CN118095789A - Cloud computing-based intelligent garden data management method and system - Google Patents

Cloud computing-based intelligent garden data management method and system Download PDF

Info

Publication number
CN118095789A
CN118095789A CN202410487046.4A CN202410487046A CN118095789A CN 118095789 A CN118095789 A CN 118095789A CN 202410487046 A CN202410487046 A CN 202410487046A CN 118095789 A CN118095789 A CN 118095789A
Authority
CN
China
Prior art keywords
sensor
data
garden
deployment
cloud computing
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202410487046.4A
Other languages
Chinese (zh)
Other versions
CN118095789B (en
Inventor
王娟
潘金萍
黄恒军
徐磊
柴洪亮
王健
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tai'an Landscaping Management Service Center
Original Assignee
Tai'an Landscaping Management Service Center
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tai'an Landscaping Management Service Center filed Critical Tai'an Landscaping Management Service Center
Priority to CN202410487046.4A priority Critical patent/CN118095789B/en
Publication of CN118095789A publication Critical patent/CN118095789A/en
Application granted granted Critical
Publication of CN118095789B publication Critical patent/CN118095789B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G06F16/9024—Graphs; Linked lists
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/909—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/251—Fusion techniques of input or preprocessed data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/092—Reinforcement learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Databases & Information Systems (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Educational Administration (AREA)
  • Artificial Intelligence (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Biology (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Primary Health Care (AREA)
  • Computational Linguistics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Library & Information Science (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a cloud computing-based intelligent garden data management method and system, and belongs to the technical field of data processing. The intelligent management method comprises the following steps: dividing a garden space into a plurality of subareas by adopting a Voronoi graph algorithm; according to the division result of the Voronoi graph algorithm, and combining with the actual requirements of gardens, adjusting the initial deployment position of the sensor; analyzing historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of sensor detection data, and defining sub-areas belonging to the same cluster as a 'presumable area'; constructing a reinforcement learning model; applying the trained reinforcement learning model to actual garden management, and dynamically adjusting the sensor layout according to the current garden state and the optimal action output by the model; fusing sensor data acquired in real time; based on complete garden environment data, data analysis, modeling and prediction are carried out through a cloud computing platform, and a scientific garden management strategy is formulated.

