Greening engineering construction monitoring system and method based on Internet of Things

Abstract

The invention discloses a greening engineering construction monitoring system and method based on the Internet of Things, and relates to the technical field of construction monitoring, and the monitoring method comprises the following steps: carrying out the category division of all greening plants of each greening engineering, collecting the construction data generated in the construction process, and generating a corresponding construction log; construction link division is carried out on any construction log, abnormal construction links are identified, and abnormal greening plants are identified; obtaining influence characteristics of each construction link, and generating a plurality of influence transmission chains; analyzing the influence degree between two influence characteristics in different construction links; the method comprises the following steps: collecting construction data of a greening project at the current moment, identifying potential influence characteristics of a construction link, and predicting a subsequent abnormal construction link; accurate prediction and early warning of potential anomalies in a construction link are realized, and the initiative and risk prevention and control capability of project management are improved.

Description

Technical Field

[0001] This invention relates to the field of construction monitoring technology, specifically to a construction monitoring system and method for greening projects based on the Internet of Things. Background Technology

[0002] Currently, the accelerated urbanization process has placed higher demands on the construction quality and management efficiency of landscaping projects. In the existing technology, the monitoring of landscaping project construction mainly relies on traditional manual inspection and paper record-keeping methods, which have significant defects in achieving traceability of the entire construction process. In this approach, data collection for each key process suffers from defects such as poor real-time performance and strong subjectivity. When problems such as low seedling survival rate or substandard construction quality occur, the lack of objective and continuous data support leads to untimely problem detection, difficulty in defining responsibilities, and disconnection in process control, which seriously affects project quality and dispute resolution efficiency. The invention patent with patent number CN110393176A discloses "Intelligent Control System and Method for Plant Maintenance Based on Internet of Things", which describes the use of Internet of Things sensors to collect plant data, thereby solving the problems of manual inspection and paper record-keeping in achieving traceability of the entire construction process. However, it does not involve the division of construction links and the identification of abnormal links, and cannot achieve full traceability of the construction process. Summary of the Invention

[0003] The purpose of this invention is to provide a construction monitoring system and method for greening projects based on the Internet of Things, so as to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring the construction of greening projects based on the Internet of Things, the monitoring method comprising the following steps: Step S100: Classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; classify any construction log into construction stages and identify abnormal construction stages. Step S200: Identify abnormal green plants in any abnormal construction phase in the construction log; compare the changes of the same green plants before the abnormal construction phase to obtain the impact characteristics of each construction phase. Step S300: Connect the influence characteristics of any type of greening plant at each construction stage to generate several influence transmission chains; Step S400: Analyze the impact of any influencing feature on subsequent influencing features in each construction log, and obtain the degree of influence between two influencing features in different construction stages; Step S500: Collect construction data of the greening project at the current moment, confirm the current construction stage, identify the potential impact characteristics of the current construction stage, extract the corresponding impact transmission chain, and predict subsequent construction stages with abnormalities.

[0005] Furthermore, step S100 includes the following steps: Step S101: Before carrying out the greening project, assign a unique identification code to each green plant and obtain the plant species of each green plant. Match the identification code of any green plant with the plant species, summarize the identification codes of all green plants of the same plant species, and divide all green plants into several plant sets according to plant species. Step S102: Deploy an IoT sensor network at the construction site in advance to collect plant data and construction data of the green plants at regular intervals during the construction process. Generate a data set for each data attribute according to the data attributes, and generate corresponding construction logs. Data attributes may include soil moisture, leaf temperature, leaf color, plant height, ambient temperature and humidity, light intensity, wind speed and direction, etc. Step S103: A construction phase database is pre-built, storing several construction phases. For each construction phase, several corresponding data attributes are matched. A construction log is arbitrarily selected, and construction data collected at a random unit time point is chosen to obtain several data attributes for that unit time point. A construction phase is arbitrarily selected; if the data attributes matched by the selected construction phase are identical to those of the selected time point, then the selected construction phase is set as the construction phase at that unit time point. Two adjacent unit time points are arbitrarily selected; if the two unit time points are in the same construction phase, then the two unit time points are divided into construction time intervals for that construction phase. The construction phases at each unit time point in the selected construction log are obtained, and several adjacent and consecutive unit time points of the same construction phase are summarized to obtain the construction time intervals for the same construction phase. The selected construction log is then divided into several construction phases. Step S104: Predefine an anomaly judgment rule for any data attribute to obtain the corresponding expected value range. Randomly select a construction stage from the selected construction log, and randomly select a data attribute from the selected construction stage to obtain the value range of the selected data attribute. If the value range of the selected data attribute is not within the expected value range, then set the selected data attribute as an anomaly attribute. Count the number of anomaly attributes in the selected construction stage as m, and set the total number of data attributes in the selected construction stage as M. total The percentage of abnormal attributes in the selected construction phase, η = m / M, is calculated. totalPreset an abnormality percentage threshold η th If η≥η th If so, the selected construction stage will be set as an abnormal construction stage, and several abnormal construction stages will be generated in the selected construction log; the abnormal judgment rule can be to preset an expected value range for each data attribute and judge whether the collected data value is within the expected value range.

