A traditional dwelling environment quality evaluation method based on semantic information

Abstract

The application discloses a traditional residence environment quality evaluation method based on semantic information, first, through a mobile terminal carrying a collector, environment data of traditional residence buildings are collected, and the data are transmitted to a cloud server to pre-process original data; then, semantic information of each sampling point is calculated in combination with multi-source information, then, based on the semantic information, K-medoids clustering is carried out to obtain pollution source distribution and pollution propagation trend conditions in the whole traditional residence environment, regional division is carried out, and visual output is carried out; finally, according to the self-adaptive adjustment of environment parameter weights of each divided region, a fuzzy comprehensive evaluation method is used to evaluate the collected traditional residence environment. In the traditional residence environment quality evaluation process, the application can collect environment data with high quality, low cost and high accuracy, and can reasonably evaluate the traditional residence environment quality, so as to provide decision makers with optimization and processing.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of environmental quality monitoring, and particularly relates to a traditional dwelling environmental quality evaluation method based on semantic information. BACKGROUND

[0002] At present, in the environmental monitoring of traditional dwellings, a fixed sensor-based method is usually used to collect environmental data of traditional dwellings. However, the indoor environment is complex and variable, the environmental parameters of traditional dwellings are unevenly distributed, and the collection range of fixed sensors is limited, so it is difficult to comprehensively monitor the entire building. In addition, the evaluation of the advantages and disadvantages of the environmental quality of traditional dwellings is mostly limited to a single environmental factor, and lacks a traditional dwelling environmental evaluation system based on multiple environmental factors. SUMMARY

[0003] To solve the problems of high cost of building an environmental monitoring system, low monitoring accuracy and inaccurate evaluation in the process of monitoring the environment of traditional dwellings, the present application provides a traditional dwelling environmental quality evaluation method based on semantic information.

[0004] The present application adopts the following technical solution to solve the above technical problems: a traditional dwelling environmental quality evaluation method based on semantic information, characterized in that it comprises the following steps:

[0005] (1) Multi-source environmental information acquisition and processing;

[0006] The multi-source environmental information acquisition and processing stage comprises the following steps:

[0007] (1.1) Use a mobile carrier to carry a collection end to collect environmental data of traditional dwelling buildings, and obtain an environmental data set ; transmit the collected multi-source data to a cloud server to perform Gaussian smoothing filter processing on the original data to eliminate accidental factor interference, and then perform Kalman filter processing on the data to obtain reliable environmental parameter data;

[0008] (2) Fine-grained region division based on semantic information;

[0009] The region division stage comprises the following steps:

[0010] (2.1) According to the environmental prior physical structure information and the mobile carrier positioning information , solve the semantic label , and give each sampling data a semantic label ;

[0011] (2.2) K-medoids clustering of the processed data, specifically: conditional sorting of the sampling point information under each semantic information, obtaining the sampling points where the maximum and minimum values of each pollutant are located , clustering with these sampling points as initial clustering centers; calculating the distance between each sample and the clustering center under different semantic environments, and classifying; taking the center point as the new clustering center and repeatedly iterating

[0012] (2.3) Regional fine-grained division of traditional residential environment according to the clustering results of environmental pollutant information combined with the physical structure of the house ;

[0013] (2.4) Spatial heat map visualization output of the processed multi-source data to obtain the distribution of pollution sources and pollution transmission trend in the entire traditional residential environment

[0014] (3) Environmental assessment method based on fuzzy comprehensive evaluation

[0015] The specific steps of the environmental assessment stage are as follows:

[0016] (3.1) Fuzzy processing of the collected traditional residential environment parameters, establishing the membership function of each environmental parameter, obtaining the membership of each environmental parameter x in the fuzzy subset X, specifically: very low, low, normal, high and very high; similarly, establish fuzzy subset Y and membership function for the environment evaluation set y, and set the environment evaluation standard as excellent, good, qualified, poor and very poor

[0017] (3.2) Comprehensive membership of each environmental parameter, establish fuzzy control rules of environmental parameters and evaluation system ;

[0018] (3.3) Weight assignment using improved analytic hierarchy process, the improved analytic hierarchy process is as follows: first, establish a hierarchical structure model; according to the fine-grained regional division results and the semantic information of the environment, compare each environmental factor with each other to obtain the judgment matrix , wherein , the judgment matrix is adjusted adaptively according to the different environment semantics; calculate the maximum eigenvalue and the corresponding characteristic weight vector based on the judgment matrix

[0019] (3.4) Weighted according to the weight vector obtained and the established fuzzy control rules to obtain the fuzzy comprehensive evaluation result Finally, the maximum membership principle is used for environmental judgment to obtain the final environmental evaluation: .

[0020] As a further improvement of the present application: the original multi-source data collected in step (1.1) is specifically in the form of temperature, humidity, carbon dioxide concentration, TVOC concentration, formaldehyde concentration, time and location.