Description

Cloud computing-based intelligent garden data management method and system
Technical Field
The application relates to the technical field of data processing, in particular to a cloud computing-based intelligent garden data management method and system.
Background
Afforestation management refers to construction, maintenance and management work of greenbelts, woodlands, parks, scenic spots, nurseries and the like of gardens. Afforestation is an important component of urban construction, and a large number of greening lands are planned in many cities, and various plants and trees are planted. However, achieving accurate management of landscaping remains a challenge. The current practice is to deploy sensors in gardens to collect environmental data to achieve intelligent management. However, too many sensors may cause problems such as complex system, reduced stability, redundant data, and high maintenance cost.
The prior art publication No. CN116777116A provides a cloud computing-based intelligent management method and system for garden data, which divides gardens into a detection area and a verification area, compares and manages the measured value of the verification area with the calculated value from the detection area, and adds a sensor in the area when the accuracy of the verification area is lower than a threshold value. The method can reduce the number of required sensors, improve the stability and reduce the cost.
However, the above prior art mainly relies on a fixed area division strategy, a matching mode of the detection sensor and the verification sensor, and a judgment mechanism based on a preset threshold to decide whether to add the sensor. The scheme lacks adaptability to complex dynamic environments, is difficult to realize accurate and flexible sensor layout, lacks intelligent automatic optimization, and cannot fully utilize cloud computing and artificial intelligence technology to improve management efficiency and accuracy.
In view of the above, we provide a cloud computing-based intelligent management method and system for garden data.
Disclosure of Invention
In order to overcome a series of defects in the prior art, the application aims to provide an intelligent garden data management method based on cloud computing, which comprises the following steps of.
And S1, dividing the garden space into a plurality of subareas by adopting a Voronoi graph algorithm according to the garden scale, the management requirement and the sensor detection range database.
And S2, properly adjusting the initial deployment position of the sensor according to the division result of the Voronoi graph algorithm and the actual requirement of gardens.
And S3, analyzing the historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of the sensor detection data, and defining sub-areas belonging to the same cluster as a 'presumable area'.
And S4, constructing a reinforcement learning model, optimizing a sensor layout scheme on the premise of ensuring the detection quality, and reducing the number of required sensors.
And S5, applying the trained reinforcement learning model to actual garden management, and dynamically adjusting the sensor layout according to the current garden state and the optimal action output by the model.
And S6, fusing the sensor data acquired in real time, wherein the environment information of the presumable area is calculated and supplemented by using the clustering result and the adjacent area data.
And S7, carrying out data analysis, modeling and prediction through a cloud computing platform based on the complete garden environment data, and formulating a scientific garden management strategy.
Further, step S1 includes the following steps.
And S11, collecting geographic boundary, area and topography information of gardens, and detecting range and performance parameters of the sensor.
Step S12, creating a set of virtual sensor location points to represent the deployment location of the sensor based on the data collected in step S11.
And S13, applying a Voronoi graph algorithm, constructing a Voronoi graph and dividing the garden space into a plurality of subareas.
And S14, checking whether the Voronoi diagram covers the whole garden space, ensuring no monitoring blind area, analyzing the size and shape of the Voronoi polygon and the uniformity of sensor distribution, and evaluating the rationality and effectiveness of the virtual deployment scheme.
And S15, when the virtual deployment effect is poor, adjusting the position of the virtual sensor or supplementing the virtual sensor according to the information provided by the Voronoi diagram, and optimizing the deployment scheme until a satisfactory virtual deployment scheme is obtained.
And S16, recording a final Voronoi diagram segmentation result, wherein the final Voronoi diagram segmentation result comprises boundary coordinates of each sub-region and dominant sensor information, and providing a basis for the subsequent actual deployment of the sensor.
Further, step S2 includes the following steps.
And S21, determining important monitoring areas and landscape characteristics of gardens.
And S22, according to the Voronoi diagram division result and in combination with actual garden requirements, properly adjusting the positions of the sensors to determine a final virtual deployment scheme.
And S23, converting the sensor position and quantity information determined in the final virtual deployment scheme into an actual deployment instruction.
And step S24, actually installing and debugging the sensor in the garden according to the actual deployment instruction, and ensuring that the position and the orientation of the sensor are consistent with the virtual deployment scheme.
And S25, after the deployment is completed, verifying whether the actually deployed sensor achieves the expected coverage effect or not through an on-site test and data acquisition mode.
And S26, performing fine adjustment on the individual sensors according to the actual verification result, so as to ensure that at least one sensor in each sub-area is responsible for data acquisition, and all the fine-adjusted sensors are deployed to meet the requirements of garden management and coverage.
Further, step S3 includes the following steps.
Step S31, collecting historical detection data of each sensor in a period of time, and performing cleaning and standardization processing.
Step S32, setting parameters of a DBSCAN clustering algorithm according to the characteristics of the garden environment and the type of the sensor.
Step S33, clustering the preprocessed historical detection data by using a DBSCAN clustering algorithm with set parameters, and dividing the preprocessed historical detection data into a plurality of clusters according to the correlation of the sensor detection data.
Step S34, checking whether the clustering result accords with the spatial distribution characteristics of the garden environment, and if necessary, adjusting DBSCAN parameters for re-clustering.
In step S35, the sub-regions belonging to the same cluster are defined as "presumable regions".
Further, step S4 includes the following steps.
In step S41, a reinforcement learning environment is constructed, the state space includes the current sensor deployment situation and the presumable region division, and the actions executable by the agent include adjusting the sensor deployment and setting parameters of the kriging method.
