Dynamic environment real-time optimization method and system based on intelligent agent

Abstract

The invention relates to the technical field of artificial intelligence, and discloses a dynamic environment real-time optimization method and system based on an intelligent agent, and the method comprises the steps: analyzing the breathing comfort of a control region; the air cleanliness of the management and control area is analyzed, an air quality index corresponding to the management and control area is calculated, and the requirement satisfaction degree of personnel about the environment in the management and control area is analyzed; evaluating the environment regulation efficiency corresponding to the environment regulation equipment, and analyzing the cooperative compatibility between the environment regulation equipment; setting an environment regulation sequence of the control area, and determining environment optimization combination equipment in the environment regulation equipment; and performing environment optimization processing on the management and control area based on the environment optimization combination equipment, monitoring equipment energy consumption corresponding to the environment optimization combination equipment so as to perform energy scheduling processing on the environment optimization combination equipment, and stopping the environment optimization processing until the demand satisfaction degree is higher than a preset threshold value so as to obtain an optimization result. According to the invention, the dynamic environment real-time optimization efficiency based on the intelligent agent can be improved.

Description

Technical Field

[0001] The present invention relates to a method and system for real-time optimization of a dynamic environment based on an intelligent agent, and belongs to the technical field of artificial intelligence. Background Art

[0002] An intelligent agent is an entity with the ability to perceive, learn, make decisions and act. It can be a software program or a hardware device. It can perceive environmental information, make decisions based on its own knowledge and algorithms, and take actions to achieve specific goals. It is often used to optimize the processing of dynamic environments.

[0003] At present, the real-time optimization method of dynamic environment through intelligent agents adopts the method of rule engine and feedback control. This method pre-sets a certain environmental parameter threshold and the corresponding control strategy. When the monitoring data triggers the threshold, the corresponding operation is performed. However, this method does not take into account multiple parameters of the environment, which leads to poor flexibility of environmental optimization. It also does not consider the collaborative work of multiple devices, which leads to unclear optimization effect. Therefore, there is a need for a method that can improve the efficiency of real-time optimization of dynamic environment based on intelligent agents. Summary of the Invention

[0004] The present invention provides a method and system for real-time optimization of a dynamic environment based on an intelligent agent, the main purpose of which is to improve the efficiency of real-time optimization of a dynamic environment based on an intelligent agent.

[0005] To achieve the above objectives, the present invention provides an agent-based real-time optimization method for a dynamic environment, comprising:

[0006] Obtaining the controlled area where the agent is located, collecting regional environmental data corresponding to the controlled area, the regional environmental data including: carbon dioxide concentration data, multi-source environmental data, and environmental personnel data; and analyzing the breathing comfort of the controlled area based on the carbon dioxide concentration data;

[0007] Based on the multi-source environmental data, analyze the air cleanliness of the controlled area, combine the breathing comfort and the air cleanliness to calculate the air quality index corresponding to the controlled area, identify the behavior pattern corresponding to the environmental personnel data, and analyze the satisfaction of the environmental needs of the personnel in the controlled area based on the behavior pattern and the air quality index;

[0008] When the demand satisfaction is not higher than a preset threshold, the agent retrieves the environmental conditioning equipment in the control area, collects the equipment operating parameters corresponding to the environmental conditioning equipment in real time, evaluates the environmental conditioning performance corresponding to the environmental conditioning equipment based on the equipment operating parameters, and analyzes the collaborative compatibility between the environmental conditioning equipment;

[0009] Based on the demand satisfaction, setting the environmental adjustment sequence of the control area, and determining the environmental optimization combination device in the environmental adjustment device in combination with the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility;

[0010] Based on the environmental optimization combination device, environmental optimization processing of the control area is performed, and the energy consumption of the equipment corresponding to the environmental optimization combination device is monitored to perform energy scheduling processing on the environmental optimization combination device until the demand satisfaction is higher than the preset threshold, stop the environmental optimization processing, and obtain the optimization result.

[0011] Optionally, analyzing the breathing comfort of the controlled area based on the carbon dioxide concentration data includes:

[0012] performing data cleaning on the carbon dioxide concentration data to obtain cleaned concentration data;

[0013] Based on a preset health threshold, the cleaning concentration data is marked as exceeding the concentration limit to obtain an exceeding-limit state marking matrix;

[0014] Calculating the exposure risk index corresponding to the exceeding-standard state marking matrix, and performing spatial interpolation processing on the exposure risk index to obtain a respiratory risk heat map;

[0015] Calculating the respiratory pressure coefficient of the controlled area based on the respiratory risk heat map;

[0016] Based on the respiratory pressure coefficient, the respiratory comfort of the controlled area is analyzed.

[0017] Optionally, calculating the respiratory pressure coefficient of the controlled area based on the respiratory risk heat map includes:

[0018] Performing color decoupling processing on the respiratory risk heat map to obtain risk mapping color components;

[0019] Binarizing the color components of the risk map to obtain a binary map of the risk area;

[0020] Counting the number of regional pixels corresponding to the risk area binary map, and calculating the risk coverage in the respiratory risk heat map based on the number of regional pixels;

[0021] Based on the risk coverage, the respiratory pressure coefficient of the control area is calculated using the following formula:

[0022] β=A base ×(1+α×γ)

[0023] Among them, β represents the respiratory pressure coefficient of the control area, A baserepresents the pressure coefficient baseline value, α represents the risk response coefficient, and γ represents the risk coverage.