[0006] Furthermore, step S200 includes the following steps: Step S201: Randomly select a construction log, extract several abnormal construction steps from the selected construction log, arbitrarily select one abnormal construction step, and obtain each abnormal attribute from the selected abnormal construction step; extract each data attribute corresponding to the green plant, compare the abnormal attributes with the data attributes of the green plant, and count the number of identical attributes as m1. s Let M1 be the number of data attributes for green plants, and calculate the proportion of plants with abnormal attributes η1=m1. s / M1, preset an abnormal correlation percentage threshold η1 th If η1≥η1 th If abnormal construction processes are detected, green plants in those processes will be marked as abnormal. The percentage of abnormal plants will be used to reflect the abnormality of the green plants, thus allowing for the elimination of occasional abnormalities in the green plants. Step S202: Obtain the plant species included in the selected construction log, and arbitrarily select one plant species, and arbitrarily select one green plant from the selected plant species; arbitrarily select one with the same attribute, obtain the expected value range of the same attribute, and extract the data value corresponding to the same attribute of the selected green plant at any unit time point under the selected abnormal construction stage. If there is a data value at a unit time point that is not within the expected value range, then the selected green plant is set as an abnormal plant with the same attribute. Step S203: Count the number of abnormal plants with the same attribute among the selected plant types as n1, set the total number of plants among the selected plant types as N, calculate the abnormal proportion of the selected plant types δ=n1 / N, and preset an abnormal proportion threshold δ. th If δ≥δ th Then the selected plant species will be set to select abnormal plant species in abnormal construction processes. Step S204: Set the construction log containing the selection of abnormal construction steps and the selection of plant types within those abnormal construction steps as the target log. If the selected plant type is an abnormal plant type, set the target log as an abnormal target log; otherwise, set it as a normal target log. Randomly select an abnormal target log and extract the previous construction step from it to obtain several abnormal attributes of the previous construction step. Set each abnormal attribute as an abnormal feature to obtain several abnormal features of the previous construction step. Summarize the several abnormal features of the same construction step in all abnormal target logs to obtain the set of abnormal features of the previous construction step. There will be multiple abnormal features between each step, but this does not mean that there is an influence relationship between any two abnormal features. Therefore, it is necessary to filter the influence between abnormal features through all historical logs to find the accurate influence relationship. Step S205: Extract any normal target log, extract several abnormal features from the previous construction stage of the selected abnormal construction stage in the extracted normal target log, if there is a certain abnormal feature with the same feature in the abnormal feature set, remove the certain abnormal feature from the abnormal feature set, and set each abnormal feature in the corrected abnormal feature set as the influence feature of the selected plant species.

[0007] Furthermore, step S300 includes the following steps: Step S301: Randomly select a construction log, acquire the plant species in the selected construction log, and arbitrarily select one plant species as the target species. Then, arbitrarily select a construction stage from the selected construction log, and extract the various impact features of the target species in the selected construction stage. At the same time, extract the various abnormal features in the previous construction stage of the selected construction stage, compare the impact features with the abnormal features, and set the same features as the actual impact features of the selected construction stage. Step S302: Extract the actual impact features of each construction stage in the selected construction log, connect the actual impact features between any two adjacent construction stages, and generate several impact transmission chains for the target type.

