset radius, calculate the sum of the radial displacement velocities of all agricultural products moving toward the target sales node relative to the target sales node per unit time, determine the agricultural product circulation convergence degree of the target sales node, and use this agricultural product circulation convergence degree to characterize the resource competition intensity of the agricultural product promotion position corresponding to the target sales node. Step S104: When the convergence of agricultural product circulation exceeds the preset promotion resource carrying capacity threshold, the promotion weight parameter for the target sales node in the agricultural technology promotion service platform is reduced, and the promotion guidance weight of adjacent redundant nodes is increased simultaneously, so as to generate a dynamic scheduling instruction containing path offset data that guides agricultural products to transfer to redundant nodes. Step S105: When agricultural products enter the area where satellite positioning signals are restricted, extract the inertial parameters of the operation before entering the area where satellite positioning signals are restricted, calculate the quality activity index of agricultural products during the signal interruption phase by combining the local time statistics unit, and correct the supply quantity statistical deviation between the place of origin and the place of destination based on the determined stagnation time at the moment the signal is restored. Step S106: Monitor the displacement change rate of agricultural products at the origin node of the production area. When the displacement change rate is lower than the preset displacement fluctuation threshold, send a harvesting progress query request to the production control terminal of the production area. Calculate the harvesting completion rate of the production area based on the time delay change characteristics of the feedback response. Input the harvesting completion rate of the production area as a supply-side variable into the dynamic scheduling command.

2. The method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data according to claim 1, characterized in that, Step S102, which calculates the quality activity index of agricultural products, includes: obtaining the average temperature and humidity fluctuations and the cumulative frequency of road bumps that characterize the transportation environment; establishing a mapping relationship between the quality activity index of agricultural products and transportation mileage, environmental conditions, and cumulative storage time; and calculating the quality loss rate of agricultural products in different logistics periods by calculating the distance the agricultural products travel in each segment of the production and sales chain and combining the mapping relationship.

3. The method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data according to claim 1, characterized in that, The determination of the convergence degree of agricultural product circulation at the target sales node in step S103 includes: establishing a geofence space centered on the target sales node, obtaining the GPS coordinates and timestamp data of all agricultural products in transit within the geofence space; converting the coordinates and timestamp data into radial movement rates relative to the center, and summing the radial movement rates to obtain the inflow flux representing the state of the promotion position being squeezed.

4. The method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data according to claim 3, characterized in that, Resource competition intensity is positively correlated with inflow flux; the method also includes: dynamically adjusting the promotion resource carrying capacity threshold based on the historical average daily throughput of the target sales node under different seasonal phenological periods, so that the trigger sensitivity of dynamic scheduling instructions is aligned with the maturity window of agricultural products.

5. The method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data according to claim 1, characterized in that, The reduction of the promotion weight parameters of the target sales node in step S104 includes: identifying the market value level of agricultural products through agricultural product quality activity indicators; when the circulation and aggregation degree of agricultural products exceeds the limit, prioritizing the entry of agricultural products with agricultural product quality activity indicators at the quality deterioration threshold into the warehousing channel of the target sales node, and suspending the display rights of agricultural products with agricultural product quality activity indicators higher than the preset safety value at the target sales node.

6. The method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data according to claim 1, characterized in that, Step S105, simulating the quality loss process of agricultural products during the signal interruption phase, includes: establishing a quality loss function relationship describing the relationship between the signal interruption duration and the quality loss variable; the quality loss function relationship is expressed by the following formula: ,in, This represents the quality loss variable during the signal interruption phase. This is a motion state correction factor determined based on the average moving speed before entering an area with limited satellite positioning signals. and These are the signal interruption time and the recovery time, respectively. This represents the metabolic loss coefficient as it evolves over time.

7. The method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data according to claim 1, characterized in that, Step S106, which calculates the harvest completion rate at the production site based on the time delay variation characteristics of the feedback response, includes: obtaining the feedback response time of the production control terminal at the production site to the harvest progress query request, and identifying the changing trend of the feedback response time; when the feedback response time shows a linear increasing trend, it is determined that the harvest completion rate at the production site is at the end of the harvesting period, and the transportation capacity allocation ratio for the starting node at the production site is reduced.

8. The method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data according to claim 1, characterized in that, The method also includes: constructing a resource allocation equilibrium logic diagram based on the convergence degree of agricultural product circulation at all target sales nodes within the entire chain coverage; when multiple target sales nodes within a local administrative region are simultaneously at the upper limit of resource competition intensity, adjusting the transportation weight and forcibly delaying the delivery time interval of the originating node in the production area, so as to reduce the convergence degree of agricultural product circulation from the source of supply.

9. A method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data as described in claim 1, characterized in that, Dynamic scheduling instructions are sent to the agricultural product transport terminals through the back-end server of the agricultural technology extension service platform. The dynamic scheduling instructions also include: alternative sales node location data and premium suggestion data based on agricultural product quality activity indicators, so as to maintain the consistency of the control logic of agricultural technology extension services in the event of unstable communication.

10. A method for dynamic tracking and control of the entire agricultural production and sales chain based on spatiotemporal big data as described in claim 1, characterized in that, In step S101, temperature and humidity data are obtained by reusing base station signaling of the mobile communication network. The method maps the location data sequence to a multi-dimensional grid in the geographic information system, identifies the temperature offset value in each grid, and uses the temperature offset value as a correction parameter for the agricultural product quality loss calculation model to improve the calculation accuracy of agricultural product quality activity indicators.

Images