[0024] Optionally, analyzing the air cleanliness of the controlled area based on the multi-source environmental data includes:

[0025] Performing data noise reduction processing on the multi-source environmental data to obtain noise-reduced multi-source environmental data;

[0026] performing clustering processing on the noise-reduced multi-source environmental data to obtain clustered environmental data;

[0027] Extracting environment characterization parameters corresponding to the clustered environment data, and analyzing parameter semantics corresponding to the environment characterization parameters;

[0028] Based on the parameter semantics, filtering out air purity-related parameters from the environmental characterization parameters;

[0029] Calculating a time series variation value corresponding to the air purity associated parameter, and calculating an environmental fluctuation entropy corresponding to the control area based on the time series variation value;

[0030] Performing parameter scoring processing on the air purity related parameters to obtain parameter scoring values;

[0031] Calculate the air purity index corresponding to the control area by combining the parameter score value and the environmental fluctuation entropy;

[0032] Based on the air purity index, the air cleanliness of the controlled area is analyzed.

[0033] Optionally, identifying the behavior pattern corresponding to the environmental personnel data includes:

[0034] Identify data tags corresponding to the environmental personnel data, and determine behavior tags in the data tags;

[0035] Extracting personnel behavior data from the environmental personnel data based on the behavior tags;

[0036] Extracting behavioral features from the personnel behavior data to obtain key behavioral features;

[0037] Analyzing the dynamic interaction characteristics between the key features of the behavior;

[0038] Based on the dynamic interaction characteristics, the behavior pattern corresponding to the key behavior feature is analyzed.

[0039] Optionally, analyzing the dynamic interaction characteristics between the key behavioral features includes:

[0040] Performing vectorization processing on the key features of the behavior to obtain a behavior feature vector;

[0041] Based on the behavior feature vector, the feature correlation coefficient between the key behavior features is calculated using the following formula:

[0042]

[0043] Among them, E represents the feature correlation coefficient between key behavioral features, F b represents the bth behavior feature vector in the Fth behavior key feature, H d represents the dth behavior feature vector in the Hth behavior key feature, b represents the serial number of the behavior feature vector of the Fth behavior key feature, d represents the serial number of the behavior feature vector of the Hth behavior key feature, m and n represent the number of behavior feature vectors of the Fth behavior key feature and the Hth behavior key feature, respectively. The squared norm of the behavioral feature vector representing the Fth behavioral key feature, The squared norm of the behavior feature vector representing the Hth behavior key feature;

[0044] Based on the feature correlation coefficients, constructing a correlation coefficient matrix between the key features of the behavior;

[0045] Analyzing the feature correlation relationship between the key features of the behaviors based on the correlation coefficient matrix;

[0046] Calculating the feature time series entropy corresponding to the key feature of the behavior, and analyzing the feature change trend corresponding to the key feature of the behavior based on the feature time series entropy;

[0047] The dynamic interaction characteristics between the key behavior features are analyzed in combination with the feature association relationship and the feature change trend.

[0048] Optionally, the evaluating the environmental conditioning performance of the environmental conditioning device based on the device operating parameters includes:

[0049] Identify the device operation identifier corresponding to the device operation parameter and its corresponding identifier association parameter;

[0050] Querying the device operation characteristics corresponding to the environmental conditioning device, and calculating the matching degree between the device operation identifier and the device operation characteristics;

[0051] Based on the matching degree, assigning an identification importance corresponding to the device operation identification, and querying a benchmark operation parameter corresponding to the device operation identification;

[0052] Calculating a parameter deviation of the environmental conditioning device by combining the benchmark operating parameters and the identification-related parameters;

[0053] Calculating the performance deviation value of the environmental conditioning device by combining the parameter deviation and the identification importance;

[0054] Based on the performance deviation value, the environmental adjustment performance corresponding to the environmental adjustment device is evaluated.

[0055] Optionally, analyzing the cooperative compatibility between the environmental adjustment devices includes:

[0056] querying a device adjustment attribute corresponding to the environment adjustment device, performing semantic analysis on the device adjustment attribute, and obtaining an attribute semantic vector;

[0057] Performing cluster analysis on the attribute semantic vectors to obtain the divided device functions corresponding to the environmental adjustment devices;

[0058] Based on the divided device functions, constructing a device association map of the environmental conditioning device;

[0059] Performing information flow simulation on the device association map to determine the data interaction path between the environmental adjustment devices;

[0060] Analyzing the control timing relationship of the environmental conditioning device based on the data interaction path;

[0061] Performing conflict optimization on the regulation timing relationship to obtain an optimized regulation timing relationship;

[0062] The collaborative compatibility between the environmental conditioning devices is determined by combining the optimized control timing relationship with the device association map.