Step S42, after the intelligent agent selects the action under a certain state, a Kriging model is constructed to predict the subarea of the removed sensor according to the selected Kriging parameters and the sensor data of the currently known subarea, and a predicted value and a predicted variance are obtained.
Step S43, designing a reward function, and reducing the number of sensors on the premise of guaranteeing the detection quality.
And S44, constructing an intelligent body by using a deep Q learning algorithm, and performing a large number of iterative training in a simulation environment to learn a strategy of selecting the optimal action in a given state.
Further, in step S42, the sub-region of the sensor is still reserved as the known point in the reinforcement learning training process, the sub-region of the sensor is removed as the unknown point for prediction, and the prediction value formula is: z++s) =μ+Σ i=1 nλi(Z(si) - μ), where: z(s) is the predicted value of the unknown point s, μ is the global mean or trend model predicted value, Z (s i) is the observed value of the known data point s i, λ i is the weight, solved by the Keli Jin Jitong equation, reflecting the contribution of the known point s i to the predicted value of the unknown point s, and n is the number of known data points used for the speculation.
The prediction variance formula is: σ 2(s)=C(s,s)－∑i=1 nλiC(si, s), wherein σ 2(s) represents the predicted variance at the unknown point s; c (s, s) is the value of the semi-variance function at the unknown point s itself, and C (s i, s) is the semi-variance function value between the known point s i and the unknown point s.
Further, the formula of the bonus function R is: r=w 1Rcoverage+w2Raccuracy－w3Pnum, where R coverage represents whether all missing sub-regions are presumed to be covered; r accuracy represents the overall accuracy of the current speculative task, namely RMSE; p num is a penalty term, related to the number of sensors used; w 1,w2,w3 is a weight coefficient for adjusting the importance of each fractional prize.
Further, for a single unknown point s, the prediction error e is: e=z(s) -z≡s, the root mean square error is: rmse= [1/m Σ j=1 m(Z(sj)－Z^(sj))2]1/2, where m is the number of unknown points, Z (s j) is the actual observation of the jth unknown point, and Z (s j) is the prediction of the jth unknown point.
The application also aims to provide a cloud computing-based intelligent garden data management system which specifically comprises the following modules.
And the sensor detection range database module is used for storing effective detection range data of various sensors and providing basic data support for space division and sensor layout optimization.
And the Voronoi graph algorithm module divides the park space according to the detection range data of the sensor, so that no monitoring blind area exists in the whole area.
The sensor deployment adjustment module is used for carrying out initial deployment layout on the sensors according to the division result of the park space and the actual garden requirements, and ensuring that at least one sensor exists in each sub-area.
And the cluster analysis module is used for analyzing the historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of the sensor data, reasonably setting the parameters of the clustering algorithm, and ensuring that the clustering result can effectively reflect the spatial heterogeneity of the garden environment.
And the reinforcement learning model module is used for constructing a reinforcement learning model, reducing the number of required sensors by adjusting the sensor layout on the premise of ensuring the monitoring quality, and improving the resource utilization efficiency.
And the sensor layout dynamic adjustment module is used for applying the trained reinforcement learning model to the actual gardens, executing corresponding operations of sensor reservation, removal, addition and movement according to the current garden state and the optimal action output by the model, and dynamically optimizing and adjusting the deployment layout of the sensors.
The data fusion and analysis prediction module is used for fusing sensor data acquired in real time, calculating by utilizing a clustering result and adjacent sensor data, supplementing missing environment information, analyzing, modeling and predicting data through the cloud computing platform based on complete garden environment data, and formulating a scientific garden management strategy.
Compared with the prior art, the application has the following beneficial effects.
The intelligent and refined management of garden environment data is realized by comprehensively utilizing a plurality of technical means such as a Voronoi graph algorithm, a DBSCAN clustering algorithm, reinforcement learning and the like; scientific division of garden space is realized through a Voronoi graph algorithm, so that division areas are more flexible and more targeted, and the actual monitoring requirements can be met; by introducing the data clustering technology, understanding of spatial heterogeneity of garden environment is enhanced, definition of a presumption area is more scientific, correlation of sensor detection data is fully utilized, utilization efficiency and accuracy of the data are improved, sensor layout optimization is performed by reinforcement learning, adjustment of environment change and management targets can be dynamically adapted, and finer and intelligent sensor layout decision is realized.
Drawings
Fig. 1 is a flowchart of a cloud computing-based intelligent garden data management method according to an embodiment of the present application.
Fig. 2 is a block diagram of a cloud computing-based intelligent garden data management system according to an embodiment of the present application.
Fig. 3 is a schematic diagram of Voronoi diagram division provided in an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention become more apparent, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some, but not all, embodiments of the invention.
All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiments described below, together with the words of orientation, are exemplary and intended to explain the invention and should not be taken as limiting the invention.
In one broad embodiment of the invention, a cloud computing-based intelligent management method for garden data comprises the following steps.
And S1, dividing the garden space into a plurality of subareas by adopting a Voronoi graph algorithm according to the garden scale, the management requirement and the sensor detection range database.
And S2, properly adjusting the initial deployment position of the sensor according to the division result of the Voronoi graph algorithm and the actual requirement of gardens.
And S3, analyzing the historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of the sensor detection data, and defining sub-areas belonging to the same cluster as a 'presumable area'.
And S4, constructing a reinforcement learning model, optimizing a sensor layout scheme on the premise of ensuring the detection quality, and reducing the number of required sensors.
