Search

CN-121998475-A - Urban green space intelligent monitoring system and method based on Internet of things

CN121998475ACN 121998475 ACN121998475 ACN 121998475ACN-121998475-A

Abstract

The invention discloses an intelligent urban green space monitoring system and method based on the Internet of things, and relates to the field of green space monitoring; the method comprises the steps of taking difference parameters as a basis, screening positive correlation data as an evaluation index, carrying out standardized treatment on the evaluation index, calculating a growth evaluation score, determining a difference threshold value by combining historical data, dividing a growth stage according to the difference threshold value, obtaining data by a weather station, calculating the evapotranspiration quantity, calculating the water demand by combining crop coefficients, drawing a time-dependent change curve of the water demand, carrying out curve fitting by a polynomial algorithm after preprocessing curve data, constructing a water demand prediction model, combining real-time weather and growth stage, determining irrigation quantity according to the model prediction water demand, correcting the model if the irrigation evaluation result is unqualified, and realizing urban green plant growth monitoring and accurate irrigation management.

Inventors

  • LI HAIJUN
  • CAO QIN
  • YU XIAOMIN
  • Jiang Feiya

Assignees

  • 上海交大慧谷信息产业股份有限公司

Dates

Publication Date
20260508
Application Date
20251219

Claims (10)