[0063] Optionally, the monitoring of the energy consumption of the equipment corresponding to the environment optimization combination equipment to optimize the energy scheduling of the environment optimization combination equipment further includes:

[0064] Performing filtering on the device energy consumption to obtain filtered device energy consumption;

[0065] Constructing an energy consumption scatter plot corresponding to the energy consumption of the filtering device, and performing fitting processing on the energy consumption scatter plot to obtain an energy consumption curve graph;

[0066] Calculating the slope of an energy consumption curve corresponding to the energy consumption curve graph, and identifying abnormal energy consumption of devices in the environmental optimization combination device based on the slope of the energy consumption curve;

[0067] Analyzing the energy consumption trend of the environmental optimization combined equipment based on the slope of the energy consumption curve;

[0068] The energy scheduling of the environmental optimization combination equipment is optimized based on the energy consumption trend and the abnormal equipment energy consumption.

[0069] In order to solve the above problems, the present invention further provides a dynamic environment real-time optimization system based on an intelligent agent, the system comprising:

[0070] A breathing comfort analysis module is used to obtain the control area where the intelligent agent is located, collect regional environmental data corresponding to the control area, and analyze the breathing comfort of the control area based on the carbon dioxide concentration data.

[0071] a demand satisfaction analysis module for analyzing the air cleanliness of the controlled area based on the multi-source environmental data, calculating an air quality index corresponding to the controlled area in combination with the breathing comfort and the air cleanliness, identifying behavioral patterns corresponding to the environmental personnel data, and analyzing the environmental demand satisfaction of personnel in the controlled area based on the behavioral patterns and the air quality index;

[0072] a collaborative compatibility analysis module configured to, when the degree of satisfaction of the demand is not higher than a preset threshold, retrieve the environmental conditioning equipment of the agent in the control area, collect the equipment operating parameters corresponding to the environmental conditioning equipment in real time, evaluate the environmental conditioning efficiency corresponding to the environmental conditioning equipment based on the equipment operating parameters, and analyze the collaborative compatibility between the environmental conditioning equipment;

[0073] An equipment combination module is used to set the environmental adjustment sequence of the control area based on the demand satisfaction, and determine the environmental optimization combination equipment among the environmental adjustment equipment in combination with the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility;

[0074] The environmental optimization module is used to perform environmental optimization processing on the controlled area based on the environmental optimization combination device, and monitor the energy consumption of the equipment corresponding to the environmental optimization combination device, so as to perform energy scheduling processing on the environmental optimization combination device until the demand satisfaction is higher than the preset threshold, stop the environmental optimization processing, and obtain the optimization result.

[0075] Compared with the problems described in the background technology, the present invention can fully understand the environmental conditions of the area by collecting the regional environmental data corresponding to the control area, analyze the breathing comfort of the control area based on the carbon dioxide concentration data, and then understand the breathing comfort of the people corresponding to the control area in the environment. Furthermore, the present invention can understand the air cleanliness of the control area by analyzing the air cleanliness of the control area based on the multi-source environmental data, and timely discover the pollution source, providing a basis for the intelligent control of air purification equipment, ventilation system optimization and environmental risk warning. It should be understood that when the demand satisfaction is not higher than the preset threshold, it means that the comfort, health and safety of the people in the control area for the environment (such as air quality, etc.) are not effectively met. Then the present invention retrieves the environmental control data of the intelligent body in the control area. Energy-saving equipment, thereby facilitating the subsequent improvement of the environmental quality in the control area through equipment linkage, improving personnel comfort and health and safety protection, further, the present invention determines the environmental optimization combination equipment in the environmental adjustment equipment by combining the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility, and then obtains the best equipment combination, which is convenient for subsequent improvement of the efficiency of environmental optimization processing, further, the present invention monitors the energy consumption of the equipment corresponding to the environmental optimization combination equipment to perform energy scheduling processing on the environmental optimization combination equipment, thereby achieving dynamic adjustment of equipment operation status and energy distribution under the premise of ensuring environmental optimization effect, reducing overall energy consumption costs, and improving energy utilization efficiency, until the demand satisfaction is higher than the preset threshold, stopping the environmental optimization processing, and obtaining the optimization result, thereby achieving improvement in the real-time optimization efficiency of the dynamic environment. Therefore, the real-time optimization method and system for dynamic environment based on intelligent agent provided in the embodiment of the present invention can improve the real-time optimization efficiency of dynamic environment based on intelligent agent. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 A flow chart of a method for real-time optimization of a dynamic environment based on an agent according to an embodiment of the present invention;

[0077] Figure 2 Schematic diagram of the clinical diagnosis and treatment data collection process of the agent-based dynamic environment real-time optimization method provided by the present invention

[0078] Figure 3 A schematic diagram of modules for implementing the agent-based real-time optimization method for a dynamic environment according to an embodiment of the present invention.

[0079] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0080] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0081] The embodiment of the present application provides a method for real-time optimization of a dynamic environment based on an intelligent agent. The execution subject of the method for real-time optimization of a dynamic environment based on an intelligent agent includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for real-time optimization of a dynamic environment based on an intelligent agent can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0082] Example 1:

[0083] Reference Figure 1 FIG. 1 is a flow chart of a method for real-time optimization of a dynamic environment based on an agent according to an embodiment of the present invention. In this embodiment, the method for real-time optimization of a dynamic environment based on an agent includes:

[0084] S1. Obtain the control area where the intelligent agent is located, collect regional environmental data corresponding to the control area, and analyze the breathing comfort of the control area based on the carbon dioxide concentration data.