And S5, applying the trained reinforcement learning model to actual garden management, and dynamically adjusting the sensor layout according to the current garden state and the optimal action output by the model.
And S6, fusing the sensor data acquired in real time, wherein the environment information of the presumable area is calculated and supplemented by using the clustering result and the adjacent area data.
And S7, carrying out data analysis, modeling and prediction through a cloud computing platform based on the complete garden environment data, and formulating a scientific garden management strategy.
Further, step S1 includes the following steps.
And S11, collecting geographic boundary, area and topography information of gardens, and detecting range and performance parameters of the sensor.
Step S12, creating a set of virtual sensor location points to represent the deployment location of the sensor based on the data collected in step S11.
And S13, applying a Voronoi graph algorithm, constructing a Voronoi graph and dividing the garden space into a plurality of subareas.
And S14, checking whether the Voronoi diagram covers the whole garden space, ensuring no monitoring blind area, analyzing the size and shape of the Voronoi polygon and the uniformity of sensor distribution, and evaluating the rationality and effectiveness of the virtual deployment scheme.
And S15, when the virtual deployment effect is poor, adjusting the position of the virtual sensor or supplementing the virtual sensor according to the information provided by the Voronoi diagram, and optimizing the deployment scheme until a satisfactory virtual deployment scheme is obtained.
And S16, recording a final Voronoi diagram segmentation result, wherein the final Voronoi diagram segmentation result comprises boundary coordinates of each sub-region and dominant sensor information, and providing a basis for the subsequent actual deployment of the sensor.
Further, step S2 includes the following steps.
And S21, determining important monitoring areas and landscape characteristics of gardens.
And S22, according to the Voronoi diagram division result and in combination with actual garden requirements, properly adjusting the positions of the sensors to determine a final virtual deployment scheme.
And S23, converting the sensor position and quantity information determined in the final virtual deployment scheme into an actual deployment instruction.
And step S24, actually installing and debugging the sensor in the garden according to the actual deployment instruction, and ensuring that the position and the orientation of the sensor are consistent with the virtual deployment scheme.
And S25, after the deployment is completed, verifying whether the actually deployed sensor achieves the expected coverage effect or not through an on-site test and data acquisition mode.
And S26, performing fine adjustment on the individual sensors according to the actual verification result, so as to ensure that at least one sensor in each sub-area is responsible for data acquisition, and all the fine-adjusted sensors are deployed to meet the requirements of garden management and coverage.
Further, step S3 includes the following steps.
Step S31, collecting historical detection data of each sensor in a period of time, and performing cleaning and standardization processing.
Step S32, setting parameters of a DBSCAN clustering algorithm according to the characteristics of the garden environment and the type of the sensor.
Step S33, clustering the preprocessed historical detection data by using a DBSCAN clustering algorithm with set parameters, and dividing the preprocessed historical detection data into a plurality of clusters according to the correlation of the sensor detection data.
Step S34, checking whether the clustering result accords with the spatial distribution characteristics of the garden environment, and if necessary, adjusting DBSCAN parameters for re-clustering.
In step S35, the sub-regions belonging to the same cluster are defined as "presumable regions".
Further, step S4 includes the following steps.
In step S41, a reinforcement learning environment is constructed, the state space includes the current sensor deployment situation and the presumable region division, and the actions executable by the agent include adjusting the sensor deployment and setting parameters of the kriging method.
Step S42, after the intelligent agent selects the action under a certain state, a Kriging model is constructed to predict the subarea of the removed sensor according to the selected Kriging parameters and the sensor data of the currently known subarea, and a predicted value and a predicted variance are obtained.
Step S43, designing a reward function, and reducing the number of sensors on the premise of guaranteeing the detection quality.
And S44, constructing an intelligent body by using a deep Q learning algorithm, and performing a large number of iterative training in a simulation environment to learn a strategy of selecting the optimal action in a given state.
Further, in step S42, the sub-region of the sensor is still reserved as the known point in the reinforcement learning training process, the sub-region of the sensor is removed as the unknown point for prediction, and the prediction value formula is: z++s) =μ+Σ i=1 nλi(Z(si) - μ), where: z(s) is the predicted value of the unknown point s, μ is the global mean or trend model predicted value, Z (s i) is the observed value of the known data point s i, λ i is the weight, solved by the Keli Jin Jitong equation, reflecting the contribution of the known point s i to the predicted value of the unknown point s, and n is the number of known data points used for the speculation.
The prediction variance formula is: σ 2(s)=C(s,s)－∑i=1 nλiC(si, s), wherein σ 2(s) represents the predicted variance at the unknown point s; c (s, s) is the value of the semi-variance function at the unknown point s itself, and C (s i, s) is the semi-variance function value between the known point s i and the unknown point s.
Further, the formula of the bonus function R is: r=w 1Rcoverage+w2Raccuracy－w3Pnum, where R coverage represents whether all missing sub-regions are presumed to be covered; r accuracy represents the overall accuracy of the current speculative task, namely RMSE; p num is a penalty term, related to the number of sensors used; w 1,w2,w3 is a weight coefficient for adjusting the importance of each fractional prize.
Further, for a single unknown point s, the prediction error e is: e=z(s) -z≡s, the root mean square error is: rmse= [1/m Σ j=1 m(Z(sj)－Z^(sj))2]1/2, where m is the number of unknown points, Z (s j) is the actual observation of the jth unknown point, and Z (s j) is the prediction of the jth unknown point.
A cloud computing-based intelligent garden data management system is used for realizing the cloud computing-based intelligent garden data management method and comprises the following steps.
And the sensor detection range database module is used for storing effective detection range data of various sensors and providing basic data support for space division and sensor layout optimization.
And the Voronoi graph algorithm module divides the park space according to the detection range data of the sensor, so that no monitoring blind area exists in the whole area.