  1. 1. The urban green space intelligent monitoring method based on the Internet of things is characterized by comprising the following steps of: Step S1, setting a time window, collecting growth data of the same plant in different time periods, generating a growth record table of each time period, and comparing the growth record tables of different time periods; s2, taking the difference parameters obtained by comparison as growth evaluation indexes, calculating growth evaluation scores, and judging the plant growth stage according to the growth evaluation scores; s3, collecting water demand of plants in different growth stages in each time window, and drawing a curve of water demand changing along with time; S4, fitting a time-dependent change curve of the water demand, constructing a water demand prediction model, and predicting the water demand in a next time window of the plant according to the water demand prediction model; And S5, irrigating the plants according to the water demand predicted by the water demand prediction model by combining the real-time meteorological data and the plant growth stage, evaluating the irrigation result, and carrying out model correction according to the result.
  2. 2. The urban green space intelligent monitoring method based on the Internet of things, which is characterized in that the specific steps of the step S1 are as follows: s1-1, setting a time window t, and dividing the growth period of plants into a plurality of continuous time windows; s1-2, recording plant growth data in each time window, and generating a plant growth record table corresponding to the time window; Step S1-3, sorting the plant growth record tables of all the time windows according to a time sequence, traversing and comparing the plant growth record tables of two adjacent time windows, defining the data names as different parameters, and extracting the different parameters.
  3. 3. The urban green space intelligent monitoring method based on the Internet of things, which is characterized in that the specific steps of the step S2 are as follows: s2-1, integrating difference parameters to construct a basic plant growth data set; S2-2, taking plant height increment as a reference variable Y, extracting all data in a plant growth data set as candidate data, setting the candidate data as X, and selecting synchronous observation data of n plants to form a sample pair (X 1 ,Y 1 ),(X 2 ,Y 2 ),...,(X n ,Y n ),(X 1 ,Y 1 ),(X 2 ,Y 2 ),...,(X n ,Y n ) as candidate data and reference variables of 1 st, 2 nd, third and n plants; s2-3, calculating the association direction of the candidate data and the reference variable, wherein the association direction specifically comprises the following steps: ; in the formula, For the pearson correlation coefficient between the ith candidate data and the reference variable Y, n is the number of samples, A specific value of candidate data X i , which is the kth sample, k is the sample index, As the mean value of the candidate data, The reference variable Y value for the kth sample, Is the sample mean of the reference variable Y, For the discrete product sum of the candidate index X and the reference variable Y, Is the sum of squares of the dispersion of the candidate data X, Is the sum of squares of the dispersion of the reference variable Y; Step S2-4, when When the plant growth evaluation index is greater than 0, judging that the candidate index and the plant growth state are positively correlated, and screening positively correlated data from a basic plant growth data set to serve as a plant growth evaluation index; s2-5, carrying out standardization treatment on plant growth evaluation indexes; Step S2-6, calculating the growth evaluation score of each time window of the plant based on the selected plant growth evaluation index, wherein the growth evaluation score is specifically: ; wherein, C 1 、C 2 、...、C n is a growth evaluation index, 、 、...、 The weight allocated to the growth evaluation index is S, which is a growth evaluation score; S2-7, selecting n plants of the same variety, collecting growth evaluation indexes of time windows of complete growth cycles of the n plants, and calculating growth evaluation scores of each time window; S2-8, arranging time windows and scores of each plant according to time sequence, calculating score difference values of each adjacent time window, integrating the difference values of all adjacent time windows of n historical plants, constructing a total difference value set, calculating an arithmetic average value of the set, and taking the arithmetic average value as a difference value threshold value; And S2-9, comparing each difference value with a difference value threshold T according to the adjacent time window score difference value sequence of each historical plant, if the difference value is larger than the difference value threshold, judging that the adjacent time window boundary corresponding to the difference value is a growth stage dividing node, and dividing the growth stage of the whole plant cycle by taking the dividing node of each plant as a boundary.
  4. 4. The urban green space intelligent monitoring method based on the Internet of things, according to claim 3, wherein the specific steps of the step S3 are as follows: S3-1, acquiring meteorological data through a meteorological station, and calculating green land evapotranspiration PET by using a FAO Penman-Montetith formula; step S3-2, collecting water required by each growth stage of the plants in each time window, wherein the water required by each growth stage of the plants in each time window is specifically: ; Wherein ETc is water demand, PET is evapotranspiration, and Kc is crop coefficient; And S3-3, sequencing the time windows, taking time as a horizontal axis and water demand as a vertical axis, and drawing a time-dependent change curve of the water demand according to the time window sequence and the water demand.
  5. 5. The intelligent urban green space monitoring method based on the Internet of things, which is characterized in that the specific steps of the step S4 are as follows: step S4-1, preprocessing data in a time-dependent change curve of the water storage amount, wherein the preprocessing specifically comprises the following steps: for the identification and processing of abnormal values, the box diagram method is used for screening abnormal ETc values, and the quartered bit distance of ETc data is calculated to be exceeded The numerical value of (1) is marked as abnormal, the abnormal value is removed, Q 1 is the first quartile, IQR is the quartile range, and 1.5 is the coefficient for judging the abnormal value in the box diagram method; For missing values, a linear interpolation method is used for filling, specifically: let the missing window be t0, the adjacent active window be t 1 (ETc=y 1 )、t 2 (ETc=y 2 ), ; In the formula, For the water demand deficiency value to be estimated, For the water demand corresponding to the first active time window adjacent to the missing time window, For the time window corresponding to the water demand deficiency value, For the first valid time window adjacent to the corresponding time window of missing values, For the water demand corresponding to the second effective time window adjacent to the time window corresponding to the missing value, A second valid time window adjacent to the time window corresponding to the missing value; s4-2, carrying out normalization processing on the pretreated water demand data; s4-3, fitting each scattered point in a time-dependent change curve of the water demand, and constructing a water demand prediction model by using a polynomial algorithm, wherein the method specifically comprises the following steps: ; in the formula, A n 、a n-1 、...、a 1 、a 0 is a polynomial coefficient, n is the order of a polynomial, and t is a time characteristic independent variable; And S4-4, predicting the water demand in the next time window of the plant according to the water demand prediction model.