[0085] The present invention can fully understand the environmental conditions of the area by collecting the regional environmental data corresponding to the control area, analyze the breathing comfort of the control area based on the carbon dioxide concentration data, and then understand the breathing comfort of the personnel corresponding to the control area in the environment, wherein the control area refers to the specific spatial range that the intelligent body is responsible for monitoring and regulating, such as the office area, the carbon dioxide concentration data is the key indicator data corresponding to the control area to measure the degree of influence of air quality on breathing, the multi-source environmental data is the data corresponding to the control area that comprehensively reflects the conditions of multiple environmental factors, such as formaldehyde, benzene, radon, ammonia, TVOC and other pollutants, and the environmental personnel data is the control area's The data should reflect the relevant conditions of the people in the area. The breathing comfort is an indicator of the control area that reflects the degree of comfort people feel when breathing in the area. Furthermore, the regional environmental data corresponding to the control area can be collected through various sensors, such as distributed sensors in multi-layer spaces (floor, bedside, and ceiling deployment can be used in homes, and public passages in public buildings can be installed next to power sockets). Through the OFDMA RU channel of WIFImesh, the subcarrier is the core unit of OFDMA (orthogonal frequency division multiplexing) technology. Each subcarrier independently modulates the data stream to avoid interference through orthogonality. The 26-subcarrier RU corresponds to approximately 2MHz bandwidth, and each subcarrier can independently transmit different environmental data.

[0086] As an embodiment of the present invention, analyzing the breathing comfort of the controlled area based on the carbon dioxide concentration data includes:

[0087] performing data cleaning on the carbon dioxide concentration data to obtain cleaned concentration data;

[0088] Based on a preset health threshold, the cleaning concentration data is marked as exceeding the concentration limit to obtain an exceeding-limit state marking matrix;

[0089] Calculating the exposure risk index corresponding to the exceeding-standard state marking matrix, and performing spatial interpolation processing on the exposure risk index to obtain a respiratory risk heat map;

[0090] Calculating the respiratory pressure coefficient of the controlled area based on the respiratory risk heat map;

[0091] Based on the respiratory pressure coefficient, the respiratory comfort of the controlled area is analyzed.

[0092] Among them, the cleaning concentration data is the data obtained after removing noise, outliers and other interference information from the carbon dioxide concentration data; the preset health threshold is the carbon dioxide concentration limit value set according to the human health standard, which is used to determine whether it exceeds the standard. The exceeding standard status marking matrix is ​​the exceeding standard status information presented in matrix form after the cleaning concentration data is marked as exceeding the concentration standard; the exposure risk index is a quantitative indicator corresponding to the exceeding standard status marking matrix, which is used to measure the degree of health risk of personnel exposed to the current carbon dioxide concentration environment. The respiratory risk heat map is a visual graph that intuitively displays the distribution of respiratory risks at different locations in the control area after the exposure risk index is spatially interpolated; the respiratory pressure coefficient is a quantitative parameter calculated based on the respiratory risk heat map to reflect the degree of pressure exerted on personnel by the respiratory environment in the control area.

[0093] Furthermore, the carbon dioxide concentration data can be cleaned based on statistical analysis and outlier detection algorithms (such as the 3σ criterion) to obtain cleaned concentration data; based on a preset health threshold, the cleaned concentration data can be marked as exceeding the concentration standard by writing a conditional judgment program (using logical judgment structures such as if statements in programming languages) to obtain an exceeding-standard state marking matrix; the exposure risk index corresponding to the exceeding-standard state marking matrix can be calculated by constructing a risk assessment model (such as a formula model that is weighted based on factors such as the duration of exceeding the standard and the number of times exceeding the standard); the exposure risk index can be spatially interpolated using a spatial interpolation algorithm (such as the inverse distance weighted interpolation method, the Kriging interpolation method, etc.) to obtain a respiratory risk heat map; based on the respiratory pressure coefficient, the respiratory comfort of the controlled area can be analyzed according to a pre-set comfort grading standard (such as dividing the pressure coefficient interval into comfortable, relatively comfortable, uncomfortable, etc. levels).

[0094] Furthermore, as an optional embodiment of the present invention, calculating the respiratory pressure coefficient of the controlled area based on the respiratory risk heat map includes:

[0095] Performing color decoupling processing on the respiratory risk heat map to obtain risk mapping color components;

[0096] Binarizing the color components of the risk map to obtain a binary map of the risk area;

[0097] Counting the number of regional pixels corresponding to the risk area binary map, and calculating the risk coverage in the respiratory risk heat map based on the number of regional pixels;

[0098] Based on the risk coverage, the respiratory pressure coefficient of the control area is calculated using the following formula:

[0099] β=A base ×(1+α×γ)

[0100] Among them, β represents the respiratory pressure coefficient of the control area, A base represents the pressure coefficient baseline value, α represents the risk response coefficient, and γ represents the risk coverage.