The sensor deployment adjustment module is used for carrying out initial deployment layout on the sensors according to the division result of the park space and the actual garden requirements, and ensuring that at least one sensor exists in each sub-area.
And the cluster analysis module is used for analyzing the historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of the sensor data, reasonably setting the parameters of the clustering algorithm, and ensuring that the clustering result can effectively reflect the spatial heterogeneity of the garden environment.
And the reinforcement learning model module is used for constructing a reinforcement learning model, reducing the number of required sensors by adjusting the sensor layout on the premise of ensuring the monitoring quality, and improving the resource utilization efficiency.
And the sensor layout dynamic adjustment module is used for applying the trained reinforcement learning model to the actual gardens, executing corresponding operations of sensor reservation, removal, addition and movement according to the current garden state and the optimal action output by the model, and dynamically optimizing and adjusting the deployment layout of the sensors.
The data fusion and analysis prediction module is used for fusing sensor data acquired in real time, calculating by utilizing a clustering result and adjacent sensor data, supplementing missing environment information, analyzing, modeling and predicting data through the cloud computing platform based on complete garden environment data, and formulating a scientific garden management strategy.
The invention will be described in further detail below with reference to the attached drawings, which illustrate preferred embodiments of the invention.
As shown in fig. 1, the intelligent management method for garden data based on cloud computing comprises the following steps.
And S1, dividing the garden space into a plurality of subareas by adopting a Voronoi graph algorithm according to the garden scale, the management requirement and the sensor detection range database, wherein each subarea is covered by a sensor master in the subarea, so that the whole garden space is ensured to have no monitoring blind area.
And S2, properly adjusting the initial deployment position of the sensor according to the division result of the Voronoi graph algorithm and the actual requirement of gardens, and ensuring that at least one sensor exists in each sub-area.
And S3, analyzing historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of sensor detection data, reasonably setting parameters of the clustering algorithm based on garden environment characteristics and sensor types, ensuring that a clustering result effectively reflects the spatial heterogeneity of the garden environment, and defining subareas belonging to the same cluster as a 'supposedly area'.
And S4, constructing a reinforcement learning model, optimizing a sensor layout scheme on the premise of ensuring the detection quality, and reducing the number of required sensors.
And S5, applying the trained reinforcement learning model to actual garden management, executing corresponding operations of sensor reservation, removal, addition and movement according to the current garden state and the optimal action output by the model, and dynamically adjusting the sensor layout.
And S6, fusing the sensor data acquired in real time, and calculating a presumable area by using a clustering result and adjacent sensor data to supplement missing environment information.
And S7, carrying out data analysis, modeling and prediction through a cloud computing platform based on the complete garden environment data, and formulating a scientific garden management strategy.
Further, step S1 includes the following steps.
And S11, collecting geographic boundary, area and topography information of gardens, and detecting range and performance parameters of the sensor.
Step S12, creating a set of virtual sensor location points to represent the deployment location of the sensor based on the data collected in step S11.
And S13, applying a Voronoi graph algorithm, constructing a Voronoi graph by taking points in the virtual sensor position point set as vertexes, and dividing the garden space into a plurality of subareas.
And S14, checking whether the Voronoi diagram covers the whole garden space, ensuring no monitoring blind area, analyzing the size and shape of the Voronoi polygon and the uniformity of sensor distribution, and evaluating the rationality and effectiveness of the virtual deployment scheme.
And S15, when the virtual deployment effect is poor, adjusting the position of the virtual sensor or supplementing the virtual sensor according to the information provided by the Voronoi diagram, and optimizing the deployment scheme until a satisfactory virtual deployment scheme is obtained.
And S16, recording a final Voronoi diagram segmentation result, wherein the final Voronoi diagram segmentation result comprises boundary coordinates of each sub-region and dominant sensor information, and providing a basis for the subsequent actual deployment of the sensor.
Fig. 3 shows a Voronoi diagram division diagram.
Further, step S2 includes the following steps.
And S21, determining important monitoring areas and landscape characteristics of gardens.
And S22, according to the Voronoi diagram division result and in combination with actual garden requirements, properly adjusting the positions of the sensors to determine a final virtual deployment scheme.
Step S23, converting the sensor position and quantity information determined in the virtual deployment scheme into an actual deployment instruction.
And step S24, actually installing and debugging the sensor in the garden according to the actual deployment instruction, and ensuring that the position and the orientation of the sensor are consistent with the virtual deployment scheme.
And S25, after the deployment is completed, verifying whether the actually deployed sensor achieves the expected coverage effect or not through an on-site test and data acquisition mode.
And S26, performing fine adjustment on individual sensors according to an actual verification result, so as to ensure that at least one sensor in each Voronoi subarea is responsible for data acquisition, and all the sensor deployment after fine adjustment meets the requirements of garden management and coverage.
Further, step S3 includes the following steps.
Step S31, collecting historical detection data of each sensor in a period of time, and performing cleaning and standardization processing to prepare for subsequent cluster analysis.
And step S32, reasonably setting parameters of a DBSCAN clustering algorithm according to the characteristics of the garden environment and the type of the sensor.
Step S33, clustering the preprocessed historical detection data by using a DBSCAN clustering algorithm with set parameters, and dividing the preprocessed historical detection data into a plurality of clusters according to the correlation of the sensor detection data.
Step S34, checking whether the clustering result accords with the spatial distribution characteristics of the garden environment, and if necessary, adjusting DBSCAN parameters for re-clustering.
