Landscaping maintenance system and method based on big data

Abstract

The invention relates to the technical field of forestry data processing, in particular to a landscaping maintenance system and method based on big data, and the system comprises a water potential monitoring module, a block division module, a pruning adjustment module, a man-hour scheduling module and an anomaly recognition module. According to the method, by tracking the consistency of soil moisture fluctuation in real time, accurately analyzing the dynamic influence of the root system type and the soil layer structure on water absorption and combining plant evapotranspiration physiological indexes and meteorological condition differences to finely divide water demand grades, efficient matching of water supply and plant physiological statuses is achieved; the pruning period is dynamically adjusted according to the actual change of plant photosynthesis and tissue expansion, the coordination error between maintenance working hour calculation and geographic path planning is effectively reduced, the continuity and timeliness of illumination change abnormal response recognition are improved, the accuracy and refinement level of garden plant maintenance decision making are remarkably improved, and the method is suitable for popularization and application. And the scientific decision-making capability, the space adaptive capability and the intelligent response level of greening maintenance management are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of forestry data processing, and particularly relates to a garden greening maintenance system and method based on big data. BACKGROUND

[0002] The technical field of forestry data processing includes the collection, storage, analysis and utilization of various data information in the forestry ecosystem, mainly for improving the scientificity and refinement level of forestry resource management, including forest resource investigation, forest health monitoring, vegetation growth analysis, disaster warning, forest environment factor monitoring, etc., involving data sources including remote sensing images, sensor data, geographic information system data, artificial collection data, forestry data processing realizes informatization management and decision support in forestry activities through the establishment of data models and information system platforms, forming a complete technical system covering front-end collection, middle-end processing and back-end application, and being applied to forest resource management, ecological restoration, carbon sink assessment, pest monitoring, greening maintenance and other subfields, wherein the garden greening maintenance system based on big data refers to a data management and decision support platform for urban garden greening management, which is constructed by integrating environmental monitoring nodes, greening area GIS map data, plant variety database and maintenance behavior record information, and is used for identifying vegetation growth conditions, setting maintenance frequency, warning pests, scheduling irrigation and fertilization, etc. in garden greening, through deploying Internet of Things monitoring equipment to obtain parameters such as light, temperature and humidity, soil moisture, etc., combining the spatial position of the greening area with the characteristics of the vegetation, using a maintenance strategy model based on rule matching and a database query algorithm, completing the classification management and dynamic maintenance task generation of the greening area, and combining a fixed block-based patrol path planning method to support the partition storage and time sequence analysis of the maintenance data, realizing the generation of daily maintenance operation suggestions and task scheduling for the area.

[0003] The traditional garden greening maintenance technology relies on static rule matching and database query to carry out garden greening area management, and lacks real-time processing capability for the actual dynamic feedback of soil water potential fluctuation and plant growth in plant growth and maintenance task arrangement, resulting in time lag and incoordination between plant maintenance operation frequency and environmental changes, and the technology cannot adapt to the scene of rapid environmental condition change, which reduces the accuracy of maintenance operation, causes waste of water resources, decline of plant health and increase of maintenance cost, etc. SUMMARY

[0004] The purpose of the present application is to solve the problems existing in the prior art, and to provide a garden greening maintenance system and method based on big data.

[0005] In order to achieve the above purpose, the technical scheme adopted by the present application is as follows: the garden greening maintenance system based on big data comprises: The water potential monitoring module divides the garden area into multiple blocks, analyzes the fluctuation consistency of soil moisture change of multiple humidity and water potential sensing nodes in the block, combines the influence of plant root type and soil layer structure on water absorption behavior, detects block water anomaly events, and generates water absorption deviation comparison results; The block division module calls the water absorption deviation comparison results, obtains the leaf area index, unit area transpiration rate, air temperature and wind speed value of the plant sample in each block, calculates the water evaporation intensity of the plant in each block, classifies the blocks, and generates water demand level aggregation sections; The pruning adjustment module utilizes the water demand level aggregation sections, obtains daily plant leaf area and chlorophyll index data in each section, analyzes the consistency and change trend of plant photosynthesis and tissue expansion, calculates the pruning demand intensity of the plant, adjusts the pruning cycle, and generates maintenance cycle adjustment results; The work scheduling module utilizes the maintenance cycle adjustment results, calculates the task amount and time length demand of each section according to the corresponding relationship between plant type and maintenance work density, plans a maintenance task list combined with geographic location information, and generates work task planning records.

[0006] As a further scheme of the present application, the water absorption deviation comparison results are specifically humidity change frequency indicators, water absorption behavior difference coefficients and node consistency judgment results, the water demand level aggregation sections include evaporation intensity distribution ranges, water demand level classification numbers and spatial continuous segment information, the maintenance cycle adjustment results are specifically leaf area growth coordination coefficients, photosynthesis expansion matching degrees and pruning rhythm change cycles, and the work task planning records include plant type work density distribution, task combination arrangement order and path continuity sorting information.

[0007] As a further scheme of the present application, the water potential monitoring module comprises: The garden area division sub-module divides the garden area into multiple blocks, extracts data collected by multiple humidity and water potential sensing nodes, and generates a humidity node spatial distribution table combined with the geographic location coordinates of the nodes; The humidity fluctuation analysis sub-module extracts humidity and water potential node humidity change data sequences in a continuous monitoring period in each block according to the humidity node spatial distribution table, analyzes the correlation and time sequence consistency of change trends between nodes, evaluates the humidity change coordination degree inside each block, and generates a humidity change consistency index; The water anomaly detection sub-module calls the humidity change consistency index, sets a humidity deviation judgment interval for each node combined with the influence of plant root type and soil layer structure on water absorption behavior, screens abnormal data points, calculates the water anomaly degree of each block, detects block water anomaly events, and generates water absorption deviation comparison results.

[0008] As a further scheme of the present application, the block division module comprises: The sample index extraction submodule calls the water absorption offset comparison result to obtain the leaf area index, unit area transpiration rate, air temperature, and wind speed value of the plant sample in the block, and generates a sample index collection table; The water evaporation calculation submodule calculates the water evaporation intensity index of each block according to the leaf area index, transpiration rate normalized value, air temperature normalized value, and wind speed normalized value of each sample according to the sample index collection table; The grade aggregation classification submodule calls the water evaporation intensity index, merges and groups the blocks according to the index value of the adjacent blocks, marks the aggregation number of the grouped blocks, and generates a water demand grade aggregation section.

[0009] As a further scheme of the present application, the pruning adjustment module comprises: The leaf data acquisition submodule collects the daily plant leaf area and chlorophyll index data of the plants in each section using the water demand grade aggregation section, and performs normalization processing on the data to generate a leaf physiological normalized parameter sequence; The growth trend analysis submodule analyzes the consistency and change trend of plant photosynthesis and tissue expansion according to the leaf physiological normalized parameter sequence, according to the change coordination degree and growth trend of the data, to obtain growth coordination trend information; The pruning demand calculation submodule calculates the pruning demand intensity of each section by calling the growth coordination trend information, calculates the pruning demand intensity index of each section, adjusts the pruning cycle of each section, and generates a maintenance cycle adjustment result As a further scheme of the present application, the work hour scheduling module comprises: The task demand quantity calculation submodule calculates the task quantity and duration demand of each section according to the corresponding relationship between each plant type and maintenance work density, combined with the distribution of multiple plants in the section, using the maintenance cycle adjustment result, to obtain task quantity demand data; The path order adjustment submodule calls the task quantity demand data, combines the geographic location information of each block, adjusts the execution order of each section by analyzing the continuity of the path, and obtains an order optimization sequence; The maintenance task planning submodule plans a maintenance task list according to the order optimization sequence, combined with the duration demand and execution order of the maintenance task of each section, and establishes a job task planning record.

[0010] As a further scheme of the present application, the system further comprises: The abnormality identification module calls the job task planning record, collects and compares the variation difference between the sunshine intensity and the leaf color response in the continuous time period, analyzes the abnormal response state of the plant to the light change, and identifies the light response abnormal event according to the persistence of the abnormal response state, and records the starting time of the event, and generates the light response lag information; The light response lag information specifically refers to the sunshine and color response comparison difference value, the response lag continuous time period, and the abnormal state marking node.

[0011] As a further scheme of the present application, the abnormality identification module comprises: The sunshine data collection submodule calls the job task planning record, collects the sunshine intensity data and the leaf color response data in each section in the continuous period, and generates the sunshine response data set; The response difference analysis submodule compares the variation amplitude of the sunshine intensity and the leaf color response in each period according to the sunshine response data set, identifies the difference degree of each period, analyzes the abnormal response state, and obtains the light response abnormal state sequence; The abnormal state recording submodule uses the light response abnormal state sequence, records the starting time and the duration of the abnormal event according to the duration period of the abnormal state, and establishes the light response lag information.

[0012] The garden greening maintenance method based on big data is executed based on the garden greening maintenance system based on big data, and comprises the following steps: S1: The garden area is divided into multiple blocks, the soil moisture change curves of the humidity sensing nodes and the water potential sensing nodes in each block are collected, the synchronization of the soil moisture fluctuations among different nodes in the block is analyzed, the plant root type and the soil layer distribution characteristics are combined, the block water abnormal event is detected, and the water absorption offset comparison result is generated; S2: Based on the water absorption offset comparison result, the leaf area index, the unit area transpiration rate, the air temperature and the wind speed value of the plant sample in each block are collected, the water evaporation intensity of the plant in each block is calculated, the blocks are grouped and graded, and the water demand grade aggregated section is obtained; S3: Based on the water demand grade aggregated section, the daily leaf area index and the leaf green index of the plant in each section are collected, the consistency and the change trend of the plant photosynthesis and the tissue expansion are analyzed, the pruning demand of the plant is quantified, the pruning maintenance period is adjusted, and the maintenance period adjustment result is established; S4: Based on the maintenance cycle adjustment result, the total amount of maintenance work and the predicted operation time of each section are calculated according to the correspondence between the plant type and the maintenance work density, combined with the number distribution of each type of plant in the section, the maintenance work content, and the geographical location, the execution order of the maintenance task is adjusted, the maintenance task list is obtained, and the operation task planning record is obtained; S5: Based on the operation task planning record, the variation difference between the sunshine intensity and the leaf color response in the continuous time period is collected and compared, the abnormal response state of the plant to the light change is analyzed, and the light response abnormal event is identified according to the persistence of the abnormal response state, and the starting time of the event is recorded, and the light response lag information is established.

[0013] Compared with the prior art, the advantages and positive effects of the present application are: In the present application, by tracking the consistency of soil moisture fluctuation in real time, the dynamic influence of root type and soil structure on water absorption is accurately analyzed, the water demand grade is finely divided combined with the physiological index of plant transpiration and the difference of meteorological conditions, the efficient matching of water supply and plant physiological state is realized, the pruning cycle is dynamically adjusted according to the actual change of plant photosynthesis and tissue expansion, the coordination error between maintenance work hour calculation and geographical path planning is effectively reduced, the continuity and timeliness of abnormal response identification of light change are improved, and the accuracy and fine level of landscape plant maintenance decision are significantly improved, and the scientific decision-making ability, spatial adaptability and intelligent response level of green maintenance management are improved. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The system flowchart of the present application is shown in the figure; Figure 2 The water potential monitoring module flowchart of the present application is shown in the figure; Figure 3 The block division module flowchart of the present application is shown in the figure; Figure 4 The pruning adjustment module flowchart of the present application is shown in the figure; Figure 5 The work hour scheduling module flowchart of the present application is shown in the figure; Figure 6 The abnormality identification module flowchart of the present application is shown in the figure. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0016] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation of the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0017] Please refer to Figure 1 The present application provides a technical solution: a garden maintenance system based on big data includes: The water potential monitoring module divides the garden area into multiple blocks, analyzes the fluctuation consistency of the soil moisture change of multiple humidity and water potential sensing nodes in the block, combines the influence of plant root type and soil structure on water absorption behavior, detects block water anomaly events, and generates water absorption offset comparison results; The block division module calls the water absorption offset comparison result, obtains the leaf area index, unit area transpiration rate, air temperature, and wind speed value of the plant sample in the block, calculates the water evaporation intensity of the plant in each block, classifies the blocks, and generates a water demand level aggregation section; The pruning adjustment module uses the water demand level aggregation section to obtain daily plant leaf area and chlorophyll index data in each section, analyzes the consistency and change trend of plant photosynthesis and tissue expansion, calculates the pruning demand intensity of the plant, adjusts the pruning cycle, and generates a maintenance cycle adjustment result; The work scheduling module uses the maintenance cycle adjustment result, calculates the task amount and time length demand of each section according to the corresponding relationship between plant type and maintenance work density, combines geographic location information, plans a maintenance task list, and generates a work task planning record; The abnormality recognition module calls the work task planning record, collects and compares the change difference between the solar intensity and the leaf color response in the continuous time period, analyzes the abnormal response state of the plant to the light change, and according to the persistence of the abnormal response state, identifies the light response abnormal event, and records the starting time of the event, and generates light response lag information.

[0018] The results of water absorption offset comparison include the frequency index of moisture change, the coefficient of water absorption behavior difference, and the node consistency judgment result. The water requirement level aggregation segment includes the evaporation intensity distribution range, the water requirement level classification number, and the spatial continuous segment information. The maintenance cycle adjustment results include the leaf area growth coordination coefficient, the photosynthetic expansion matching degree, and the pruning rhythm change cycle. The operation task planning records include the plant type operation density distribution, the task combination arrangement order, and the path continuity sorting information. The light response lag information specifically refers to the difference value of the sunlight and color response comparison, the response lag continuous time period, and the abnormal state marking node.

[0019] See also Figure 2 , the water potential monitoring module includes: The garden area division submodule divides the garden area into multiple blocks, extracts the data collected by multiple humidity and water potential sensor nodes, and combines the geographical location coordinates of the nodes to generate a humidity node spatial distribution table; First, a GPS positioning instrument is used to measure the overall outline of the garden area to obtain the longitude and latitude coordinates of the area. Then, data processing is performed using GIS software tools. The entire garden area is divided into multiple sub-areas using a regular rectangular grid method. The area of ​​each sub-area is set to 25 square meters and numbered 1, 2, 3, etc. For example, block 1 is located in the southeast corner of the garden, with longitude and latitude coordinates ranging from (31.212345, 121.567890) to (31.212500, 121.568045). And so on. After the above grid division, the humidity preset in the garden soil is matched with the water potential sensor nodes for geographic coordinates. , the latitude and longitude coordinates of the nodes are automatically extracted according to the GIS tool, and the specific blocks where the nodes are located are marked. For example, in block 1, there are humidity sensors numbered A1 and A2 and a water potential sensor B1. Then the node data of A1 (31.212400, 121.567950), A2 (31.212480, 121.568030), and B1 (31.212420, 121.567970) are recorded. After all nodes are located, the complete data set corresponding to the latitude and longitude coordinates and node numbers is exported using the GIS tool. Combined with the numbering rules and node types, the spatial distribution table of humidity nodes is finally generated.

[0020] Table 1 Humidity node spatial distribution table: ; As shown in Table 1, the humidity node spatial distribution table clearly describes the location distribution of each node in the garden space and the block to which it belongs, providing a basis for subsequent monitoring data processing.

[0021] The wetness fluctuation analysis submodule extracts the wetness change data sequence of the humidity and water potential nodes in each block in the continuous monitoring period according to the humidity node spatial distribution table, analyzes the correlation and time sequence consistency of the change trend between the nodes, evaluates the coordination degree of the wetness change in each block, and generates a wetness change consistency index; According to the humidity node spatial distribution table, the humidity and water potential sensor node data of each block in the garden area are continuously collected. Each monitoring node automatically records data once an hour. Assuming that 24 hours is a complete monitoring period, for example, A1 and A2 nodes in block 1 record humidity data at 00:00, 01:00, …, 23:00, respectively, to generate the data sequence of the nodes, such as the A1 node data sequence [32.1%, 31.8%, 31.5%, …, 30.9%], and the A2 node data sequence [32.2%, 32.0%, 31.6%, …, 30.7%]. Then, the correlation between the wetness change data trends of the nodes is calculated. First, the change value of the adjacent time is calculated, such as the humidity change of A1 node from 00:00 to 01:00 is 0.3%. Then, the correlation between the change trends of the two nodes at each time is calculated, such as the correlation coefficient of A1 and A2 nodes in the 24-hour period is calculated using the Pearson correlation coefficient, and the value is 0.93, indicating that the wetness change trends of the two nodes are highly coordinated. Finally, the average value of the correlation coefficients is taken as the wetness change consistency index of each block after the similar method is used to process all block nodes. For example, the wetness change consistency index of block 1 is 0.89. Similarly, the data processing of all blocks is completed, and the wetness change consistency index table is finally generated.

[0022] The water anomaly detection submodule calls the wetness change consistency index, combines the influence of plant root types and soil structure on water absorption behavior, sets a wetness offset judgment interval for each node, filters abnormal data points, and uses the formula: ; The water anomaly degree of each block is calculated, the water anomaly event of the block is detected, and the water absorption offset comparison result is generated. wherein, is the water anomaly degree of the qth block, is the wetness monitoring value of the ath abnormal node, which is obtained by the acquisition data of the humidity or water potential sensor of the node in the current monitoring period, is the wetness offset judgment boundary of the ath node based on the root type and soil structure, which is set by the root type and soil structure parameter query table of the node, is the wetness difference data between the cth pair of adjacent nodes, which is calculated by the difference between the monitoring values of the two adjacent nodes, is the spatial continuity level of the bth abnormal node, which is calculated by analyzing the spatial distance and arrangement direction of the node and surrounding abnormal nodes, is the number of abnormal nodes within the neighborhood range of the bth abnormal node, which is obtained by counting the number of abnormal nodes within a fixed radius range, A is the total number of nodes identified as abnormal in the block, which is obtained by judging whether the moisture deviation exceeds the set judgment boundary and counting the number of nodes that meet the condition, B is the number of spatially aggregated abnormal groups in the block, which is obtained by clustering the continuously distributed abnormal nodes and counting the number of aggregated groups, C is the number of adjacent node pairs in the block for difference calculation, which is obtained by extracting node pairs that meet the distance condition based on spatial layout and counting, a is the index number of the abnormal node, b is the index number of the abnormal node in the abnormal aggregation group, and c is the index number of the adjacent node pair; Based on the moisture change consistency index, for example, the moisture change consistency index of block 1 is 0.89, first query the preset moisture deviation judgment boundary, according to the plant root type (soil root plant) and soil layer structure (sandy loam soil) set the moisture deviation judgment boundary is 28%-33%, then the moisture monitoring value of node A1 in block 1 is 30.9%, A2 node is 30.7%, both nodes are within the limit range, assuming that there is A3 node located in block 2, its monitoring value is 27.0%, which is obviously lower than the lower limit value 28% of the judgment boundary, then the node is defined as an abnormal node, then calculate the moisture difference data between adjacent nodes in block 2, for example, the moisture difference between node A3 (27.0%) and adjacent node B2 (29.2%) is 2.2%, after summing the squares of all adjacent node differences, take the square root, assuming that the calculation result is 3.6%, and evaluate the spatial continuity level, the spatial distance between node A3 and the surrounding abnormal nodes is calculated using Euclidean distance, for example, the distance to the nearest abnormal node is 0.8 meters, then the spatial continuity level is defined as level 2, if the number of abnormal nodes within a fixed radius of 1 meter is 3, then the number of regional aggregation abnormal groups B is 1, the abnormal degree is calculated using the formula In the formula, is the monitoring value of node A3, is the lower limit of the moisture judgment boundary, is the difference between adjacent nodes, is the spatial continuity level, ​​​is an abnormal node number; wherein the water abnormality degree refers to a comprehensive expression of the intensity and concentration trend of the deviation of the soil water state in the plant root zone in the garden block from the normal water absorption behavior, not only reflecting whether the wetting state of a single node is abnormal, but also considering whether these abnormalities have regional aggregation performance, so as to identify whether there is a local water supply imbalance, soil permeability barrier or system irrigation failure problem. In the garden maintenance system, this parameter can be used as a priority level identification basis in task scheduling, to dynamically adjust the inspection frequency, irrigation resource input or control system response rhythm, so that the management behavior focuses on the water imbalance risk significant section. The calculated abnormality degree value is 1.94, then compare this value with the preset abnormality judgment reference value 1.5, higher than the reference value, then confirm that the block has a water abnormality event, generate the water absorption deviation comparison result of the block as "abnormal event occurs", mark the starting time of the event, for example, 23:00 on the same day. The formula accurately quantifies the node abnormality degree through the joint operation of the wetness difference between nodes and the node abnormal aggregation condition, so that the detection result is more sensitive and accurate.

[0023] Please refer to Figure 3 , the block division module comprises: The sample index extraction submodule calls the water absorption deviation comparison result to obtain the leaf area index, unit area transpiration rate, air temperature, and wind speed value of the plant samples in the block, and generates a sample index collection table; The water absorption deviation comparison result is the input basis, and block 2 is taken as an example for specific description. First, call the water absorption deviation comparison result, block 2 is a water abnormality event occurrence area, the system automatically selects and determines the plant sample number of this block from the garden plant database, for example, three plant samples with numbers Z1, Z2 and Z3, then call the specific index data of the plant sample data respectively, and perform data collection and processing operation item by item, wherein the leaf area index (LAI) is obtained by unmanned aerial vehicle remote sensing image recognition and processing, taking Z1 sample as an example, through the plant canopy photo and using image analysis tool to perform pixel analysis and area conversion, the LAI value is 2.4, the unit area transpiration rate is measured by a transpiration instrument, for example, the transpiration rate value of Z1 plant is 4.2 mmol·m -2 ·s -1 , the air temperature index is measured by a temperature sensor in the block, for example, the sensor T1 measures 30.0°C, the wind speed index is measured by a wind speed instrument, for example, the wind speed instrument F1 records the wind speed as 2.3 m·s -1 , after collecting all the plant sample data one by one, use the database management software to collect and form a unified data table, that is, the sample index collection table, to record the specific monitoring index and number of each sample, for subsequent calculation and calling.

[0024] Table 2 sample index collection table: ; As shown in Table 2, the sample index aggregation table clearly embodies the specific parameter data of each sample, providing basic data for the next step of evapotranspiration intensity calculation.

[0025] The water evapotranspiration calculation submodule calls the leaf area index, transpiration rate normalized value, air temperature normalized value, and wind speed normalized value of each sample according to the sample index aggregation table, and uses the formula: ; Calculate the water evapotranspiration intensity index of each block; wherein, is the water evapotranspiration intensity index of block x, is the leaf area index of the nth sample, obtained through remote sensing image or field measurement, is the normalized value of the transpiration rate per unit area of the nth sample, obtained by normalizing the transpiration measurement data, is the normalized value of the air temperature of the nth sample, obtained by normalizing the temperature sensor data, is the normalized value of the wind speed of the nth sample, obtained by normalizing the anemometer measurement data, is the area normalized value of block x, obtained by the ratio of the current block area to the maximum block area, N is the number of samples in the block, obtained by counting the number of all sample points in the block, and n is the sample index number; Based on the sample data of block 2 in Table 2, first, normalize each parameter, the leaf area index normalized value is based on the maximum LAI value of the whole garden 4.0, for example, the leaf area index normalized value of Z1 sample is 2.4 / 4.0=0.6, similarly, Z2 and Z3 are 0.7 and 0.75 respectively, the transpiration rate normalization is based on the maximum measurement value of 6.0 mmol·m -2 ·s -1 in the garden area, then the normalized value of Z1 sample is 4.2 / 6.0=0.7, Z2 and Z3 are 0.77 and 0.83 respectively, the air temperature normalization is based on the maximum air temperature of 38.0°C in the garden, then Z1 is 30.0 / 38.0=0.79, similarly, Z2 and Z3 are 0.80 and 0.82 respectively, the wind speed normalization value is based on the maximum wind speed of 4.0 m·s -1 in the garden history record, then Z1 is 2.3 / 4.0=0.58, similarly, Z2 and Z3 are 0.65 and 0.53 respectively, the block area normalization value is based on the maximum block area of 40 square meters in the whole garden, the area of block 2 is 25 square meters, then the area normalization value is 25 / 40=0.625, put the above normalized data into the formula to calculate the water evapotranspiration intensity index of block 2: ; ; ; ; The water evaporation intensity index reflects the numerical expression of the overall water loss ability of the vegetation community under the adjustment of the current environmental conditions at a specific garden block spatial scale. Specifically, it measures the overall demand and pressure intensity of the plants in the block for water supply. This index not only describes the evaporation output characteristics within a single block, but also ensures that the indices between different blocks can be directly compared through normalization processing. It can be used as a key input factor for subsequent water management, block water demand classification, irrigation scheduling priority division, and other scenarios, helping the management system to focus on spatial areas with higher water demand risks. The water evaporation intensity index of block 2 is 3.5152, which indicates the water evaporation capacity of the plant population in the block under the current environmental conditions, and clearly defines the relative intensity of the block's water demand. The formula considers multiple factors such as leaf area index, transpiration rate, air temperature, wind speed, etc., to obtain a more accurate and comprehensive quantitative indicator of plant evaporation intensity.