[0101] Among them, the risk mapping color component is the color information corresponding to the risk level obtained after the respiratory risk heat map is subjected to color decoupling processing; the risk area binary map is an image formed after the risk mapping color component is binarized and uses black and white to distinguish risk areas from non-risk areas; the number of regional pixels is the number of pixels representing risk areas in the risk area binary map; based on the number of regional pixels, the risk coverage is the proportion of risk area pixels to total pixels in the respiratory risk heat map; the pressure coefficient baseline value is a pre-set value used as a basic reference for calculating the respiratory pressure coefficient, which can be obtained by collecting a large amount of relevant respiratory data of the controlled area under no-risk or low-risk conditions, performing statistical analysis, and then taking the mean or median; the risk response coefficient is a parameter determined based on experience or experiments to measure the degree of influence of risk coverage on the respiratory pressure coefficient. The risk response coefficient can be determined by designing experimental scenarios with different risk coverage, observing the changes in the respiratory pressure coefficient, and combining data analysis and expert experience to determine its appropriate value.

[0102] Furthermore, the respiratory risk heat map can be color decoupled by a color space conversion algorithm (such as conversion from RGB to HSV) to obtain a risk mapping color component; the risk mapping color component can be binarized by setting a grayscale threshold (such as the threshold automatically calculated by the Otsu algorithm) to obtain a risk area binary map; the number of regional pixels corresponding to the risk area binary map can be counted by an image pixel statistics function (such as the countNonZero function in OpenCV), and based on the number of regional pixels, the risk coverage in the respiratory risk heat map is calculated by calculating the ratio of the number of pixels in the risk area to the total number of pixels;

[0103] S2. Based on the multi-source environmental data, analyze the air cleanliness of the controlled area, calculate the air quality index corresponding to the controlled area in combination with the breathing comfort and the air cleanliness, identify the behavior pattern corresponding to the environmental personnel data, and analyze the satisfaction of the environmental needs of the personnel in the controlled area based on the behavior pattern and the air quality index.

[0104] By analyzing the air cleanliness of the controlled area based on the multi-source environmental data, the present invention can understand the air cleanliness of the controlled area, promptly discover the pollution source, and provide a basis for intelligent control of air purification equipment, ventilation system optimization and environmental risk warning. Among them, the air cleanliness indicates the cleanliness or purity of pollutants in the air of the controlled area, and reflects the content of pollutants such as impurities, dust, and harmful gases in the air.

[0105] As an embodiment of the present invention, analyzing the air cleanliness of the controlled area based on the multi-source environmental data includes:

[0106] Performing data noise reduction processing on the multi-source environmental data to obtain noise-reduced multi-source environmental data;

[0107] performing clustering processing on the noise-reduced multi-source environmental data to obtain clustered environmental data;

[0108] Extracting environment characterization parameters corresponding to the clustered environment data, and analyzing parameter semantics corresponding to the environment characterization parameters;

[0109] Based on the parameter semantics, filtering out air purity-related parameters from the environmental characterization parameters;

[0110] Calculating a time series variation value corresponding to the air purity associated parameter, and calculating an environmental fluctuation entropy corresponding to the control area based on the time series variation value;

[0111] Performing parameter scoring processing on the air purity related parameters to obtain parameter scoring values;

[0112] Calculate the air purity index corresponding to the control area by combining the parameter score value and the environmental fluctuation entropy;

[0113] Based on the air purity index, the air cleanliness of the controlled area is analyzed.

[0114] Among them, the denoised multi-source environmental data is the data result after noise and outliers are removed from the multi-source environmental data after data denoising processing; the clustered environmental data is a data set divided into different categories based on data similarity after clustering processing of the denoised multi-source environmental data; the environmental characterization parameters are the various indicators corresponding to the clustered environmental data that can reflect environmental characteristics; the parameter semantics are the environmental significance and connotation represented by the indicators corresponding to the environmental characterization parameters; the air purity associated parameters are parameters directly related to the content and purity of pollutants in the air, selected from the environmental characterization parameters; the time series change value is the change in the value of the air purity associated parameters at consecutive time points; the environmental fluctuation entropy is a value that quantifies the dynamic stability of air cleanliness corresponding to the control area based on the time series change of the air purity associated parameters; the parameter score value is a quantitative score obtained after evaluating the air purity associated parameters according to the set rules; and the air purity index is a quantitative evaluation indicator that comprehensively reflects the air cleanliness and dynamic stability corresponding to the control area, calculated by combining the parameter score value and the environmental fluctuation entropy.

[0115] Furthermore, the multi-source environmental data can be subjected to data denoising by a median filtering algorithm to obtain denoised multi-source environmental data; the denoised multi-source environmental data can be clustered by a K-means algorithm to obtain clustered environmental data; the environmental characterisation parameters corresponding to the clustered environmental data can be extracted by a data feature extraction function, and the data feature extraction function is composed of a Java function; the parameter semantics corresponding to the environmental characterisation parameters can be analyzed by natural language processing technology; based on the parameter semantics, air purity associated parameters can be screened out from the environmental characterisation parameters by a preset semantic matching rule; the time series change value corresponding to the air purity associated parameter can be calculated by calculating the difference between data at adjacent time points; the environmental fluctuation entropy corresponding to the control area can be calculated by combining the time series change value with the information entropy calculation formula; the air purity associated parameters can be subjected to parameter scoring processing by comparing with a preset parameter scoring standard table and combining normalisation processing to obtain parameter scoring values, such as dividing the PM2.5 concentration parameter into different intervals (0-35μg / m 3 Excellent, 35-75 μg / m 3is good, etc.), assigned a corresponding score, and then the original value of each parameter is normalized and mapped to the interval of [0, 1] to obtain the final score value of the parameter; combining the parameter score value and the environmental fluctuation entropy, the air purity index corresponding to the control area is calculated by a weighted summation formula; based on the value of the air purity index, the air cleanliness of the control area is analyzed; if the air purity index is greater than 0.8, the air cleanliness of the control area is excellent; if the index is 0.6-0.8, the cleanliness is good; if the index is 0.4-0.6, the cleanliness is lightly polluted; if the index is lower than 0.4, the cleanliness is moderately polluted or above.