In step S35, the sub-regions belonging to the same cluster are defined as "presumable regions". In the pushable region, the environmental conditions within each sub-region can be used to infer the environmental status of an unknown sub-region using known sensor data of one region.
Further, step S4 includes the following steps.
In step S41, a reinforcement learning environment is constructed, the state includes the current sensor deployment situation and the presumable region division, and the actions executable by the agent include adjusting the sensor deployment and setting parameters of the kriging method.
Step S42, after the intelligent agent selects the action under a certain state, a Kriging model is constructed to predict the subarea of the removed sensor according to the selected parameters of the Kriging method and the sensor data of the currently known subarea, and a predicted value and a predicted variance are obtained.
And step S43, designing a reward function, and encouraging the model to reduce the number of sensors on the premise of ensuring the detection quality.
And S44, constructing an intelligent body by using a deep Q learning algorithm, and performing a large number of iterative training in a simulation environment to learn a strategy of selecting the optimal action in a given state.
In step S42, the agent needs to predict the environmental status of the removal area after selecting the removal sensor. The method comprises the steps of constructing a Kriging model, utilizing sensor data of a reserved area, predicting an environment value of a removed subarea in a presumable area, constructing the Kriging model through data preparation, variation function modeling and Kriging Jin Gongshi solving steps, predicting the subarea of the removal action by applying the constructed Kriging model to obtain a predicted value and a predicted variance, and predicting the subarea of the sensor as a known point in the reinforcement learning training process, wherein the subarea of the removal sensor is used as an unknown point, and a predicted value formula is as follows: z++s) =μ+Σ i=1 nλi(Z(si) - μ), where: z(s) is the predicted value of the unknown point s, μ is the global mean (for ordinary kriging) or trend model predictor (for panthering), Z (s i) is the observed value of the known data point s i, λ i is the weight, solved by the kriging Jin Jitong equation, reflecting the contribution of the known point s i to the unknown point s predictor, n is the number of known data points used for the speculation, and the prediction variance formula is: σ 2(s)=C(s,s)－∑i=1 nλiC(si, s), wherein σ 2(s) represents the predicted variance at the unknown point s; c (s, s) is the value of the semi-variance function at the unknown point s itself, and C (s i, s) is the semi-variance function value between the known point s i and the unknown point s. The true value of the sub-region is obtained by a sensor deployed at an unknown point, a prediction error and a Root Mean Square Error (RMSE) are calculated, and for a single unknown point s, the prediction error e is: e=z(s) -z≡s, the Root Mean Square Error (RMSE) is: rmse= [1/m Σ j=1 m(Z(sj)－Z^(sj))2]1/2, where m is the number of unknown points, Z (s j) is the actual observation of the jth unknown point, and Z (s j) is the prediction of the jth unknown point.
Taking Root Mean Square Error (RMSE) as an evaluation index of the presumption accuracy, constructing a reward function R as follows: r=w 1Rcoverage+w2Raccuracy－w3Pnum, where R coverage represents whether all missing sub-regions are presumed to be covered; r accuracy represents the overall accuracy of the current speculative task, namely RMSE; p num is a penalty term, related to the number of sensors used; w 1,w2,w3 is a weight coefficient for adjusting the importance of each fractional prize.
In the reinforcement learning process, the intelligent agent continuously tries different actions, and updates the strategy according to the obtained reward signals so as to learn the optimal strategy with the least sensors on the premise of ensuring the estimation accuracy, and the specific flow is as follows.
An initial kriging model is constructed, and data speculation is performed on an unknown region based on existing sensor observation data.
Comparing the estimated result with the actual observed value, and calculating the estimated accuracy as an important component of the reward function.
The agent adjusts the strategy according to the reward function, including optimizing the kriging model parameters and redeploying the location and number of sensors.
The above procedure is repeated by continually increasing the accuracy of the speculation and reducing the number of sensors required until an optimal strategy is obtained that uses the least sensors under the given accuracy constraints.
The reinforcement learning process aims to achieve the following two goals: the accuracy of spatial data speculation is maximized while minimizing the cost of data acquisition (number of sensors). Optimizing the kriging model parameters and sensor deployment locations is critical to achieving these two goals.
As shown in fig. 2, the intelligent management system for garden data based on cloud computing specifically comprises the following modules.
And the sensor detection range database module is used for storing effective detection range data of various sensors and providing basic data support for space division and sensor layout optimization.
And the Voronoi graph algorithm module divides the park space according to the detection range data of the sensor, so that no monitoring blind area exists in the whole area.
The sensor deployment adjustment module is used for carrying out initial deployment layout on the sensors according to the division result of the park space and the actual garden requirements, and ensuring that at least one sensor exists in each sub-area.
And the cluster analysis module is used for analyzing the historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of the sensor data, reasonably setting the parameters of the clustering algorithm, and ensuring that the clustering result can effectively reflect the spatial heterogeneity of the garden environment.
And the reinforcement learning model module is used for constructing a reinforcement learning model, reducing the number of required sensors by adjusting the sensor layout on the premise of ensuring the monitoring quality, and improving the resource utilization efficiency.
And the sensor layout dynamic adjustment module is used for applying the trained reinforcement learning model to the actual gardens, executing corresponding operations of sensor reservation, removal, addition and movement according to the current garden state and the optimal action output by the model, and dynamically optimizing and adjusting the deployment layout of the sensors.
The data fusion and analysis prediction module is used for fusing sensor data acquired in real time, calculating by utilizing a clustering result and adjacent sensor data, supplementing missing environment information, analyzing, modeling and predicting data through the cloud computing platform based on complete garden environment data, and formulating a scientific garden management strategy.