[0008] Furthermore, step S400 includes the following steps: Step S401: Randomly select a construction log, randomly select an influence transmission chain from the selected construction log and set it as the target transmission chain, and randomly extract two adjacent influence features from the target transmission chain and set them as the target influence feature group; extract any two adjacent influence features from the remaining influence transmission chains, and if two influence features in a certain influence transmission chain are the same as the target influence feature group, then set that influence transmission chain as the target transmission chain as well; Step S402: Obtain the plant species corresponding to each target transmission chain, count the number of plant species included in the target transmission chain as A1, set the total number of plant species as A, and calculate the plant correlation of the target transmission chain G1=A1 / A; obtain the construction logs related to each target transmission chain and set them as feature logs, arbitrarily select the i-th target transmission chain, and count the number of feature logs corresponding to the i-th target transmission chain as P. i The occurrence frequency of the i-th target transmission chain is calculated to be G2. i =P i / w, where w is the number of construction logs; Plant association G1 reflects the prevalence of the influence transmission chain. The higher the G1 value, the wider the scope of the influence chain, the more plant species involved, and the higher its importance; Occurrence frequency G2 reflects the frequency of the influence transmission chain in historical data. The higher the G2 value, the more common the transmission chain was in past construction, and the higher its predictive value and reliability. Step S403: Randomly select two adjacent construction stages from the selected construction log, and extract the target transfer chain containing the two construction stages according to the formula: ; Where u is the number of target transmission chains containing two construction stages; the degree of influence R between the two construction stages is calculated.

[0009] Furthermore, step S500 includes the following steps: Step S501: Obtain the construction data collected at the current moment, divide it into several data attribute sets, retrieve the construction stage database for comparison, and obtain the real-time construction stage at the current moment; Step S502: Extract the data values ​​of each data attribute in the real-time construction process and compare them with the predefined anomaly judgment rules of each data attribute to obtain the expected value range of each data attribute. If the data value of a certain data attribute is not in the corresponding expected data range, then the certain data attribute is set as a potential impact feature. Step S503: Extract several potential impact features in the real-time construction process, arbitrarily select one potential impact feature, and generate a real-time impact group with the selected potential impact feature and the real-time construction process; obtain each plant species in the real-time construction process, and extract the impact transmission chain of each plant species. If a certain impact transmission chain contains the real-time impact group, then set the certain impact transmission chain as the desired transmission chain. Step S504: Randomly select an expected transmission chain and acquire the next construction stage in the selected expected transmission chain to obtain the influence degree between the real-time construction stage and the next construction stage as R; set the occurrence frequency of the expected transmission chain as G2, and calculate the expected anomaly value Y = R × G2 for the next construction stage; extract the expected anomaly values ​​of the next construction stage in each expected transmission chain, and accumulate them to obtain the comprehensive anomaly value Y of the next construction stage. total A preset abnormal threshold Y th If Y total ≥Y th If an anomaly occurs, an alert will be sent to the real-time construction process.

[0010] To better implement the above methods, a greening project construction monitoring system is also proposed. The monitoring system includes a historical greening analysis module, a reverse tracing analysis module, an impact transmission analysis module, a characteristic impact analysis module, and an abnormal construction prediction module. The historical greening analysis module is used to classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; it can also classify any construction log into construction stages and identify abnormal construction stages. The reverse tracing analysis module is used to identify abnormal green plants in any abnormal construction stage in the construction log; and to compare the differences in the changes of the same green plants before the abnormal construction stage to obtain the impact characteristics of each construction stage. The impact transmission analysis module is used to connect the impact characteristics of any type of greening plant at various construction stages and generate several impact transmission chains. The feature impact analysis module is used to analyze the impact of any impact feature on subsequent impact features in various construction logs, and to obtain the degree of influence between two impact features in different construction stages. The abnormal construction prediction module is used to collect construction data of the greening project at the current moment, identify the current construction stage, identify the potential impact characteristics of the current construction stage, extract the corresponding impact transmission chain, and predict the subsequent abnormal construction stages.

[0011] Furthermore, the historical greening analysis module includes a construction log generation unit and an anomaly identification unit; The construction log generation unit is used to classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; the abnormal link identification unit is used to classify any construction log into construction links and identify abnormal construction links.

[0012] Furthermore, the reverse tracing analysis module includes an abnormal plant identification unit and an impact feature extraction unit; The abnormal plant identification unit is used to identify abnormal green plants in any abnormal construction stage in the construction log; the influence feature extraction unit is used to compare the differences in the changes of the same green plants before the abnormal construction stage to obtain the influence features of each construction stage.