[0116] The present invention calculates the air quality index corresponding to the control area by combining the breathing comfort and the air cleanliness, so as to understand the comprehensive air quality status of the control area, and further provide data support for the subsequent demand satisfaction analysis, wherein the air quality index is a quantitative indicator corresponding to the control area that comprehensively reflects the pollutant content, cleanliness and impact on human breathing comfort in the air. Furthermore, the breathing comfort and the air cleanliness are standardized and summed to obtain the air quality index corresponding to the control area.

[0117] By identifying the behavioral patterns corresponding to the environmental personnel data, the present invention can timely gain insights into the activity patterns and potential needs of personnel in the environment, provide strong support for improving environmental management efficiency, and achieve efficient coordination between the environment and personnel activities. Among them, the behavioral pattern is a collective expression of the action patterns, activity preferences and interaction methods of personnel in a specific environment presented by analyzing the environmental personnel data.

[0118] As an embodiment of the present invention, identifying the behavior pattern corresponding to the environmental personnel data includes:

[0119] Identify data tags corresponding to the environmental personnel data, and determine behavior tags in the data tags;

[0120] Extracting personnel behavior data from the environmental personnel data based on the behavior tags;

[0121] Extracting behavioral features from the personnel behavior data to obtain key behavioral features;

[0122] Analyzing the dynamic interaction characteristics between the key features of the behavior;

[0123] Based on the dynamic interaction characteristics, the behavior pattern corresponding to the key behavior feature is analyzed.

[0124] Among them, the data tag is a mark corresponding to the environmental personnel data for identifying and classifying data attributes, the behavior tag is a mark in the data tag for indicating the attributes related to the specific behavior of the personnel, the personnel behavior data is based on the behavior tag, and is extracted from the environmental personnel data based on the data part directly related to the personnel behavior. The key behavior feature is the data feature that can reflect the essential characteristics of the personnel behavior after the feature extraction of the personnel behavior data. The dynamic interaction characteristics are the mutual connection, influence and change trend of the key behavior features over time.

[0125] Furthermore, the data labels corresponding to the environmental personnel data can be identified by a label recognition tool, which is compiled by a programming language; the behavioral labels in the data labels can be determined by natural language processing technology; based on the behavioral labels, the personnel behavior data can be extracted from the environmental personnel data by a data screening algorithm, which is compiled by a programming language; the behavioral features of the personnel behavior data can be extracted by a convolutional neural network to obtain key behavioral features, such as through structures such as convolutional layers, pooling layers and fully connected layers, to automatically extract local features and spatial hierarchical structures in the data. In the processing of personnel behavior data, for spatial information such as human posture and movement in video data, CNN can extract features by sliding the convolution kernel on the image or video frame, such as detecting key information such as the position of human joints; based on the dynamic interaction characteristics, the behavior patterns corresponding to the key behavioral features are analyzed, and hierarchical aggregation can be performed according to the similarity of the key behavioral features by adopting a hierarchical clustering algorithm, and the behavior patterns can be divided in combination with the dynamic interaction characteristics.

[0126] Furthermore, as an optional embodiment of the present invention, analyzing the dynamic interaction characteristics between the key behavioral features includes:

[0127] Performing vectorization processing on the key features of the behavior to obtain a behavior feature vector;

[0128] Based on the behavior feature vector, the feature correlation coefficient between the key behavior features is calculated using the following formula:

[0129]

[0130] Among them, E represents the feature correlation coefficient between key behavioral features, F b represents the bth behavior feature vector in the Fth behavior key feature, H drepresents the dth behavior feature vector in the Hth behavior key feature, b represents the serial number of the behavior feature vector of the Fth behavior key feature, d represents the serial number of the behavior feature vector of the Hth behavior key feature, m and n represent the number of behavior feature vectors of the Fth behavior key feature and the Hth behavior key feature, respectively. The squared norm of the behavioral feature vector representing the Fth behavioral key feature, The squared norm of the behavior feature vector representing the Hth behavior key feature;

[0131] Based on the feature correlation coefficients, constructing a correlation coefficient matrix between the key features of the behavior;

[0132] Analyzing the feature correlation relationship between the key features of the behaviors based on the correlation coefficient matrix;

[0133] Calculating the feature time series entropy corresponding to the key feature of the behavior, and analyzing the feature change trend corresponding to the key feature of the behavior based on the feature time series entropy;

[0134] The dynamic interaction characteristics between the key behavior features are analyzed in combination with the feature association relationship and the feature change trend.

[0135] Among them, the behavioral feature vector is the numerical representation of the behavioral key feature; the feature correlation coefficient represents the correlation strength between the behavioral key features; the correlation coefficient matrix is ​​the feature correlation strength matrix constructed based on the feature correlation coefficient; the feature correlation relationship is the mutual influence relationship between the behavioral key features; the feature time series entropy is the uncertainty measure of the time series data corresponding to the behavioral key feature; the feature change trend is the dynamic evolution law of the time series data corresponding to the behavioral key feature.

[0136] Furthermore, the key behavioral features can be vectorized through feature encoding (such as one-hot encoding, word vector embedding) to obtain behavioral feature vectors; based on the feature correlation coefficient, the correlation coefficient matrix between the key behavioral features can be constructed through matrix filling and organization; based on the correlation coefficient matrix, the feature correlation relationship between the key behavioral features can be analyzed through network visualization; the feature time series entropy corresponding to the key behavioral features can be calculated through the Shannon entropy formula; based on the feature time series entropy, the feature change trend corresponding to the key behavioral features can be analyzed through time series analysis or trend fitting, such as calculating the entropy value fluctuations in different time periods through a sliding window, identifying the periodic changes in feature uncertainty (such as the regular differences between weekdays / weekends) to predict the long-term evolution direction of feature complexity; combining the feature correlation relationship and the feature change trend, the dynamic interaction characteristics between the key behavioral features can be analyzed through comprehensive evaluation, such as comparing the time series changes of feature correlation strength and entropy value fluctuations, to determine whether the correlation relationship is accompanied by an increase or decrease in feature uncertainty (such as a sudden drop in entropy when a strong correlation occurs, indicating that the synergy between features is enhanced).

[0137] The present invention analyzes the satisfaction of the needs of the personnel in the control area with respect to the environment based on the behavior pattern and the air quality index, and can understand the comprehensive satisfaction of the personnel in the control area with respect to the environmental comfort, thereby providing data support for the subsequent optimization of environmental governance. The satisfaction of the needs is the degree to which the personnel in the control area have achieved their expectations of the environment (such as air quality, etc.) in terms of comfort, health and safety. Furthermore, based on the behavior pattern and the air quality index, the satisfaction of the needs of the personnel in the control area with respect to the environment is analyzed, such as the statistical average of the air quality index during the period of high personnel gathering, and the comparison of the environmental comfort threshold (such as PM2.5 ≤ 35 μg / m 3 or divide the area into different functional areas (such as office areas and leisure areas), cross-analyze the length of stay of personnel and the air quality level of the corresponding area, locate low satisfaction areas (such as areas with a stay of more than 2 hours and slightly polluted air quality), and comprehensively judge the degree of demand satisfaction in each area and the whole area.

[0138] S3. When the demand satisfaction is not higher than the preset threshold, the environmental conditioning equipment of the intelligent body in the control area is called up, and the equipment operating parameters corresponding to the environmental conditioning equipment are collected in real time. Based on the equipment operating parameters, the environmental conditioning performance corresponding to the environmental conditioning equipment is evaluated, and the collaborative compatibility between the environmental conditioning equipment is analyzed.

[0139] It should be understood that when the demand satisfaction is not higher than the preset threshold, it means that the comfort, health and safety needs of the people in the control area for the environment (such as air quality, etc.) are not effectively met. The present invention then calls the environmental conditioning equipment of the intelligent body in the control area, thereby facilitating the subsequent improvement of the environmental quality in the control area through equipment linkage, thereby improving the comfort and health and safety of the people. Among them, the preset threshold is the minimum qualification standard for the demand satisfaction (that is, the critical value for judging whether the demand is effectively met); the environmental conditioning equipment is the execution device of the intelligent body in the control area for improving the environmental quality (such as air conditioning, air purifier, fresh air system, etc.); further, the environmental conditioning equipment of the intelligent body in the control area can be called through the Internet of Things platform.

[0140] The present invention evaluates the environmental conditioning efficiency corresponding to the environmental conditioning equipment based on the equipment operating parameters, thereby understanding the actual improvement ability and working status of the environmental conditioning equipment on the environmental quality of the controlled area, thereby facilitating the determination of subsequent environmental optimization combination equipment, wherein the equipment operating parameters are real-time status data of the environmental conditioning equipment during operation (such as power, working hours, output indicators, etc.), and the environmental conditioning efficiency is the ability of the environmental conditioning equipment to improve environmental indicators (such as pollutant removal rate, temperature and humidity control accuracy, etc.). Furthermore, the equipment operating parameters corresponding to the environmental conditioning equipment can be collected in real time through a sensor network.

[0141] As an embodiment of the present invention, the step of evaluating the environmental conditioning performance of the environmental conditioning device based on the device operating parameters includes:

[0142] Identify the device operation identifier corresponding to the device operation parameter and its corresponding identifier association parameter;

[0143] Querying the device operation characteristics corresponding to the environmental conditioning device, and calculating the matching degree between the device operation identifier and the device operation characteristics;

[0144] Based on the matching degree, assigning an identification importance corresponding to the device operation identification, and querying a benchmark operation parameter corresponding to the device operation identification;

[0145] Calculating a parameter deviation of the environmental conditioning device by combining the benchmark operating parameters and the identification-related parameters;

[0146] Calculating the performance deviation value of the environmental conditioning device by combining the parameter deviation and the identification importance;

[0147] Based on the performance deviation value, the environmental adjustment performance corresponding to the environmental adjustment device is evaluated.