[0021] As a further improvement of the present application: the trapezoidal function is selected as the membership function model in step (3.1) according to the traditional residential environment evaluation system .

[0022] Compared with the prior art, the present application has the following technical effects:

[0023] After adopting the above technical means, the present application can obtain the following advantages: due to the adoption of the present technical solution, the present application uses a mobile carrier to carry a collector to traverse the environment and simultaneously collect multi-source environmental information under the target environment, so that non-fixed-point environmental monitoring can be realized; the data processing module is used to remove gross errors and smooth filtering of the multi-source information, and a multi-feature self-adaptive clustering method is used to cluster and divide the original data according to each multi-source information, and the original data is outputted and displayed visually; then the whole environment is divided into zones, the weights of each environmental parameter are determined by combining the clustering results and environmental semantic information and using hierarchical analysis, and the whole traditional residential environment is comprehensively evaluated by using fuzzy control, so that the target environment can be evaluated to the greatest extent, the environmental quality of the traditional residence can be scientifically and reasonably reflected, data reference can be provided for subsequent environmental reconstruction and regulation, and compared with the background technical method, the present technical solution has high efficiency, low cost and high precision. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 It is a monitoring system architecture diagram;

[0025] Figure 2 A traditional residential environment quality evaluation method flow chart based on semantic information

[0026] Figure 3 It is a fuzzy comprehensive judgment algorithm flow chart;

[0027] Figure 4 It is a hierarchical analysis model of the environmental evaluation problem. DETAILED DESCRIPTION

[0028] The present application will be further described in detail below in combination with the drawings and specific embodiments:

[0029] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0030] The present application adopts the mode of mobile carriers carrying collectors to replace the mode of fixed environmental monitoring base stations to improve the monitoring efficiency and accuracy of environmental monitoring. Meanwhile, the environmental quality evaluation problem of complex indoor environment is solved by combining environmental semantic information, fuzzy comprehensive evaluation method and improved analytic hierarchy process, so that the environmental quality status of traditional dwellings can be scientifically and reasonably reflected, thereby providing a reference for environmental regulation and reconstruction.

[0031] The specific technical solutions are as follows: a traditional dwelling environmental quality evaluation method based on semantic information, comprising the following steps:

[0032] Step 1: multi-source environmental information acquisition and processing, specifically comprising:

[0033] 1.1 The present example uses a mobile carrier carrying a collection end to collect environmental data of traditional dwelling building environment, and obtains an environmental data set ; wherein the position information of the mobile carrier is solved in real time by a combined positioning system. The sensor modules carried are mainly selected according to the environmental characteristics, including temperature and humidity sensors, carbon dioxide sensors, TVOC sensors and formaldehyde sensors;

[0034] 1.2 The data collected by the environmental sensors and the time and position data are transmitted to a cloud server, and the multi-source data are integrated in the cloud server, specifically in the form of temperature, humidity, carbon dioxide concentration, TVOC concentration, formaldehyde concentration, time and position. The original data are subjected to Gaussian smoothing filter processing to eliminate accidental factor interference, and then Kalman filter processing is performed on the data to obtain reliable environmental parameter data.

[0035] Step 2: fine-grained region division based on semantic information,

[0036] specifically comprising

[0037] 2.1 multi-dimensional K-medoids clustering of the processed data, specifically: first, according to the environmental prior physical structure information and the mobile carrier positioning information , the semantic label is solved , and each sampling data is given a semantic label ;

[0038] 2.2 The sampling point information under each semantic information is conditionally sorted to obtain the sampling points where the maximum and minimum values of each pollutant are located . These sampling points are used as initial clustering centers for clustering; the distances between each sample and the clustering center under different semantic environments are calculated for class division; and the center point is used as a new clustering center for repeated iteration

[0039] 2.3 The traditional dwelling environment is regionally and finely divided according to the environmental pollutant information clustering results and the physical structure of the house

[0040] 2.4 The processed multi-source data is spatially heat-mapped and visualized to obtain the pollution source distribution and pollution transmission trend in the entire traditional dwelling environment

[0041] Step 3: Environmental evaluation method based on fuzzy comprehensive evaluation

[0042] The specific steps are as follows

[0043] 3.1 The collected traditional dwelling environment parameters are fuzzified to establish the membership functions of each environmental parameter, obtaining the membership of each environmental parameter x in the fuzzy subset X. Specifically: very low (NB), low (NS), normal (ZO), high (PS), and very high (PB). Similarly, the fuzzy subset Y and membership function are established for the environmental evaluation set y, and the environmental evaluation standard is set as excellent (VS), good (S), qualified (M), poor (L), and very poor (VL). The membership function model is selected according to the traditional dwelling environment evaluation system

[0044] 3.2 The membership degrees of each environmental parameter are integrated to establish the fuzzy control rules of environmental parameters and evaluation system

[0045] 3.3 Improved analytic hierarchy process is used to assign weights. The improved analytic hierarchy process is as follows: first, a hierarchical structure model is established, as shown in Figure 4 ; each environmental factor is compared with each other according to the preliminary regional division results and environmental semantic information to obtain a judgment matrix , wherein , the judgment matrix is adjusted according to the different environmental semantics; the maximum eigenvalue and the corresponding characteristic weight vector are calculated based on the judgment matrix

[0046] 3.4 The fuzzy comprehensive evaluation result is obtained by weighting the obtained weight vector and the established fuzzy control rules ​​​​ Finally, the environment is judged by using the maximum membership principle to obtain the final environment evaluation: .