[0013] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention constructs an impact transmission chain based on historical data, enabling accurate prediction and early warning of potential anomalies in the construction process. This helps to take intervention measures in advance, effectively avoid the occurrence of construction quality problems, and improve the initiative and risk control capabilities of project management. 2. This invention provides a scientific basis for construction decisions through multi-dimensional data analysis and impact degree calculation, and supports the quantitative evaluation of the response characteristics of various green plants in different construction stages, thereby optimizing construction technology and improving the survival rate and overall quality of greening projects. 3. This invention enables traceability and refined management of the entire construction process. By identifying abnormal links and tracing back, it clarifies the root cause of the problem and the attribution of responsibility, enhances the transparency and accountability mechanism of project management, and provides technical support for the standardization and normalization of greening project construction. Attached Figure Description

[0014] Figure 1 A schematic diagram illustrating the steps of a construction monitoring method for greening projects based on the Internet of Things; Figure 2 This is a schematic diagram of a construction monitoring system for greening projects based on the Internet of Things. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Example: Figures 1 to 2 As shown, this invention provides a method for monitoring the construction of greening projects based on the Internet of Things. The monitoring method includes the following steps: Step S100: Classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; classify any construction log into construction stages and identify abnormal construction stages. Step S100 includes the following steps: Step S101: Before carrying out the greening project, assign a unique identification code to each green plant and obtain the plant species of each green plant. Match the identification code of any green plant with the plant species, summarize the identification codes of all green plants of the same plant species, and divide all green plants into several plant sets according to plant species. Example 1: Before construction, each plant is assigned a unique identification code; for example, ginkgo trees are assigned identification codes G001 and G002, rose shrubs are assigned identification codes R001 and R002, and lawns are assigned identification codes such as C001 and C002; then the identification codes of the same type are summarized to form a ginkgo tree set, a rose shrub set, and a lawn set. Step S102: Deploy an IoT sensor network in advance at the construction site to collect plant data and construction data of green plants at each unit time point during the construction process of the greening project, generate a data set for each data attribute according to the data attributes, and generate corresponding construction logs. Step S103: A construction phase database is pre-built, storing several construction phases. For each construction phase, several corresponding data attributes are matched. A construction log is arbitrarily selected, and construction data collected at a random unit time point is chosen to obtain several data attributes for that unit time point. A construction phase is arbitrarily selected; if the data attributes matched by the selected construction phase are identical to those of the selected time point, then the selected construction phase is set as the construction phase at that unit time point. Two adjacent unit time points are arbitrarily selected; if the two unit time points are in the same construction phase, then the two unit time points are divided into construction time intervals for that construction phase. The construction phases at each unit time point in the selected construction log are obtained, and several adjacent and consecutive unit time points of the same construction phase are summarized to obtain the construction time intervals for the same construction phase. The selected construction log is then divided into several construction phases. Step S104: Predefine an anomaly judgment rule for any data attribute to obtain the corresponding expected value range. Randomly select a construction stage from the selected construction log, and randomly select a data attribute from the selected construction stage to obtain the value range of the selected data attribute. If the value range of the selected data attribute is not within the expected value range, then set the selected data attribute as an anomaly attribute. Count the number of anomaly attributes in the selected construction stage as m, and set the total number of data attributes in the selected construction stage as M. total The percentage of abnormal attributes in the selected construction phase, η = m / M, is calculated. total Preset an abnormality percentage threshold η th If η≥η th If so, the selected construction stage will be set as an abnormal construction stage, and several abnormal construction stages will be generated in the selected construction log. Example 2: Anomaly detection rules were defined for the data attribute "soil moisture," with an expected value range of 20%-40%. The value range of "soil moisture" was extracted from the "planting" stage and found to be 10%-15%, which is outside the expected range; therefore, "soil moisture" was marked as an abnormal attribute. The number of abnormal attributes in this stage was counted as 1, and the total number of data attributes was 3, resulting in an abnormal attribute percentage η = 33.3%. A preset anomaly percentage threshold η was set. th =50%, since η < η th This step is not marked as abnormal.

[0017] Step S200: Identify abnormal green plants in any abnormal construction phase in the construction log; compare the changes of the same green plants before the abnormal construction phase to obtain the impact characteristics of each construction phase. Step S200 includes the following steps: Step S201: Randomly select a construction log, extract several abnormal construction steps from the selected construction log, arbitrarily select one abnormal construction step, and obtain each abnormal attribute from the selected abnormal construction step; extract each data attribute corresponding to the green plant, compare the abnormal attributes with the data attributes of the green plant, and count the number of identical attributes as m1. s Let M1 be the number of data attributes for green plants, and calculate the proportion of plants with abnormal attributes η1=m1. s / M1, preset an abnormal correlation percentage threshold η1 th If η1≥η1 th If so, green plants from abnormal construction phases will be selected and marked as abnormal. Step S202: Obtain the plant species included in the selected construction log, and arbitrarily select one plant species, and arbitrarily select one green plant from the selected plant species; arbitrarily select one with the same attribute, obtain the expected value range of the same attribute, and extract the data value corresponding to the same attribute of the selected green plant at any unit time point under the selected abnormal construction stage. If there is a data value at a unit time point that is not within the expected value range, then the selected green plant is set as an abnormal plant with the same attribute. Step S203: Count the number of abnormal plants with the same attribute among the selected plant types as n1, set the total number of plants among the selected plant types as N, calculate the abnormal proportion of the selected plant types δ=n1 / N, and preset an abnormal proportion threshold δ. th If δ≥δ th Then the selected plant species will be set to select abnormal plant species in abnormal construction processes. Step S204: Set the construction log containing the selection of abnormal construction steps and the selection of plant types within the selected abnormal construction steps as the target log. If the selected plant type is an abnormal plant type, set the target log as an abnormal target log; otherwise, set it as a normal target log. Randomly select an abnormal target log, extract the previous construction step in the selected abnormal construction step in the selected abnormal target log, obtain several abnormal attributes of the previous construction step, and set each abnormal attribute as an abnormal feature, thus obtaining several abnormal features of the selected previous construction step. Summarize the several abnormal features of the same construction step in all abnormal target logs to obtain the abnormal feature set of the previous construction step. Step S205: Extract any normal target log, extract several abnormal features from the previous construction stage of the selected abnormal construction stage in the extracted normal target log, if there is a certain abnormal feature with the same feature in the abnormal feature set, remove the certain abnormal feature from the abnormal feature set, and set each abnormal feature in the corrected abnormal feature set as the influence feature of the selected plant species.

[0018] Step S300: Connect the influence characteristics of any type of greening plant at each construction stage to generate several influence transmission chains; Step S300 includes the following steps: Step S301: Randomly select a construction log, acquire the plant species in the selected construction log, and arbitrarily select one plant species as the target species. Then, arbitrarily select a construction stage from the selected construction log, and extract the various impact features of the target species in the selected construction stage. At the same time, extract the various abnormal features in the previous construction stage of the selected construction stage, compare the impact features with the abnormal features, and set the same features as the actual impact features of the selected construction stage. Step S302: Extract the actual impact features of each construction stage in the selected construction log, and connect the actual impact features between any two adjacent construction stages to generate several impact transmission chains for the target type. Example 3: Select a construction log, with the target species being ginkgo trees, and select the construction stage "watering". Extract the influencing feature "soil moisture too low" for ginkgo trees in the "watering" stage, and simultaneously extract the abnormal feature "soil hardness too high" from the previous stage "planting". After comparison, the same feature "soil hardness too high" is set as the actual influencing feature of the "watering" stage. Connect the actual influencing features of all construction stages in the construction log to obtain the actual influencing feature of "land leveling" stage as "soil hardness too high", which is transmitted to the "planting" stage as "insufficient plant depth", and then to the "watering" stage as "soil moisture too low", generating an influence transmission chain.

[0019] Step S400: Analyze the impact of any influencing feature on subsequent influencing features in each construction log, and obtain the degree of influence between two influencing features in different construction stages; Step S400 includes the following steps: Step S401: Randomly select a construction log, randomly select an influence transmission chain from the selected construction log and set it as the target transmission chain, and randomly extract two adjacent influence features from the target transmission chain and set them as the target influence feature group; extract any two adjacent influence features from the remaining influence transmission chains, and if two influence features in a certain influence transmission chain are the same as the target influence feature group, then set that influence transmission chain as the target transmission chain as well; Step S402: Obtain the plant species corresponding to each target transmission chain, count the number of plant species included in the target transmission chain as A1, set the total number of plant species as A, and calculate the plant correlation of the target transmission chain G1=A1 / A; obtain the construction logs related to each target transmission chain and set them as feature logs, arbitrarily select the i-th target transmission chain, and count the number of feature logs corresponding to the i-th target transmission chain as P. i The occurrence frequency of the i-th target transmission chain is calculated to be G2. i =P i / w, where w is the number of construction logs; Step S403: Randomly select two adjacent construction stages from the selected construction log, and extract the target transfer chain containing the two construction stages according to the formula: ; Where u is the number of target transmission chains containing two construction stages; the degree of influence R between the two construction stages is calculated.