  6. 6. The intelligent urban green space monitoring method based on the Internet of things, which is characterized in that the specific steps of the step S5 are as follows: S5-1, collecting plant growth evaluation indexes by using a sensor, calculating plant growth evaluation scores, and judging plant growth stages according to the plant growth evaluation scores; S5-2, acquiring weather data of a current time window and a future time window through a weather station, calculating the green land evapotranspiration and the water demand of the current time window, predicting the water demand of the next time window by using a water demand prediction model, deducting the precipitation of the current stage, and determining the final actual irrigation quantity, wherein the method specifically comprises the following steps of: ; wherein I is the actual irrigation quantity, For the actual water demand of the plant in the current time window, For the predicted water demand for the next time window, Setting a pre-irrigation coefficient according to actual conditions, wherein P is the current natural precipitation; S5-3, collecting growth evaluation scores of n historical adjacent two time windows of normal growth of the same plant, calculating a difference value, setting a fluctuation coefficient x according to an actual service scene by a professional, and setting a threshold value interval by combining the difference value; s5-4, after irrigation is finished and a next time window is entered, collecting plant growth evaluation indexes again, and calculating the growth evaluation score after irrigation; s5-5, calculating a difference value between the post-irrigation growth evaluation score and the pre-irrigation growth evaluation score, judging that the irrigation is qualified when the difference value is in a threshold value interval, and judging that the irrigation is unqualified when the difference value is not in the threshold value interval; S5-6, when judging that irrigation is unqualified, correcting the water demand prediction model, wherein the method specifically comprises the following steps: and re-collecting meteorological data to calculate PET, calculating ETc, re-drawing a water demand time-varying curve, re-fitting the water demand time-varying curve by using a polynomial algorithm, and constructing a new water demand prediction model.
  7. 7. The urban green space intelligent monitoring system based on the Internet of things is characterized by comprising a growth data difference extraction module, a plant growth stage judgment module, a plant water demand curve drawing module, a water demand prediction model construction module and an irrigation execution and model correction module; The growth data difference extraction module is used for setting a time window for collecting plant growth data, recording the data and extracting difference parameters; The plant growth stage judging module is used for calculating scores by taking the difference parameters as evaluation indexes so as to judge the plant growth stage; the plant water demand curve drawing module is used for calculating water demand of each time window of the plant and drawing a water demand time change curve; The water demand prediction model construction module is used for preprocessing water demand data and fitting a curve to construct a model to predict water demand; and the irrigation execution and model correction module is used for executing irrigation after determining the irrigation quantity, evaluating the irrigation result and correcting the water demand prediction model.
  8. 8. The urban green space intelligent monitoring system based on the Internet of things, according to claim 7, wherein the growth data difference extraction module comprises a growth data recording unit and a difference parameter extraction unit; The growth data recording unit is used for dividing a plant growth period into a plurality of time windows and recording data to generate a growth recording table; The difference parameter extraction unit is used for sorting the growth record table according to time, comparing adjacent tables and extracting difference parameters; the plant growth stage judging module comprises a growth evaluation score calculating unit and a growth stage dividing unit; The growth evaluation score calculation unit is used for integrating the difference parameter screening indexes and calculating a growth evaluation score after standardization; the growth stage dividing unit is used for dividing the growth stage according to the historical plant fraction difference value threshold value and the comparison difference value.
  9. 9. The urban green space intelligent monitoring system based on the Internet of things of claim 7, wherein the plant water demand curve drawing module comprises a water demand calculation unit and a water demand curve drawing unit; The water demand calculating unit is used for obtaining meteorological data to calculate the evapotranspiration and calculating the water demand of plants by combining crop coefficients; The water demand curve drawing unit is used for sequencing time windows according to time, and drawing a change curve by taking time and water demand as axes; the water demand prediction model construction module comprises a water demand data preprocessing unit and a water demand prediction model construction unit; the water demand data preprocessing unit is used for recognizing and eliminating water demand abnormal values, filling missing values and carrying out normalization processing; the water demand prediction model construction unit is used for fitting a water demand curve by using a polynomial algorithm and constructing a model to predict the water demand of the next time window.
  10. 10. The intelligent urban green space monitoring system based on the Internet of things of claim 7, wherein the irrigation execution and model correction module comprises an irrigation scheme execution unit and an irrigation result evaluation and model correction unit; The irrigation scheme execution unit is used for judging the plant growth stage and determining the actual irrigation quantity by combining the meteorological data and the predicted water demand; And the irrigation result evaluation and model correction unit is used for evaluating the irrigation qualification, and recalculating the water demand and constructing a new prediction model when the irrigation qualification is not qualified.