Claims (10)

1. A cloud computing-based intelligent garden data management method is characterized by comprising the following steps:
Step S1, dividing a garden space into a plurality of subareas by adopting a Voronoi graph algorithm according to a garden scale, management requirements and a sensor detection range database;
Step S2, according to the division result of the Voronoi graph algorithm, and combining with actual requirements of gardens, adjusting the initial deployment position of the sensor;
Step S3, analyzing the historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the correlation of the sensor detection data, and defining sub-areas belonging to the same cluster as a 'presumable area';
s4, constructing a reinforcement learning model, optimizing a sensor layout scheme on the premise of guaranteeing detection quality, and reducing the number of required sensors;
step S5, applying the trained reinforcement learning model to actual garden management, and dynamically adjusting the sensor layout according to the current garden state and the optimal action output by the model;
S6, fusing sensor data acquired in real time, wherein the environmental information of the presumable area is calculated and supplemented by using a clustering result and adjacent area data;
and S7, carrying out data analysis, modeling and prediction through a cloud computing platform based on the complete garden environment data, and formulating a scientific garden management strategy.
2. The intelligent garden data management method based on cloud computing as claimed in claim 1, wherein the step S1 comprises the following steps of;
step S11, collecting geographic boundary, area and topography information of gardens, and detecting range and performance parameters of sensors;
Step S12, creating a virtual sensor position point set to represent the deployment position of the sensor based on the data collected in the step S11;
step S13, a Voronoi graph algorithm is applied, a Voronoi graph is constructed, and a garden space is divided into a plurality of subareas;
Step S14, checking whether the Voronoi diagram covers the whole garden space, ensuring no monitoring blind area, analyzing the size and shape of the Voronoi polygon and the uniformity of sensor distribution, and evaluating the rationality and effectiveness of the virtual deployment scheme;
Step S15, when the virtual deployment effect is poor, according to the information provided by the Voronoi diagram, adjusting the position of the virtual sensor or supplementing the virtual sensor, and optimizing the deployment scheme until a satisfactory virtual deployment scheme is obtained;
And S16, recording a final Voronoi diagram segmentation result, wherein the final Voronoi diagram segmentation result comprises boundary coordinates of each sub-region and dominant sensor information, and providing a basis for the subsequent actual deployment of the sensor.
3. The intelligent garden data management method based on cloud computing as claimed in claim 1, wherein the step S2 comprises the following steps;
Step S21, defining important monitoring areas and landscape characteristics of gardens;
Step S22, according to the Voronoi diagram division result and in combination with actual garden requirements, the position of the sensor is adjusted to determine a final virtual deployment scheme;
Step S23, converting the position and quantity information of the sensors determined in the final virtual deployment scheme into an actual deployment instruction;
step S24, actually installing and debugging the sensor in gardens according to the actual deployment instruction, and ensuring that the position and the orientation of the sensor are consistent with the virtual deployment scheme;
Step S25, after deployment is completed, verifying whether the actually deployed sensor achieves the expected coverage effect or not through an on-site test and data acquisition mode;
and S26, performing fine adjustment on the individual sensors according to the actual verification result, so as to ensure that at least one sensor in each sub-area is responsible for data acquisition, and all the fine-adjusted sensors are deployed to meet the requirements of garden management and coverage.
4. The intelligent garden data management method based on cloud computing as claimed in claim 1, wherein the step S3 comprises the following steps;
step S31, collecting historical detection data of each sensor in a period of time, and performing cleaning and standardization processing;
Step S32, setting parameters of a DBSCAN clustering algorithm according to the characteristics of the garden environment and the type of the sensor;
Step S33, clustering the preprocessed historical detection data by using a DBSCAN clustering algorithm with set parameters, and dividing the preprocessed historical detection data into a plurality of clusters according to the correlation of the sensor detection data;
Step S34, checking whether the clustering result accords with the spatial distribution characteristics of the garden environment;
in step S35, the sub-regions belonging to the same cluster are defined as "presumable regions".
5. The intelligent garden data management method based on cloud computing as claimed in claim 1, wherein the step S4 comprises the following steps;
step S41, constructing a reinforcement learning environment, wherein a state space comprises the current sensor deployment condition and the presumable region division, and the actions executable by an intelligent agent comprise the steps of adjusting the sensor deployment and setting parameters of a Kriging method;
Step S42, after the intelligent agent selects actions in a certain state, a Kriging model is constructed to predict the subareas of which the sensors are removed according to the selected Kriging parameters and the sensor data of the currently known subareas, so as to obtain a predicted value and a predicted variance;
step S43, designing a reward function, and reducing the number of sensors on the premise of ensuring the detection quality;
And S44, constructing an intelligent body by using a deep Q learning algorithm, and performing a large number of iterative training in a simulation environment to learn a strategy of selecting the optimal action in a given state.
6. The intelligent management method for garden data based on cloud computing according to claim 5, wherein in step S42, a sub-region of the sensor is still reserved as a known point in the reinforcement learning training process, the sub-region of the sensor is removed as the unknown point for prediction, and the prediction value formula is: z++s) =μ+Σ i=1 nλi(Z(si) - μ), where: z(s) is the predicted value of the unknown point s, μ is the global mean or trend model predicted value, Z (s i) is the observed value of the known data point s i, λ i is the weight, solved by the Keli Jin Jitong equation, reflecting the contribution of the known point s i to the predicted value of the unknown point s, and n is the number of known data points used for the speculation.
7. The intelligent management method for garden data based on cloud computing as claimed in claim 6, wherein in step S42, the prediction variance formula is: σ 2(s)=C(s,s)－∑i=1 nλiC(si, s), wherein σ 2(s) represents the predicted variance at the unknown point s; c (s, s) is the value of the semi-variance function at the unknown point s itself, and C (s i, s) is the semi-variance function value between the known point s i and the unknown point s.
8. The intelligent management method for garden data based on cloud computing as claimed in claim 7, wherein the formula of the reward function R is: r=w 1Rcoverage+w2Raccuracy－w3Pnum, where R coverage represents whether all missing sub-regions are presumed to be covered; r accuracy represents the overall accuracy of the current speculation task; p num is a penalty term, related to the number of sensors used; w 1,w2,w3 is a weight coefficient for adjusting the importance of each fractional prize.
9. The intelligent management method for garden data based on cloud computing as claimed in claim 8, wherein for a single unknown point s, the prediction error e is: e=z(s) -z≡s, the root mean square error is: rmse= [1/m Σ j=1 m(Z(sj)－Z^(sj))2]1/2, where m is the number of unknown points, Z (s j) is the actual observation of the jth unknown point, and Z (s j) is the prediction of the jth unknown point.
10. A cloud computing-based intelligent management system for garden data, configured to implement the intelligent management method for garden data according to any one of claims 1 to 9, comprising:
The sensor detection range database module is used for storing effective detection range data of various sensors and providing basic data support for space division and sensor layout optimization;
The Voronoi graph algorithm module divides the park space according to the detection range data of the sensor, and ensures that the whole area has no monitoring blind area;
The sensor deployment adjustment module is used for carrying out initial deployment layout on the sensors according to the division result of the park space and the actual garden requirements, so that at least one sensor is ensured in each sub-area;
the cluster analysis module is used for analyzing the historical environment data by adopting a DBSCAN clustering algorithm, dividing the historical environment data into a plurality of clusters according to the relativity of the sensor data, reasonably setting parameters of the clustering algorithm, and ensuring that the clustering result can effectively reflect the spatial heterogeneity of the garden environment;
the reinforcement learning model module is used for constructing a reinforcement learning model, reducing the number of required sensors by adjusting the sensor layout on the premise of ensuring the monitoring quality, and improving the resource utilization efficiency;
The sensor layout dynamic adjustment module is used for applying the trained reinforcement learning model to the actual gardens, executing corresponding operations of sensor reservation, removal, addition and movement according to the current garden state and the optimal action output by the model, and dynamically optimizing and adjusting the deployment layout of the sensors;
The data fusion and analysis prediction module is used for fusing sensor data acquired in real time, calculating by utilizing a clustering result and adjacent sensor data, supplementing missing environment information, analyzing, modeling and predicting data through the cloud computing platform based on complete garden environment data, and formulating a scientific garden management strategy.
CN202410487046.4A 2024-04-23 2024-04-23 Cloud computing-based intelligent garden data management method and system Active CN118095789B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410487046.4A CN118095789B (en) 2024-04-23 2024-04-23 Cloud computing-based intelligent garden data management method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410487046.4A CN118095789B (en) 2024-04-23 2024-04-23 Cloud computing-based intelligent garden data management method and system