[0020] Step S500: Collect construction data of the greening project at the current moment, confirm the current construction stage, identify the potential impact characteristics of the current construction stage, extract the corresponding impact transmission chain, and predict the subsequent construction stages with abnormalities. Step S500 includes the following steps: Step S501: Obtain the construction data collected at the current moment, divide it into several data attribute sets, retrieve the construction stage database for comparison, and obtain the real-time construction stage at the current moment; Step S502: Extract the data values ​​of each data attribute in the real-time construction process and compare them with the predefined anomaly judgment rules of each data attribute to obtain the expected value range of each data attribute. If the data value of a certain data attribute is not in the corresponding expected data range, then the certain data attribute is set as a potential impact feature. Step S503: Extract several potential impact features in the real-time construction process, arbitrarily select one potential impact feature, and generate a real-time impact group with the selected potential impact feature and the real-time construction process; obtain each plant species in the real-time construction process, and extract the impact transmission chain of each plant species. If a certain impact transmission chain contains the real-time impact group, then set the certain impact transmission chain as the desired transmission chain. Step S504: Randomly select an expected transmission chain and acquire the next construction stage in the selected expected transmission chain to obtain the influence degree between the real-time construction stage and the next construction stage as R; set the occurrence frequency of the expected transmission chain as G2, and calculate the expected anomaly value Y = R × G2 for the next construction stage; extract the expected anomaly values ​​of the next construction stage in each expected transmission chain, and accumulate them to obtain the comprehensive anomaly value Y of the next construction stage. total A preset abnormal threshold Y th If Y total ≥Y th If an anomaly occurs, an alert will be sent to the real-time construction process.

[0021] A greening project construction monitoring system includes a historical greening analysis module, a reverse tracing analysis module, an impact transmission analysis module, a characteristic impact analysis module, and an abnormal construction prediction module. The historical greening analysis module is used to classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; it can also classify any construction log into construction stages and identify abnormal construction stages. The reverse tracing analysis module is used to identify abnormal green plants in any abnormal construction stage in the construction log; and to compare the differences in the changes of the same green plants before the abnormal construction stage to obtain the impact characteristics of each construction stage. The impact transmission analysis module is used to connect the impact characteristics of any type of greening plant at various construction stages and generate several impact transmission chains. The feature impact analysis module is used to analyze the impact of any impact feature on subsequent impact features in various construction logs, and to obtain the degree of influence between two impact features in different construction stages. The abnormal construction prediction module is used to collect construction data of the greening project at the current moment, identify the current construction stage, identify the potential impact characteristics of the current construction stage, extract the corresponding impact transmission chain, and predict the subsequent abnormal construction stages.

[0022] The historical greening analysis module includes a construction log generation unit and an anomaly identification unit. The construction log generation unit is used to classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; the abnormal link identification unit is used to classify any construction log into construction links and identify abnormal construction links.

[0023] The reverse tracing analysis module includes an abnormal plant identification unit and an impact feature extraction unit. The abnormal plant identification unit is used to identify abnormal green plants in any abnormal construction stage in the construction log; the influence feature extraction unit is used to compare the differences in the changes of the same green plants before the abnormal construction stage to obtain the influence features of each construction stage.

[0024] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for monitoring the construction of a green project based on the Internet of Things, characterized in that: The monitoring method includes the following steps: Step S100: Classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; classify any construction log into construction stages and identify abnormal construction stages. Step S200: Identify abnormal green plants in any abnormal construction phase in the construction log; compare the changes of the same green plants before the abnormal construction phase to obtain the impact characteristics of each construction phase. Step S300: Connect the influence characteristics of any type of greening plant at each construction stage to generate several influence transmission chains; Step S400: Analyze the impact of any influencing feature on subsequent influencing features in each construction log, and obtain the degree of influence between two influencing features in different construction stages; Step S500: Collect construction data of the greening project at the current moment, confirm the current construction stage, identify the potential impact characteristics of the current construction stage, extract the corresponding impact transmission chain, and predict subsequent construction stages with abnormalities. 2.The method of claim 1, wherein the method further comprises: Step S100 includes the following steps: Step S101: Before carrying out the greening project, assign a unique identification code to each green plant and obtain the plant species of each green plant. Match the identification code of any green plant with the plant species, summarize the identification codes of all green plants of the same plant species, and divide all green plants into several plant sets according to plant species. Step S102: Deploy an IoT sensor network in advance at the construction site to collect plant data and construction data of green plants at each unit time point during the construction process of the greening project, generate a data set for each data attribute according to the data attributes, and generate corresponding construction logs. Step S103: A construction phase database is pre-built, storing several construction phases. For each construction phase, several corresponding data attributes are matched. A construction log is arbitrarily selected, and construction data collected at a random unit time point is chosen to obtain several data attributes for that unit time point. A construction phase is arbitrarily selected; if the data attributes matched by the selected construction phase are identical to those of the selected time point, then the selected construction phase is set as the construction phase at that unit time point. Two adjacent unit time points are arbitrarily selected; if the two unit time points are in the same construction phase, then the two unit time points are divided into construction time intervals for that construction phase. The construction phases at each unit time point in the selected construction log are obtained, and several adjacent and consecutive unit time points of the same construction phase are summarized to obtain the construction time intervals for the same construction phase. The selected construction log is then divided into several construction phases. Step S104: defining an abnormality judgment rule for any data attribute in advance to obtain a corresponding expected value range, selecting any construction link from the selected construction log, and selecting any data attribute from the selected construction link to obtain the value range of the selected data attribute; if the value range of the selected data attribute is not in the expected value range, the selected data attribute is set as an abnormal attribute, the number of abnormal attributes in the selected construction link is counted as m, and the total number of data attributes in the selected construction link is set as M total , the abnormal attribute proportion η of the selected construction link is calculated as m / M total , an abnormal proportion threshold η th is preset, and if η ≥ η th , the selected construction link is set as an abnormal construction link, and a plurality of abnormal construction links of the selected construction log are generated. 3.The method of claim 2, wherein the method further comprises: Step S200 includes the following steps: Step S201: randomly select a construction log, extract a plurality of abnormal construction links in the selected construction log, randomly select an abnormal construction link, and obtain each abnormal attribute from the selected abnormal construction link; extract each data attribute corresponding to the green plant, compare the abnormal attribute with the data attribute of the green plant, and count the number of the same attributes as m1 s , set the number of data attributes of the green plant as M1, calculate the plant-related proportion of the abnormal attribute as η1=m1 s / M1, preset an abnormal-related proportion threshold η1 th , if η1≥η1 th , the green plant of the selected abnormal construction link is marked as abnormal. Step S202: Obtain the plant species included in the selected construction log, and arbitrarily select one plant species, and arbitrarily select one green plant from the selected plant species; arbitrarily select one with the same attribute, obtain the expected value range of the same attribute, and extract the data value corresponding to the same attribute of the selected green plant at any unit time point under the selected abnormal construction stage. If there is a data value at a unit time point that is not within the expected value range, then the selected green plant is set as an abnormal plant with the same attribute. Step S203: count the number of abnormal plants with the same attribute in the selected plant species as n1, set the total number of plants in the selected plant species as N, calculate the abnormal proportion δ = n1 / N of the selected plant species, and preset an abnormal proportion threshold δ th If δ ≥ δ th , the selected plant species is set as the abnormal plant species of the selected abnormal construction link. Step S204: Set the construction log containing the selection of abnormal construction steps and the selection of plant types within the selected abnormal construction steps as the target log. If the selected plant type is an abnormal plant type, set the target log as an abnormal target log; otherwise, set it as a normal target log. Randomly select an abnormal target log, extract the previous construction step in the selected abnormal construction step in the selected abnormal target log, obtain several abnormal attributes of the previous construction step, and set each abnormal attribute as an abnormal feature, thus obtaining several abnormal features of the selected previous construction step. Summarize the several abnormal features of the same construction step in all abnormal target logs to obtain the abnormal feature set of the previous construction step. Step S205: Extract any normal target log, extract several abnormal features from the previous construction stage of the selected abnormal construction stage in the extracted normal target log, if there is a certain abnormal feature with the same feature in the abnormal feature set, remove the certain abnormal feature from the abnormal feature set, and set each abnormal feature in the corrected abnormal feature set as the influence feature of the selected plant species.

4. The method according to claim 3, characterized in that: Step S300 includes the following steps: Step S301: Randomly select a construction log, acquire the plant species in the selected construction log, and arbitrarily select one plant species as the target species. Then, arbitrarily select a construction stage from the selected construction log, and extract the various impact features of the target species in the selected construction stage. At the same time, extract the various abnormal features in the previous construction stage of the selected construction stage, compare the impact features with the abnormal features, and set the same features as the actual impact features of the selected construction stage. Step S302: Extract the actual impact features of each construction stage in the selected construction log, connect the actual impact features between any two adjacent construction stages, and generate several impact transmission chains for the target type.