Description

Urban green space intelligent monitoring system and method based on Internet of things Technical Field The invention relates to the field of green land monitoring, in particular to an intelligent urban green land monitoring system and method based on the Internet of things. Background In the field of urban green land monitoring, accurate judgment and irrigation as required in the plant growth stage are key links for guaranteeing healthy growth of plants. In the traditional plant growth management, division of growth stages depends on manual observation of plant morphological characteristics, the mode is high in subjectivity and low in quantification degree, growth differences under different time windows are difficult to systematically compare, a standardized 'growth stage-evaluation index' correlation system cannot be established, and stage judgment errors are large, so that accuracy of subsequent management decisions is affected. In the aspect of water demand calculation and irrigation decision, the prior art has obvious limitations that firstly, crop coefficient dynamic change characteristics of plants in different growth stages are not fully combined, water demand calculation and actual physiological demands of the plants are disjointed, secondly, the application of meteorological data is mostly limited to historical static data, the integration of real-time meteorological conditions and meteorological trends in future time periods is lacking, water demand estimation cannot be matched with environment dynamic change, thirdly, data preprocessing links are not standard, and a unified and scientific recognition and processing scheme is lacking aiming at abnormal values and missing values in water demand data, so that the quality and model accuracy of subsequent data are directly affected. In addition, the traditional irrigation amount determination only considers the water demand of the current period, the predicted water demand and the natural precipitation of the next period are not comprehensively formed, excessive or insufficient irrigation is easily caused, and accurate and efficient plant water management is difficult to realize. In order to solve the problems of inaccurate judgment, large deviation between water demand calculation and prediction, lack of dynamic feedback in irrigation decisions and the like in the growth stage in the prior art, a set of systematic technical scheme capable of integrating time window division, quantitative growth evaluation, dynamic water demand calculation and prediction model iterative optimization is needed so as to realize standardization, precision and intellectualization of plant growth management. Disclosure of Invention The invention aims to provide an intelligent urban green space monitoring system and method based on the Internet of things, so as to solve the problems in the prior art. In order to achieve the purpose, the invention provides the technical scheme that the urban green space intelligent monitoring method based on the Internet of things comprises the following steps: Step S1, setting a time window, collecting growth data of the same plant in different time periods, generating a growth record table of each time period, and comparing the growth record tables of different time periods; s1-1, setting a time window t, and dividing the growth period of plants into a plurality of continuous time windows; s1-2, recording plant growth data in each time window, and generating a plant growth record table corresponding to the time window; Step S1-3, sorting the plant growth record tables of all the time windows according to a time sequence, traversing and comparing the plant growth record tables of two adjacent time windows, defining the data names as different parameters, and extracting the different parameters. The growth period is divided through setting a time window, the growth data is recorded, and the difference parameters are compared and extracted, so that the growth data change of the plant in different continuous time periods is obtained, and data support is provided for the determination of the subsequent growth evaluation index. S2, taking the difference parameters obtained by comparison as growth evaluation indexes, calculating growth evaluation scores, and judging the plant growth stage according to the growth evaluation scores; s2-1, integrating difference parameters to construct a basic plant growth data set; S2-2, taking plant height increment as a reference variable Y, extracting all data in a plant growth data set as candidate data, setting the candidate data as X, and selecting synchronous observation data of n plants to form a sample pair (X1,Y1),(X2,Y2),...,(Xn,Yn),(X1,Y1),(X2,Y2),...,(Xn,Yn) as candidate data and reference variables of 1 st, 2 nd, third and n plants; s2-3, calculating the association direction of the candidate data and the reference variable, wherein the association direction specifically comprises the following steps