Publications (2)

Publication Number Publication Date
CN118095789A true CN118095789A (en) 2024-05-28
CN118095789B CN118095789B (en) 2024-07-23

Family

ID=91149543

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410487046.4A Active CN118095789B (en) 2024-04-23 2024-04-23 Cloud computing-based intelligent garden data management method and system

Country Status (1)

Country Link
CN (1) CN118095789B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118261401A (en) * 2024-05-29 2024-06-28 福建拓海建设工程有限公司 Intelligent management system for landscaping engineering project

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060190224A1 (en) * 2005-02-24 2006-08-24 International Business Machines Corporation Integrated circuit layout critical area determination using voronoi diagrams and shape biasing
CN115423148A (en) * 2022-07-29 2022-12-02 江苏大学 Agricultural machinery operation performance prediction method and device based on kriging method and decision tree
CN116299704A (en) * 2023-03-20 2023-06-23 四川西南交大铁路发展股份有限公司 Earthquake single-unit positioning correction method, system, storage medium and terminal
CN116777116A (en) * 2023-06-29 2023-09-19 金乡县园林绿化服务中心 Cloud computing-based intelligent garden data management method and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060190224A1 (en) * 2005-02-24 2006-08-24 International Business Machines Corporation Integrated circuit layout critical area determination using voronoi diagrams and shape biasing
CN115423148A (en) * 2022-07-29 2022-12-02 江苏大学 Agricultural machinery operation performance prediction method and device based on kriging method and decision tree
CN116299704A (en) * 2023-03-20 2023-06-23 四川西南交大铁路发展股份有限公司 Earthquake single-unit positioning correction method, system, storage medium and terminal
CN116777116A (en) * 2023-06-29 2023-09-19 金乡县园林绿化服务中心 Cloud computing-based intelligent garden data management method and system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118261401A (en) * 2024-05-29 2024-06-28 福建拓海建设工程有限公司 Intelligent management system for landscaping engineering project
CN118261401B (en) * 2024-05-29 2024-08-09 福建拓海建设工程有限公司 Intelligent management system for landscaping engineering project

Also Published As

Publication number Publication date
CN118095789B (en) 2024-07-23

Similar Documents

Publication Publication Date Title
CN118095789B (en) Cloud computing-based intelligent garden data management method and system
Hickey Slope angle and slope length solutions for GIS
CN113902580B (en) Historical farmland distribution reconstruction method based on random forest model
KR102214360B1 (en) Land-cover Change prediction Method Using Regional Land-use Change Expected Values
CN115907436B (en) Quality coupling prediction-based water resource water environment regulation and control method and system
CN111967696A (en) Neural network-based electric vehicle charging demand prediction method, system and device
CN114970302B (en) Regional groundwater condition prediction method based on groundwater monitoring system
KR102211157B1 (en) optimaized water cooling apparatus and method for solar photovoltatic power generation module using rain water
CN109063388A (en) The micro climate architecture design addressing design method of wind environment simulation
CN115829812A (en) Carbon sequestration amount calculation method and system based on ecosystem simulation
CN117688844B (en) Urban waterlogging real-time simulation method and system based on deep neural network
Yue et al. Simulation and prediction of land use change in Dongguan of China based on ANN cellular automata-Markov chain model
Tehrani et al. Predicting urban Heat Island in European cities: A comparative study of GRU, DNN, and ANN models using urban morphological variables
CN111914465A (en) Data-free regional hydrological parameter calibration method based on clustering and particle swarm optimization
CN117150600A (en) Roof water storage and drainage control method for urban waterlogging prevention
Rajan et al. A GIS based integrated land use/cover change model to study human-land interactions
CN116307018A (en) Wind speed prediction method and system based on WRF mode sensitivity parameter adjustment
CN115496128A (en) Urban waterlogging risk forecasting method based on raininess-raininess characteristic parameter combined distribution
CN115860248A (en) Wind speed numerical assimilation short-term prediction method and system
CN115034159A (en) Power prediction method, device, storage medium and system for offshore wind farm
CN111914430A (en) Data-based regional hydrological parameter calibration method based on clustering and particle swarm optimization
CN117455041B (en) Multi-target management and control utilization method for future water resources obtained through prediction
Margini et al. Water allocation in multi-purpose and multi-year reservoir using ant colony optimization
CN115422848B (en) Urban water system simulation method and device, electronic equipment and storage medium
CN113592664B (en) Crop production space prediction simulation method, device, model and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant