Mulberry planting mode adaptive matching system

Abstract

The invention discloses a mulberry planting mode self-adaptive matching system which comprises a park coordination node, a land parcel end device, a field communication network and a planting rule base. The system retrieves candidate modes, issues parameter ranges, collects specific indexes such as leaf temperatures and leaf area indexes according to microenvironment clustering based on a mulberry planting mode library and a phenological window, and updates a template in a closed loop mode according to feedback variables. Field edge nodes identify transplanting stress or soil water potential abnormity and are linked with ecological loads to trigger emergency rearrangement; and combined reporting and idempotent processing are supported in a weak network scene. According to the system, the survival rate and post-harvest recovery are improved, and the energy consumption and mode mismatching risk are reduced.

Description

Technical Field

[0001] The present invention relates to the fields of agricultural and forestry informatization and refined cultivation, and in particular to a mulberry tree planting pattern adaptive matching system. Background Art

[0002] As the core raw material crop in the sericulture industry chain, mulberry trees have biological characteristics such as a significant seedling growth period after transplanting, frequent leaf picking, a short post-harvest recovery period, and sensitivity to changes in moisture and ground temperature. Existing digital systems for industrial parks are mostly based on general work orders and basic monitoring: the platform issues tasks based on empirical templates, the terminal transmits the results back, and manual adjustments are made when anomalies are encountered. This approach has the following main pain points in the mulberry tree scenario: Most existing templates are empirical or single-region based, and are difficult to link with the site microenvironment (slope position, soil texture, wind and light conditions, historical pests and diseases) and phenological windows (seedling acclimatization, greening, rapid growth period, and post-harvest recovery), resulting in overly rough configurations of row spacing, irrigation and drainage, and leaf-picking intensity.

[0003] The monitoring items are mainly general environmental quantities, and there is a lack of mulberry-specific field indicators (such as leaf temperature / canopy temperature difference, leaf area index, bud germination rate, root activity indicators, etc.) to quantitatively identify the advantages and disadvantages of the model, making it difficult to support accurate template switching.

[0004] Abnormalities are disconnected from the ecology: Although abnormal equipment or water pressure can generate alarms, there is a lack of linkage with ecological loads (soil and water loss sensitivity, water quota red lines, and surrounding ecological constraints), which can easily lead to suboptimal decisions such as excessive leaf harvesting / irrigation to meet construction deadlines. Summary of the Invention

[0005] The object of the present invention is to overcome the deficiencies of the prior art and provide a mulberry tree planting pattern adaptive matching system, comprising a park coordination node, a plot end device, a field communication module and a cloud data server; the cloud data server includes a planting rule library; the park coordination node, the plot end device and the cloud data server are respectively connected to the field communication module for communication; The planting rule library includes a mulberry planting pattern library, parameter templates, threshold and weight sub-libraries, phenological window configuration, ecological load coefficient configuration, and a data dictionary; The park coordination node is used to retrieve candidate planting patterns from the mulberry tree planting pattern library based on the plot microenvironment and the phenological window, and generate and issue an allocation sheet containing a pattern template identifier and a parameter range to the plot end device; The plot-end device is used to collect mulberry tree-specific field indicators within the phenological window and form field records, and report the field records and execution results to the park coordination node; The park coordination node matches, scores and updates the ranking of candidate planting patterns based on the reported feedback variables, and after reaching the sampling coverage and verification period set by the rule base, determines the target planting pattern and rolls out the parameter template and threshold.

[0006] Preferably, the mulberry planting pattern library includes: row spacing schemes, variety or stock / scion combinations, mulch / cover configurations, irrigation and drainage strategy templates, pruning and stem setting rules, seedling management strategies and upper limits on leaf picking intensity; pattern entries are mapped to plot microenvironmental zoning and phenological windows.

[0007] Preferably, the park coordination node clusters the microenvironment of the plots through plot environmental data and soil type, terrain slope, light period distribution, wind conditions and historical pest and disease information to obtain plot cluster identification; the cluster identification participates in the pattern library retrieval and initial screening of candidate patterns.

[0008] Preferably, the candidate planting patterns are retrieved from the mulberry planting pattern library based on the plot microenvironment and the phenological window, including: the phenological window includes the transplanting seedling slow-growing period, green growth, spring shoot rapid growth period and leaf picking recovery period; pattern retrieval and parameter update are only performed within the task set that matches the current window, and parameter changes outside the window are postponed or enter a pending confirmation state.

[0009] Preferably, the mulberry tree-specific field indicators include leaf temperature or canopy temperature difference, leaf area index, bud germination rate, root activity indicator, soil moisture / water potential and ground temperature; the park coordination node performs quality verification and consistency check on the mulberry tree-specific field indicators according to the data dictionary, and generates a matching element set for model evaluation.

[0010] Preferably, the feedback variables include the survival rate, plant height or diameter at breast height increase, the improvement range of leaf area index during the preset observation period and the recovery time after leaf harvesting; when the feedback variables meet the improvement threshold and robustness conditions set by the rule base, the system improves the ranking of the corresponding mode entry, otherwise it will reduce or freeze its optional status.

[0011] Preferably, it also includes abnormal ecological load linkage. When the field edge node identifies transplanting stress or abnormal soil water potential and the ecological load coefficient of the target plot reaches the linkage condition set by the rule base, the park coordination node will suppress or limit the leaf picking, topdressing or irrigation and drainage subtasks related to the plot, and trigger a temporary switch of the mode template mapped to the plot.

[0012] The beneficial effect of the present invention is that the park coordination node retrieves candidate patterns based on the plot microenvironment clustering and the current phenological window, significantly reducing the probability of mismatching configurations such as row spacing, coverage, irrigation and drainage, and leaf picking intensity.

[0013] A matching factor set was constructed using mulberry-specific field indicators such as leaf temperature / canopy temperature difference, leaf area index, bud germination rate, root activity indicators, soil moisture / water potential and ground temperature. This enabled the evaluation of model quality to be transformed from empirical judgment to data verification, and the model was particularly sensitive to the seedling acclimatization, greening and post-harvest recovery stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a schematic diagram of the principle of an adaptive matching system for mulberry tree planting patterns; Figure 2 Schematic diagram of the process of clustering plots for park coordination nodes and using cluster identification to conduct preliminary screening of planting patterns. DETAILED DESCRIPTION

[0015] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0016] The features and performance of the present invention are further described in detail below with reference to the embodiments.

[0017] like Figure 1 As shown, a mulberry tree planting pattern adaptive matching system includes a park coordination node, a plot terminal device, a field communication network and a planting rule library; The planting rule library includes a mulberry planting pattern library, a parameter template, a threshold and weight sub-library, a phenological window configuration, an ecological load coefficient configuration, and a data dictionary; The park coordination node is configured to retrieve candidate planting patterns from the mulberry tree planting pattern library based on the plot microenvironment and the phenological window, generate and issue an allocation sheet including a pattern template identifier and a parameter range to the plot end device; The plot-side device is configured to collect mulberry tree-specific field indicators within the phenological window and form field records, and report the field records and execution results to the park coordination node; The park coordination node matches and scores candidate planting patterns based on the reported feedback variables, and after reaching the sampling coverage and verification period set by the rule base, determines the target planting pattern and rolls out the parameter template and threshold; The mulberry planting pattern library includes at least the following fields: row spacing scheme, variety or rootstock / scion combination, mulch / cover configuration, irrigation and drainage strategy template, pruning and stem setting rules, seedling management strategy and leaf picking intensity upper limit; the pattern entries establish a mapping relationship between the plot microenvironment zoning and the phenological window.

[0018] The park coordination node clusters the microenvironment of the plots through plot environmental data and soil type, terrain slope, light period distribution, wind conditions and historical pest and disease information to obtain plot cluster identification; the cluster identification participates in the pattern library retrieval and initial screening of candidate patterns.

[0019] The phenological window includes at least the transplanting seedling slow-down period, green growth, rapid spring shoot growth period and leaf-picking recovery period; the system limits the execution of pattern retrieval and parameter updates only within the task set that matches the current window, and parameter changes outside the window are postponed or enter a pending confirmation state.

[0020] The mulberry tree-specific field indicators include at least leaf temperature or canopy temperature difference, leaf area index, bud germination rate, root activity indicator, soil moisture / water potential and ground temperature; the park coordination node performs quality verification and consistency check on the above indicators according to the data dictionary, and generates a matching feature set for model evaluation.

[0021] The feedback variables include at least the survival rate, plant height or diameter at breast height increase, leaf area index improvement within a preset observation period, and recovery time after leaf harvesting; when the feedback variables meet the improvement threshold and robustness conditions set by the rule base, the system improves the ranking of the corresponding mode entry, otherwise it will reduce or freeze its optional status.

[0022] When forming an allocation order, the park coordination node refines the parameter range into lightweight rules on the plot side that can be issued, allowing the plot-side device to perform fine-grained fine-tuning based on real-time observations without exceeding the upper limit constraints, and transmit the fine-tuning records back in the report for closed-loop learning.

[0023] Abnormal ecological load linkage is satisfied: when the field edge node identifies transplanting stress or abnormal soil water potential and the ecological load coefficient of the target plot (including soil and water loss sensitivity, water quota occupancy rate or adjacent ecological red line status) reaches the linkage conditions set by the rule library, the park coordination node will suppress or limit the leaf picking, topdressing or irrigation and drainage subtasks related to the plot, and trigger a temporary switch of the mode template mapped to the plot.

[0024] The field edge nodes and the park coordination nodes adopt an asymmetric division of labor: the field edge nodes only perform local inferences on transplanting stress and soil water potential anomalies and form template switching suggestions, while the park coordination nodes perform global rearrangement and pattern determination across plots after aggregating data from multiple plots; when the conclusions of the two are inconsistent, the decision of the park coordination node shall prevail.

[0025] The field communication network supports low-bandwidth merged reporting: when the uplink quality is lower than the low-bandwidth threshold set by the rule base, the field-end device merges and sends field records and a new round of task applications or template switching suggestions in a single reporting transaction, and carries an idempotent identifier and retransmission sequence number to ensure consistency.

[0026] The park coordination node sets the priority coverage effect and validity period for emergency rearrangement: within the validity period, emergency sorting overrides regular mode sorting; when the period expires or the release conditions are met, the system restores regular sorting according to the most recent stability assessment results.

[0027] The system implements version control and audit backtracking for allocation orders, model entries, parameter templates and data dictionaries, and performs deduplication and idempotence processing in re-task applications and exception backchecks to avoid statistical deviations caused by repeated receipt or repeated evaluation.

[0028] Example 1: Adaptive Matching of Seedling-Greening Periods in Hilly Red Soil Mulberry Gardens The park's environment is located in southwestern hilly terrain, with red soil and a slope of 8–12°. The farm experiences strong annual winds and significant diurnal temperature fluctuations. The field communication network utilizes LoRa with intermittent 2G backhaul, which results in significant link fluctuations. One field-side device (including multi-source sensing and execution control) is deployed for every 2–3 mu (approximately 1.5–1.5 acres), and one field edge node is deployed for every 30–50 mu (approximately 1.5–2.5 acres). The edge nodes are equipped with local anomaly detection and buffering capabilities.

[0029] The park coordination nodes and planting rule base are centrally deployed in the park computer room. The rule base includes pattern library, thresholds and weights, phenological window configuration, ecological load coefficient configuration, and data dictionary.

[0030] Mode entry: M1 (slow seedling growth and water conservation type): row and plant spacing 3.0×0.6 m; mulching; drip irrigation with low flow rate and high frequency; upper limit of leaf picking intensity prohibited; topdressing suspended.

[0031] M2 (greening and root promotion type): row and plant spacing is the same as M1; drip irrigation with medium flow and medium frequency; trace amount of topdressing; leaves are harvested before they open.

[0032] Phenological Window: W1 (transplanting and seedling acclimatization) 0–21 days; W2 (greening) 22–45 days. Only maintenance tasks are allowed outside this window, and template switching is prohibited.

[0033] Field indicators include leaf temperature or canopy temperature difference, leaf area index, bud initiation rate, root activity indicator, soil moisture / water potential, and ground temperature. Field-side devices collect data every 10 minutes. Upon reaching edge nodes, missing data are filled and out-of-bounds data are cleared. Only data that passes quality verification is included in the matching feature set.

[0034] Feedback variables: survival rate, plant height increment, increase in leaf area index during the 14-day observation period, and post-harvest recovery time (in this case, leaves harvested during the seedling hardening / greening period were not opened, so only potential recovery capacity was recorded).

[0035] Microenvironment clustering: The platform divides mulberry fields into three types of plots (A / B / C) based on slope position, soil texture, wind exposure, and historical pest and disease data, and generates cluster identifiers.

[0036] Within W1, Class A plots retrieve M1 and M2 candidates from the pattern library based on cluster identifiers. The platform generates an allocation list (including pattern template identifiers and parameter ranges) and sends it to the plot-side devices. Without exceeding the upper limit, the plot-side devices fine-tune the drip irrigation frequency within the high-frequency range and record the fine-tuning entries. When the link is normal, reporting is performed on a 10-minute cycle; when the link is poor, low-bandwidth combined reporting is initiated. Before the end of W1, the platform compiles statistics on the compliance of feedback variables. If the survival rate and leaf area index of Class A plots meet both the improvement threshold and robustness criteria, M1's ranking in the Class A mapping is adjusted upward, and some parameters are solidified to the new template version. After entering W2, the platform re-searches and sends the M2 parameter range. The terminal releases the micro-topdressing amount and the medium flow and frequency of drip irrigation, continuing the closed-loop evaluation.

[0037] Edge nodes only perform local inferences on soil water potential anomalies related to transplanting stress and generate template switching recommendations. If the edge identifies soil water potential anomalies and the ecological load factor (including slope erosion sensitivity and water quota usage) for the plot meets the linkage conditions, the platform triggers topdressing / irrigation and drainage restrictions for the plot and maintains M1 mode. The temporary override is valid for 48 hours and will be reviewed and restored after the expiration date.

[0038] When the uplink quality is lower than the low bandwidth threshold of the rule base, the field-side device merges the reported transactions into a single transaction: the latest field record + re-task application / template switching suggestion, and carries the idempotent flag and retransmission sequence number; the platform side activates the deduplication strategy to ensure consistent statistical caliber.

[0039] Example 2: Leaf Harvesting-Recovery Period Adaptation and Emergency Rearrangement in a Sandy Soil Mulberry Garden in a River Valley The park's environment is located on an inland river valley terrace with sandy loam soil. It features an underground drip irrigation network and a small weather station. Summers are hot and dry, with strong evapotranspiration. 4G communications are the primary method, but local obstructions lead to frequent interruptions. One field-side device is deployed for every 5 mu (approximately 1.5 acres) of land, with two field edge nodes deployed in separate zones. The edge nodes provide local models, merged caching, and alarm reporting.

[0040] Model entry: M3 (conventional leaf-collecting type): row and plant spacing 2.8×0.7 m; medium leaf-collecting intensity; drip irrigation with medium flow and medium frequency; conventional topdressing.

[0041] M4 (conservative leaf harvesting - high temperature period): Extend the leaf harvesting interval; drip irrigation with medium to low flow and high frequency; reduce the topdressing gear; and thicken the mulch.

[0042] Phenological window: W3 (spring shoot rapid growth period) 46–90 days; W4 (leaf harvest recovery period) 91–150 days. This example covers the cross-window operation from W3 to W4.

[0043] On-site indicators: leaf area index, leaf temperature or canopy temperature difference, soil moisture / water potential, ground temperature, branch length and pest and disease indicators.

[0044] Feedback variables: plant height increment within W3, postharvest recovery time within W4, and increase in leaf area index.

[0045] Ecological load: The water quota occupancy rate and the ecological red line status of the river valley area are used as the source of the ecological load coefficient.

[0046] The W3 platform uses the microenvironment clustering results as an index to retrieve M3 / M4, generating an allocation list (including parameter ranges) and sending it to the field site. To accommodate local shading variations, the field site device fine-tunes the leaf-picking intensity of M3 within the permitted range and transmits back a record of the fine-tuning. Field sites collect field metrics every 10 minutes, and the platform calculates sampling coverage and robustness on a rolling basis every 30 minutes.

[0047] There were three consecutive days of high temperatures and the edge nodes detected abnormal soil water potential, and the ecological load coefficient reached the linkage conditions; the edge generated a template switching suggestion, and the platform temporarily switched the plot from M3 to M4, and implemented emergency rescheduling of leaf picking, topdressing, and irrigation and drainage tasks (valid for 72 hours).

[0048] When 4G is briefly congested, the terminal enables low-bandwidth combined reporting, combining field records and handover recommendations into a single transaction report to avoid frame loss and duplicate statistics.

[0049] Before W4 ends, the platform analyzes the performance of feedback variables for that plot during and after the emergency validity period. If post-harvest recovery time decreases and the leaf area index improves to meet the improvement threshold and robustness criteria, M4's ranking in the high temperature and low water potential scenario will be adjusted upward. If recovery after expiration is average, M3 will be restored and M4's selectable status in that scenario will be frozen for an observation period. The platform applies cascading resource constraints to adjacent plots, prioritizing water, fertilizer, and operation time for the plots undergoing emergency rescheduling. Lower-priority tasks for the remaining plots will be postponed to avoid network congestion.

[0050] Only transplanting stress and abnormal soil water potential patterns are inferred locally, generating template switching recommendations. Pest and disease indications are merely prompts, not triggering template switching. After integrating data from multiple plots, global rearrangement and finalization of the model are performed. When edge and platform judgments differ, the platform's decision prevails, and the cause of the conflict and backtracking evidence are recorded.

[0051] The foregoing description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein and should not be construed as excluding other embodiments. Rather, the present invention can be used in various other combinations, modifications, and environments and can be modified within the scope of the concept described herein through the above teachings or techniques or knowledge in the relevant field. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention are intended to be protected by the appended claims.

Claims

1. A mulberry planting pattern adaptive matching system, characterized in that: It includes a park coordination node, a plot end device, a field communication module and a cloud data server; the cloud data server includes a planting rule library; the park coordination node, the plot end device, and the cloud data server are respectively connected to the field communication module for communication; The planting rule library includes a mulberry planting pattern library, parameter templates, threshold and weight sub-libraries, phenological window configuration, ecological load coefficient configuration, and a data dictionary; The park coordination node is used to retrieve candidate planting patterns from the mulberry tree planting pattern library based on the plot microenvironment and the phenological window, and generate and issue an allocation sheet containing a pattern template identifier and a parameter range to the plot end device; The plot-end device is used to collect mulberry tree-specific field indicators within the phenological window and form field records, and report the field records and execution results to the park coordination node; The park coordination node matches, scores and updates the ranking of candidate planting patterns based on the reported feedback variables, and after reaching the sampling coverage and verification period set by the rule base, determines the target planting pattern and rolls out the parameter template and threshold.

2. A mulberry tree planting pattern adaptive matching system according to claim 1, characterized in that: The mulberry planting model library includes: row spacing schemes, variety or rootstock / scion combinations, mulch / cover configurations, irrigation and drainage strategy templates, pruning and stem setting rules, seedling management strategies and leaf picking intensity upper limits; model entries are mapped to plot microenvironmental zoning and phenological windows.

3. The adaptive matching system for mulberry tree planting patterns according to claim 1, characterized in that: The park coordination node clusters the microenvironment of the plots through plot environmental data and soil type, terrain slope, light period distribution, wind conditions and historical pest and disease information to obtain plot cluster identification; the cluster identification participates in the pattern library retrieval and initial screening of candidate patterns.

4. The adaptive matching system for mulberry tree planting patterns according to claim 1, characterized in that: Based on the plot microenvironment and the phenological window, candidate planting patterns are retrieved from the mulberry planting pattern library, including: the phenological window includes the transplanting seedling slow-down period, green growth, spring shoot rapid growth period and leaf picking recovery period; pattern retrieval and parameter update are only performed within the task set that matches the current window, and parameter changes outside the window are postponed or enter a pending confirmation state.

5. The adaptive matching system for mulberry tree planting patterns according to claim 1, characterized in that: The mulberry tree-specific field indicators include leaf temperature or canopy temperature difference, leaf area index, bud germination rate, root activity indicator, soil moisture / water potential and ground temperature; the park coordination node performs quality verification and consistency check on the mulberry tree-specific field indicators according to the data dictionary, and generates a matching feature set for model evaluation.

6. The adaptive matching system for mulberry tree planting patterns according to claim 1, characterized in that: The feedback variables include the survival rate, plant height or diameter at breast height increase, leaf area index increase within the preset observation period and recovery time after leaf harvesting; When the feedback variable meets the promotion threshold and robustness conditions set by the rule base, the system will improve the ranking of the corresponding pattern entry, otherwise it will lower or freeze its selectable status.

7. The adaptive matching system for mulberry planting patterns according to claim 1, characterized in that: It also includes abnormal ecological load linkage. When the field edge node identifies transplanting stress or abnormal soil water potential and the ecological load coefficient of the target plot reaches the linkage conditions set by the rule base, the park coordination node will inhibit or limit the leaf picking, topdressing or irrigation and drainage sub-tasks related to the plot, and trigger a temporary switch of the mode template mapped to the plot.

Images