[0047] The application traverses the environment by using the mobile carrier to carry the collector, and collects multi-source environment information, so that non-fixed-point environment monitoring can be realized, and the efficiency is high and the effect is good; the data processing module is used to remove gross errors and smooth filtering of multi-source information, and a multi-feature self-adaptive clustering method is used for clustering and dividing the original data according to each multi-source information, and visual output is performed for intuitive display; then the whole environment is partitioned, the weights of each environment parameter are determined by combining the clustering results and the environment semantic information and using hierarchical analysis, and the fuzzy control is used for comprehensive evaluation of the whole traditional dwelling environment, so that the target environment can be evaluated to the greatest extent, the environment quality condition of the traditional dwelling is scientifically and reasonably reflected, data reference is provided for subsequent environment reconstruction and regulation, and compared with the background technical method, the technical efficiency is high, the cost is low and the precision is high.

[0048] The above is only one of the preferred embodiments of the application, and does not limit the application in any other form, and any modification or equivalent change made according to the technical essence of the application still belongs to the scope of the application claimed.

Claims

1. A method for evaluating the environmental quality of traditional dwellings based on semantic information, characterized in that, Comprising the following steps: (1) Multi-source environmental information acquisition and processing; (1.1) Collecting the environment data of traditional residential buildings by using mobile carriers to carry the collecting end, and obtaining the environment data set ; transmitting the collected multi-source data to the cloud server to perform Gaussian smoothing filter processing on the original data, and then performing Kalman filter on the data to obtain reliable environment parameter data; (2) Fine-grained region division based on semantic information; In the region division stage, the specific steps are: (2.1) according to environmental priori physical structure information and mobile carrier positioning information solving semantic labels , giving semantic labels to each sampling data ; (2.2) K-medoids clustering on the processed data, specifically: conditional sorting of the sampling point information under each semantic information, obtaining the sampling points where the maximum and minimum values of each pollutant are located , clustering with these sampling points as initial clustering centers; Calculate the distance between each sample and the cluster center under different semantic environments, and perform class division; Take the center point as the new cluster center and repeat the iteration; (2.3) According to the environmental pollution information clustering results, the traditional residential environment is divided into regions with fine granularity combined with the physical structure of the house ; (2.4) Spatial heat map visualization output is performed on the processed multi-source data to obtain the pollution source distribution and pollution propagation trend in the whole traditional residence environment; (3) Environmental assessment method based on fuzzy comprehensive evaluation; In the environmental assessment stage, the specific steps are: (3.1) The collected traditional residence environmental parameters are fuzzified, the membership functions of each environmental parameter are established, and the membership of each environmental parameter x in the fuzzy subset X is obtained, which is: very low, low, normal, high and very high; Similarly, the fuzzy subset Y and the membership function are established for the environment evaluation set y, and the environment evaluation standard is set as excellent, good, qualified, poor and very poor; (3.2) The membership of each environmental parameter is integrated to establish the fuzzy control rules of environmental parameters and evaluation system ; (3.3) the improved analytic hierarchy process is used to assign weights, and the improved analytic hierarchy process is specifically as follows: firstly, a hierarchical structure model is established; according to the preliminary fine-grained region division result and the two-by-two comparison of each environmental factor according to environmental semantic information, a judgment matrix is obtained , wherein, the judgment matrix is adjusted adaptively according to different environmental semantics; and the maximum eigenvalue and the corresponding characteristic weight vector are calculated on the basis of the judgment matrix. (3.4) The fuzzy comprehensive evaluation result is obtained according to the weight vector and the established fuzzy control rules The weights are added The fuzzy comprehensive evaluation result is obtained Finally, the environment is judged using the maximum membership degree principle to obtain the final environment evaluation: .

2. The traditional dwelling environment quality evaluation method based on semantic information according to claim 1, characterized in that: The specific form of the original multi-source data collected in step (1.1) is: temperature, humidity, carbon dioxide concentration, TVOC concentration, formaldehyde concentration, time and position.

3. The traditional dwelling environment quality evaluation method based on semantic information according to claim 1, characterized in that: The membership function model in step (3.1) selects a trapezoidal function according to a traditional residential environment evaluation system .

Images