5. The method according to claim 4, characterized in that: Step S400 includes the following steps: Step S401: Randomly select a construction log, randomly select an influence transmission chain from the selected construction log and set it as the target transmission chain, and randomly extract two adjacent influence features from the target transmission chain and set them as the target influence feature group; extract any two adjacent influence features from the remaining influence transmission chains, and if two influence features in a certain influence transmission chain are the same as the target influence feature group, then set that influence transmission chain as the target transmission chain as well; Step S402: Obtain the plant species corresponding to each target transmission chain, count the number of plant species included in the target transmission chain as A1, set the total number of plant species as A, and calculate the plant correlation of the target transmission chain G1=A1 / A; obtain the construction logs related to each target transmission chain and set them as feature logs, arbitrarily select the i-th target transmission chain, and count the number of feature logs corresponding to the i-th target transmission chain as P. i The occurrence frequency of the i-th target transmission chain is calculated to be G2. i =P i / w, where w is the number of construction logs; Step S403: Randomly select two adjacent construction stages from the selected construction log, and extract the target transfer chain containing the two construction stages according to the formula: ； Where u is the number of target transmission chains containing two construction stages; the degree of influence R between the two construction stages is calculated.

6. The method for monitoring the construction of greening projects based on the Internet of Things according to claim 5, characterized in that: Step S500 includes the following steps: Step S501: Obtain the construction data collected at the current moment, divide it into several data attribute sets, retrieve the construction stage database for comparison, and obtain the real-time construction stage at the current moment; Step S502: Extract the data values ​​of each data attribute in the real-time construction process and compare them with the predefined anomaly judgment rules of each data attribute to obtain the expected value range of each data attribute. If the data value of a certain data attribute is not in the corresponding expected data range, then the certain data attribute is set as a potential impact feature. Step S503: Extract several potential impact features in the real-time construction process, arbitrarily select one potential impact feature, and generate a real-time impact group with the selected potential impact feature and the real-time construction process; obtain each plant species in the real-time construction process, and extract the impact transmission chain of each plant species. If a certain impact transmission chain contains the real-time impact group, then set the certain impact transmission chain as the desired transmission chain. Step S504: any selected a desired delivery chain, and to select the next construction link in the desired delivery chain to obtain, the real-time construction link between the influence degree of the next construction link is R; set the occurrence frequency of the desired delivery chain is G2, calculated to get the next construction link of the expected abnormal value Y=R×G2; to each next construction link in the desired delivery chain of the expected abnormal value is extracted, is accumulated to get the next construction link of the comprehensive abnormal value Y total , pre-set an abnormal threshold Y th , if Y total ≥Y th , the real-time construction link sends abnormal prompt.

7. A greening project construction monitoring system, used to execute the greening project construction monitoring method based on the Internet of Things as described in any one of claims 1-6, characterized in that: The monitoring system includes a historical greening analysis module, a reverse tracing analysis module, an impact transmission analysis module, a characteristic impact analysis module, and an abnormal construction prediction module; The historical greening analysis module is used to classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; it can also classify any construction log into construction stages and identify abnormal construction stages. The reverse tracing analysis module is used to identify abnormal green plants in any abnormal construction stage in the construction log; and to compare the differences in the changes of the same green plants before the abnormal construction stage to obtain the impact characteristics of each construction stage. The influence transmission analysis module is used to connect the influence characteristics of any type of greening plant in each construction stage and generate several influence transmission chains. The feature impact analysis module is used to analyze the impact of any impact feature on subsequent impact features in various construction logs, and to obtain the degree of impact between two impact features in different construction stages. The abnormal construction prediction module is used to collect construction data of the greening project at the current moment, confirm the current construction stage, identify the potential impact characteristics of the current construction stage, extract the corresponding impact transmission chain, and predict the subsequent abnormal construction stages.

8. A greening project construction monitoring system according to claim 7, characterized in that: The historical greening analysis module includes a construction log generation unit and an anomaly identification unit. The construction log generation unit is used to classify all green plants in each greening project by species, collect construction data generated during the construction process, and generate corresponding construction logs; the abnormal link identification unit is used to classify any construction log into construction links and identify abnormal construction links.

9. A greening project construction monitoring system according to claim 7, characterized in that: The reverse tracing analysis module includes an abnormal plant identification unit and an impact feature extraction unit; The abnormal plant identification unit is used to identify abnormal green plants in any abnormal construction stage in the construction log; the influence feature extraction unit is used to compare the differences in the changes of the same green plants before the abnormal construction stage to obtain the influence features of each construction stage.

Images