[0026] The grade aggregation and classification submodule calls the water evaporation intensity index, merges and groups the blocks according to the index values of adjacent blocks, marks the aggregated number for the grouped blocks, and generates water demand grade aggregated sections; According to the above water evaporation intensity index data, adjacent blocks are merged and classified based on the difference in intensity index, for example, the index of block 2 is 3.5152, the index of adjacent block 1 is 3.2100, and the index of block 3 is 3.8200. First, set the intensity index merging difference threshold to 0.5, which is obtained from historical garden maintenance experience statistics and verified by experiments. Blocks with a difference within 0.5 are merged into the same aggregated section, and blocks with a difference exceeding the threshold are separately classified. The intensity index difference between block 2 and adjacent block 1 is 0.3052, which is less than the threshold 0.5, so they are merged into the same aggregated section. The difference between block 3 and block 2 is 0.3048, which is also less than the threshold, so they form an aggregated section together, marked as aggregated section number G1. In this way, other blocks are continuously evaluated to generate a complete water demand grade aggregated section table.

[0027] Table 3: Water demand grade aggregated section table ; As shown in Table 3, the water demand grade aggregated section table reflects the aggregation of each block combination and the corresponding index interval, facilitating the subsequent development of targeted irrigation and maintenance measures.

[0028] Please refer to Figure 4 , the pruning adjustment module includes: The leaf data acquisition submodule uses the water requirement level aggregation section to collect the daily plant leaf area and leaf index data of plants in each section, and normalizes the data to generate leaf physiological normalized parameter sequences; The water requirement level aggregation section is used as the basis for leaf data collection. Taking aggregation section G1 as an example (including blocks 1, 2, and 3), first, representative plant samples (numbers Z1, Z2, and Z3) are selected, and daily leaf area data of each plant is collected. The leaf area of plant Z1 is measured by a laser leaf area meter every day for three consecutive days, which is 25.0 cm², 25.6 cm², and 26.2 cm², respectively. At the same time, the leaf index is measured every day, and the leaf index of Z1 is 40 SPAD, 42 SPAD, and 43 SPAD, respectively. The obtained data is normalized by taking the maximum leaf area of 32.0 cm² and the maximum leaf index of 50 SPAD as the reference. The normalized data of plant Z1 on the first day is 25.0 / 32.0=0.78 for leaf area and 40 / 50=0.80 for leaf index. The normalized data of the next two days is (0.80, 0.84) and (0.82, 0.86), respectively. After normalizing the data of plants Z2 and Z3, the leaf physiological normalized parameter sequence data is formed in order, which lays the data foundation for growth trend analysis.

[0029] The growth trend analysis submodule analyzes the consistency and change trend of plant photosynthesis and tissue expansion according to the leaf physiological normalized parameter sequence, the change coordination degree of the data, and the growth trend, and obtains the growth coordination trend information; Taking the leaf physiological normalized parameter sequence data as input, the daily normalized growth amount and growth trend of each plant are calculated. Taking plant Z1 as an example, the normalized growth amount of leaf area for two consecutive days is 0.80-0.78=0.02 on the second day and 0.82-0.80=0.02 on the third day. Similarly, the normalized growth amount of leaf index is 0.04 and 0.02, respectively. The data is analyzed for coordination. The specific process is to compare the absolute difference between the growth amount of leaf area and leaf index every day, such as 0.02-0.04|=0.02 on the second day and |0.02-0.02|=0.00 on the third day. The mean value of the difference value represents the coordination degree as (0.02+0.00) / 2=0.01. Further analysis of the overall growth trend shows that if the growth amount before and after is consistent, the coordination is good. Similarly, the coordination and trend of Z2 and Z3 are calculated, and the coordination indexes of each plant are summarized to obtain the overall coordination trend information of section G1, i.e., the average coordination of leaf area and leaf index of each plant is 0.015, indicating that the plants in the section are in a coordinated growth state.

[0030] The pruning demand calculation submodule calls the growth coordination trend information, calculates the pruning demand intensity of each section, and uses the formula: ; The pruning demand intensity index of each section is calculated, the pruning cycle of each section is adjusted, and the maintenance cycle adjustment result is generated; Among them, is the pruning demand intensity index of section k, is the number of plants in section k, is the plant index number in the section, is the normalized growth amount of the leaf area index of the jth plant, which is obtained by dividing the leaf area index change amount of the plant on the consecutive observation day by the maximum leaf area index, is the normalized growth amount of the leaf green index of the jth plant, which is obtained by dividing the leaf green index change amount of the plant on the consecutive observation day by the maximum leaf green index, is the normalized growth rate of the leaf area index of the jth plant, which is obtained by calculating the average of the leaf area index normalized increment of three or more days, is the normalized growth rate of the leaf green index of the jth plant, which is obtained by calculating the average of the leaf green index normalized increment of three or more days, is the section number; According to the growth coordination trend information, taking the aggregated section G1 plant data as an example, taking three plants as an example (Z1, Z2, Z3), the standardized and growth rate indicators of the normalized growth amount difference of each plant are calculated in turn. Taking plant Z1 as an example, the normalized growth amount difference data of the second day is 0.02, and the third day is 0.00. The average rate of leaf area index growth is (0.02+0.02) / 2=0.02, and the leaf green index growth rate is (0.04+0.02) / 2=0.03. The data is brought into the formula: ; ; ; ; ; The pruning demand intensity index is a comprehensive expression of the coordination and growth rate of plant structure expansion and function growth in the section range, and is used to quantify the urgency of the current need for pruning work in the region. The higher the index value, the more active the plant growth trend in the region, the worse the synchronization of structure and function, and the higher the priority of the pruning task. The index can be directly applied to the pruning cycle adjustment, task sorting, and work planning of the intelligent maintenance system, improving the initiative, accuracy, and resource allocation rationality of garden maintenance. The pruning demand intensity index calculation result is 0.0106, and through experimental data verification, the pruning demand index determination standard is set, the index interval 0.01 below is low demand, 0.01-0.02 is medium demand, and 0.02 above is high demand, then the section G1 belongs to the medium demand state, and the pruning cycle is adjusted to every 12 days according to the index, and the maintenance cycle adjustment result is generated: the pruning cycle is adjusted from every 15 days to every 12 days. The formula more accurately quantifies the urgency of plant pruning by comprehensively analyzing the normalized growth difference and growth rate of leaf area index and chlorophyll index, effectively supporting fine management decisions.

[0031] Table 4: Section plant physiological growth data table ; As shown in Table 4, the physiological growth data of plants in the section clearly shows the daily growth of plants, and the data is normalized for easy calculation and evaluation of coordination and growth trend, accurately supporting subsequent pruning demand analysis and maintenance cycle adjustment.

[0032] Referring to Figure 5 , the work hour scheduling module includes: The task demand quantity calculation submodule uses the maintenance cycle adjustment result, according to the corresponding relationship between each plant type and maintenance work density, and combines the distribution of multiple plants in the section, to calculate the task quantity and time length demand of each section, and obtain the task quantity demand data; The maintenance cycle adjustment result is the input basis for calculating the task demand, taking the aggregation section G1 as an example, which includes three plant types, shrubs, trees and lawns, first, according to the relationship between plant type and maintenance work density, the corresponding data is established, taking shrubs as an example, its maintenance density is defined as 0.5 man-hours per square meter, 0.8 man-hours for trees, and 0.3 man-hours for lawns, and the specific area proportion of each plant is recorded, for example, the area of section G1 is 75 square meters, of which the area proportion of shrubs is 40% (30 square meters), trees account for 30% (22.5 square meters), and lawns account for 30% (22.5 square meters), the maintenance man-hours of each plant is calculated by using the product of area proportion and work density, shrubs 30*0.5=15 man-hours, trees 22.5*0.8=18 man-hours, and lawns 22.5*0.3=6.75 man-hours, the total task quantity of the section is obtained by adding the man-hours of the three plants, the task quantity of section G1 is 15+18+6.75=39.75 man-hours, and the daily work quantity standard of each worker is set as 8 man-hours, so the section needs 39.75÷8≈5 workdays, the task quantity demand data of the section is obtained, and the time required for maintenance work and the task scale are determined.

[0033] The path order adjustment submodule calls the task quantity demand data, combines the geographic location information of each block, adjusts the execution order of each section by analyzing the continuity of the path, and obtains the order optimization sequence; The task quantity demand data of section G1 is called, and the actual geographic location information of each block is referred, such as block 1 (latitude and longitude 31.212345, 121.567890), block 2 (31.213000, 121.568500) and block 3 (31.213600, 121.569000), first, the straight-line distance between adjacent blocks is calculated, based on the geographic coordinate distance calculation formula, such as the distance between block 1 and block 2 is 65 meters, and the distance between block 2 and block 3 is 60 meters, according to the principle of short distance first, the path continuity between blocks is compared and analyzed, such as the sequence distance from the starting block (block 1) to the ending block (block 3) is 125 meters, which is less than other sequences (such as block 3 first and block 2 second, a total of 135 meters), the best maintenance path order is determined as block 1→block 2→block 3, and the order optimization sequence is recorded to provide specific path guidance for actual operation route planning.

[0034] The maintenance task planning submodule plans the maintenance task list according to the order optimization sequence, combines the time length demand and execution order of each section maintenance task, and establishes the operation task planning record; The sequence optimization sequence is used to refine the maintenance task arrangement according to the specific task amount of each block and the optimized path, taking section G1 (including blocks 1, 2, and 3) as an example. Based on a total task working day of 5 working days, for example, the first working day is a maintenance task for block 1, with a specific maintenance duration of 15 working hours, 2 workers are arranged, and a total of 16 working hours are completed, with 1 working hour in excess for flexible adjustment. The second and third working days are maintenance tasks for block 2, with a total of 18 working hours, which are completed by 2 workers working 8 hours per day, respectively, with 2 working hours remaining as a reserve. The fourth working day is a block 3 task, which needs to complete 6.75 working hours, and only 1 worker is arranged, with 1.25 working hours remaining for flexible tasks or emergency task reserve. The fifth working day is a comprehensive review and flexible arrangement, and so on. In this way, the daily maintenance working hours, personnel quantity, specific date, and work content of each block are planned in detail, and finally an explicit work task planning record is formed.

[0035] Table 5: Work task planning record table: ; As shown in Table 5, the work task planning record clearly shows the specific task amount, personnel arrangement, and flexible reserve time of each maintenance working day, facilitating task allocation and implementation in actual work.

[0036] Please refer to Figure 6 , the abnormality identification module comprises: The sunshine data acquisition submodule calls the work task planning record to collect the sunshine intensity data and leaf surface color response data in each section within a continuous period, and generates a sunshine response data set; Based on the work task planning record, taking the aggregated section G1 as an example, the work period of 7 consecutive days in the block is determined as the basis for sunshine data collection. The specific implementation is to collect data continuously from 8:00 to 18:00 every day. First, install a sunshine intensity sensor at the center of the section. Record the sunshine intensity every hour, for example, the sunshine intensity measured at 8:00 on the first day is 150 W·m -2 , and the data of the 7th day is [150, 220, 320, 450, 580, 600, 610, 590, 480, 350, 200] W·m -2 At the same time, 3 representative plants (Z1, Z2, Z3) are selected within the same period every day, and the RGB values of the plant leaves are measured every hour using a color analyzer and converted into green indices. For example, the green index value of Z1 leaf at 8:00 on the first day is 0.75, Z2 is 0.76, and Z3 is 0.78. The subsequent data is measured and recorded daily. After 7 days of complete data collection, the sunshine intensity and leaf surface color response data are time-synchronized and aligned, and data matching processing is performed to form a sunshine response data set, which is used for response difference analysis in the next stage.

[0037] The response difference analysis submodule compares the change amplitudes of the sunlight intensity and the leaf color response in each period according to the sunlight response data set, identifies the difference degree of each time period, analyzes the abnormal response state, and obtains the light response abnormal state sequence; According to the sunlight response data set, the change amplitudes between the sunlight intensity and the leaf color response are analyzed day by day and hour by hour. Specifically, the 2nd day and the 3rd day are taken as examples for detailed description. On the 2nd day, the sunlight intensity rises from 220 W·m -2 to 320 W·m -2 , the change amplitude is 100 W·m -2 , the average change value of the leaf green index in the same period rises from 0.76 to 0.79, and the amplitude is 0.03; the proportion of the increase of each unit of sunlight intensity to the change of the leaf green index is calculated as 0.03 / 100=0.0003. On the 3rd day, the sunlight intensity in the same period rises from 320 W·m -2 to 450 W·m -2 , the change amplitude is 130 W·m -2 , but the leaf green index only rises from 0.79 to 0.795, and the amplitude is 0.005; the calculation proportion is 0.005 / 130≈0.0000385, which is obviously smaller than the data of the previous day. Further analysis of the proportion values of the data in each period is performed, and the change proportion threshold is set as 0.0001. The threshold is determined through a large number of experimental data in the early stage. When the proportion value of the period is smaller than the threshold, it is defined as the response abnormal state. For example, the proportion of 0.0000385 on the 3rd day is smaller than the threshold, and is recorded as an abnormal response. In this way, all the data of the continuous 7 days are analyzed hour by hour, the light response abnormal state sequence is formed, the occurrence of the abnormal state in each period is recorded, and the actual response state of the plant to the light change is embodied.

[0038] The abnormal state recording submodule uses the light response abnormal state sequence, records the starting time and the duration of the abnormal event according to the duration period of the abnormal state, and establishes the light reaction lag information; Based on the abnormal state sequence data of light response, the duration of abnormal response state in each period is analyzed, for example, at 9:00 on the 3rd day, the abnormal response state is first detected, and whether the abnormal state continues in the subsequent adjacent period is analyzed, the continuous abnormal state from 10:00 to 11:00 on the 3rd day meets the definition of more than two consecutive periods of abnormal state, and is defined as a persistent abnormal event, taking the first abnormal period (9:00 on the 3rd day) as the starting time point of the abnormal event, recording the starting time and duration of 3 hours; Similarly, the abnormal response is first recorded at 15:00 on the 6th day, and still does not recover to normal at 17:00, which is also marked as an abnormal event and the time is recorded; After the method is processed on all abnormal states in the continuous 7 days, the clear light response lag information sequence is formed, the specific starting time and duration of each abnormal event are recorded, and the lag of plant physiological response and continuous abnormal performance are clear.

[0039] The garden maintenance method based on big data is executed based on the above-mentioned garden maintenance system based on big data, comprising the following steps: S1: divide the garden area into multiple blocks, collect the soil moisture change curves of the humidity sensing nodes and water potential sensing nodes in each block, analyze the synchronicity of soil moisture fluctuation between different nodes in the block, combine the plant root type and soil layer distribution characteristics, detect the water anomaly event of the block, and generate the water absorption offset comparison result; S2: based on the water absorption offset comparison result, collect the leaf area index, unit area transpiration rate, air temperature and wind speed value of the plant sample in each block, calculate the water evaporation intensity of the plant in each block, group and grade the blocks, and obtain the water demand level aggregation section; S3: based on the water demand level aggregation section, collect the daily leaf area index and chlorophyll index of the plants in each section, analyze the consistency and change trend of plant photosynthesis and tissue expansion, quantify the pruning demand of the plants, adjust the pruning maintenance period, and establish the maintenance period adjustment result; S4: based on the maintenance period adjustment result, according to the corresponding relationship between the plant type and the maintenance work density, combining the number distribution of each type of plant in the section and the maintenance work content, calculating the total amount of maintenance work and the expected operation time of each section, combining the geographical location, adjusting the execution order of the maintenance task, the maintenance task list, and obtaining the operation task planning record; S5: based on the operation task planning record, collect and compare the variation difference between the sunshine intensity and the leaf color response in the continuous time period, analyze the abnormal response state of the plant to the light change, and according to the persistence of the abnormal response state, identify the light response abnormal event, and record the starting time of the event, and establish the light response lag information.

[0040] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application in other forms. Any skilled person in the art can modify or change the disclosed technical content into equivalent embodiments with equivalent changes, and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application, without departing from the technical solution content of the present application, still falls within the protection scope of the present application.

Claims

1. A landscaping maintenance system based on big data, characterized in that, The system comprises: The water potential monitoring module divides the garden area into multiple blocks, analyzes the fluctuation consistency of the soil moisture change of multiple humidity and water potential sensing nodes in the block, combines the influence of plant root type and soil structure on water absorption behavior, detects block water anomaly events, and generates water absorption offset comparison results; The block division module calls the water absorption offset comparison results, obtains the leaf area index, unit area transpiration rate, air temperature, and wind speed value of the plant sample in the block, calculates the water evaporation intensity of the plant in each block, classifies the blocks, and generates water demand level aggregation sections; The pruning adjustment module uses the water demand level aggregation sections to obtain daily plant leaf area and chlorophyll index data in each section, analyzes the consistency and trend of plant photosynthesis and tissue expansion, calculates the pruning demand intensity of the plant, adjusts the pruning cycle, and generates maintenance cycle adjustment results; The work scheduling module uses the maintenance cycle adjustment results, calculates the task amount and time length demand of each section according to the corresponding relationship between plant type and maintenance work density, plans a maintenance task list combined with geographic location information, and generates work task planning records.

2. The big data based garden maintenance system according to claim 1, wherein, The water absorption offset comparison results are specifically humidity change frequency indicators, water absorption behavior difference coefficients, and node consistency judgment results. The water demand level aggregation sections include evaporation intensity distribution ranges, water demand level classification numbers, and spatial continuous segment information. The maintenance cycle adjustment results are specifically leaf area growth coordination coefficients, photosynthesis expansion matching degrees, and pruning rhythm change cycles. The work task planning records include plant type work density distribution, task combination arrangement order, and path continuity sorting information. 3.The big data-based garden maintenance system according to claim 1, wherein, The water potential monitoring module comprises: The garden area division submodule divides the garden area into multiple blocks, extracts data collected by multiple humidity and water potential sensing nodes, and generates a humidity node spatial distribution table combined with the geographic location coordinates of the nodes; The humidity fluctuation analysis submodule extracts humidity and water potential node humidity change data sequences in a continuous monitoring period in each block according to the humidity node spatial distribution table, analyzes the correlation and time sequence consistency of the change trend between nodes, evaluates the humidity change coordination degree within each block, and generates a humidity change consistency index; The water anomaly detection submodule calls the humidity change consistency index, sets a humidity offset judgment interval for each node combined with the influence of plant root type and soil structure on water absorption behavior, filters abnormal data points, calculates the water anomaly degree of each block, detects block water anomaly events, and generates water absorption offset comparison results.

4. The big data based garden maintenance system according to claim 3, wherein, The block division module comprises: The sample index extraction submodule calls the water absorption offset comparison results, obtains the leaf area index, unit area transpiration rate, air temperature, and wind speed value of the plant sample in the block, and generates a sample index collection table; The water evaporation calculation submodule calls the leaf area index, transpiration rate normalized value, air temperature normalized value, and wind speed normalized value of each sample according to the sample index collection table, and calculates the water evaporation intensity index of each block. The hierarchical aggregation classification submodule calls the evapotranspiration intensity index, merges and groups the blocks according to the index values of adjacent blocks, marks the aggregated numbers of the grouped blocks, and generates water demand hierarchical aggregation sections.

5. The big data based garden maintenance system according to claim 4, wherein, The pruning adjustment module comprises: The leaf data acquisition submodule collects daily plant leaf area and leaf index data of plants in each section by using the water demand hierarchical aggregation sections, and performs normalization processing on the data to generate leaf physiological normalized parameter sequences; The growth trend analysis submodule analyzes the consistency and change trend of plant photosynthesis and tissue expansion according to the leaf physiological normalized parameter sequences, according to the change coordination degree and growth trend of the data, to obtain growth coordination trend information; The pruning demand calculation submodule calculates the pruning demand intensity of each section by calling the growth coordination trend information, calculates the pruning demand intensity index of each section, adjusts the pruning period of each section, and generates maintenance period adjustment results.

6. The big data based garden maintenance system according to claim 5, wherein, The work hour scheduling module comprises: The task demand calculation submodule calculates the task amount and time length demand of each section by using the maintenance period adjustment results, according to the corresponding relationship between each plant type and maintenance work density, and combining the distribution of multiple plants in the section, to obtain task amount demand data; The path order adjustment submodule adjusts the execution order of each section by calling the task amount demand data, combining the geographic location information of each block, and analyzing the continuity of the path, to obtain a sequence of order optimization; The maintenance task planning submodule plans a maintenance task list according to the sequence of order optimization, combining the time length demand and execution order of the maintenance task of each section, and establishes a job task planning record. 7.The big data-based garden maintenance system according to claim 1, wherein, The system further comprises: The abnormality identification module identifies the abnormality of the light response event according to the abnormality of the light response state, and records the starting time of the event, and generates light response lag information; The light response lag information specifically refers to the comparison difference value of the sunshine and the color response, the response lag continuous time period, and the abnormal state marker node. 8.The big data-based garden maintenance system according to claim 7, wherein, The abnormality identification module comprises: The sunshine data acquisition submodule calls the job task planning record to collect sunshine intensity data and leaf color response data in each section in a continuous period, and generates a sunshine response data set; The response difference analysis submodule compares the change amplitudes of the sunshine intensity and the leaf color response in each period according to the sunshine response data set, identifies the difference degree of each period, analyzes the abnormal response state, and obtains a sequence of light response abnormality states; The abnormality state recording submodule uses the sequence of light response abnormality states to record the starting time and duration of the abnormal event according to the duration period of the abnormal state, and establishes the light response lag information.

9. A method for maintaining landscaping based on big data, characterized in that, The method is used to realize the garden maintenance system based on big data according to any one of claims 1-8, comprising the following steps: S1: divide the garden area into multiple blocks, collect the soil moisture change curve of the humidity sensing node and water potential sensing node in each block, analyze the synchronicity of soil moisture fluctuation between different nodes in the block, combine the plant root type and soil layer distribution characteristics, detect the water anomaly event in the block, generate the water absorption offset comparison result; S2: based on the water absorption offset comparison result, collect the leaf area index, unit area transpiration rate, air temperature and wind speed value of the plant sample in each block, calculate the water evaporation intensity of the plant in each block, group and grade the blocks, and obtain the water demand level aggregation section; S3: based on the water demand level aggregation section, collect the daily leaf area index and chlorophyll index of the plant in each section, analyze the consistency and change trend of plant photosynthesis and tissue expansion, quantify the pruning demand of the plant, adjust the pruning maintenance period, and establish the maintenance period adjustment result; S4: based on the maintenance period adjustment result, according to the corresponding relationship between the plant type and maintenance work density, combining the quantity distribution of each type of plant in the section and the maintenance work content, calculate the total maintenance work amount and the expected operation time of each section, combine the geographical position, adjust the execution order of the maintenance task, the maintenance task list, and obtain the operation task planning record; S5: based on the operation task planning record, collect and compare the change difference between the sunshine intensity and the leaf color response in the continuous time period, analyze the abnormal response state of the plant to the light change, and according to the persistence of the abnormal response state, identify the light response abnormal event, and record the starting time of the event, and establish the light response lag information.

Images