[0148] Among them, the equipment operation identifier is a unique identity tag corresponding to the equipment operation parameters (such as equipment number, operation mode code); the identifier-associated parameter is the specific status data corresponding to the equipment operation identifier in the equipment operation parameters (such as power and duration under the mode, etc.); the equipment operation characteristic is the identifiable behavioral attribute of the operation mode corresponding to the environmental conditioning device; the matching degree indicates the degree of fit between the equipment operation identifier and the equipment operation characteristic (such as whether the identification mode matches the actual functional output); the identification importance indicates the function priority or influence degree corresponding to the equipment operation identifier (such as the core function identifier has a higher weight); the benchmark operation parameter is the standard design parameter corresponding to the equipment operation identifier (such as rated power, theoretical purification efficiency); the parameter deviation indicates the difference rate between the actual parameters of the environmental conditioning device and the benchmark parameters; the efficiency deviation value indicates the quantitative difference between the comprehensive efficiency of the environmental conditioning device and the design target.

[0149] Furthermore, the device operation identifier and its corresponding identifier-associated parameters corresponding to the device operation parameters can be identified through an identification algorithm, and the identification algorithm is compiled in JAVA language; the device operation characteristics corresponding to the environmental control device can be queried through the background of the device management system; the matching degree between the device operation identifier and the device operation characteristics can be calculated through the cosine similarity algorithm; based on the matching degree, the hierarchical analysis method is used to assign the identifier importance corresponding to the device operation identifier; the benchmark operation parameters corresponding to the device operation identifier can be queried through the device standard parameter library; in combination with the benchmark operation parameters and the identifier-associated parameters, the mean square error formula is used to calculate the parameter deviation of the environmental control device; in combination with the parameter deviation and the identifier importance, the efficiency deviation value of the environmental control device is calculated through a weighted sum model; based on the efficiency deviation value, the environmental control efficiency corresponding to the environmental control device is evaluated according to a pre-set efficiency evaluation standard.

[0150] By analyzing the collaborative compatibility between the environmental conditioning devices, the present invention can understand the degree of functional complementarity when the devices are operated in conjunction, thereby facilitating the determination of subsequent environmental optimization combination devices, wherein the collaborative compatibility is the degree of mutual adaptation between the environmental conditioning devices in terms of functional coordination and operating logic.

[0151] As an embodiment of the present invention, analyzing the cooperative compatibility between the environmental adjustment devices includes:

[0152] querying a device adjustment attribute corresponding to the environment adjustment device, performing semantic analysis on the device adjustment attribute, and obtaining an attribute semantic vector;

[0153] Performing cluster analysis on the attribute semantic vectors to obtain the divided device functions corresponding to the environmental adjustment devices;

[0154] Based on the divided device functions, constructing a device association map of the environmental conditioning device;

[0155] Performing information flow simulation on the device association map to determine the data interaction path between the environmental adjustment devices;

[0156] Analyzing the control timing relationship of the environmental conditioning device based on the data interaction path;

[0157] Performing conflict optimization on the regulation timing relationship to obtain an optimized regulation timing relationship;

[0158] The collaborative compatibility between the environmental conditioning devices is determined by combining the optimized control timing relationship with the device association map.

[0159] Among them, the device adjustment attributes are the adjustment functions and parameter characteristics of the device; the attribute semantic vector is a digital expression after semantic analysis of the attribute; the division of device functions is to classify the devices according to the attributes; the device association map is a visual network of device function associations; the data interaction path is the data transmission channel between devices; the regulation timing relationship is the time sequence logic of the device operation; the optimization of the regulation timing relationship is the coordination plan after eliminating the timing conflict.

[0160] Furthermore, the device adjustment properties corresponding to the environmental adjustment device can be queried through the device attribute database; the device adjustment properties can be semantically parsed through a natural language processing algorithm library (such as NLTK, SpaCy) to obtain an attribute semantic vector; the attribute semantic vector can be clustered through a clustering analysis tool (such as the K-Means algorithm in Scikit-learn) to obtain the divided device functions corresponding to the environmental adjustment device; based on the divided device functions, a knowledge graph construction platform (such as Neo4j) can be used to construct a device association graph of the environmental adjustment device; the device association graph can be simulated for information flow through network simulation software (such as OMNeT++) to determine the environmental adjustment device. The data interaction path between the devices is determined; based on the data interaction path, a timing analysis model (such as Petri net) is used to analyze the control timing relationship of the environmental regulation device; the control timing relationship can be conflict optimized by a conflict resolution algorithm (such as genetic algorithm optimization) to obtain an optimized control timing relationship; combining the optimized control timing relationship with the device association map, the hierarchical analysis method is used to determine the collaborative compatibility between the environmental regulation devices, first determine the weight of each device in the functional collaboration according to the device association map, and then score the operation of each device according to the coordination fluency index in the optimized control timing relationship, and finally combine the weight with the score to calculate a comprehensive score reflecting the collaborative compatibility between the environmental regulation devices, so as to evaluate the collaborative compatibility level.

[0161] S4. Based on the demand satisfaction, set the environmental adjustment sequence of the control area, and determine the environmental optimization combination device in the environmental adjustment device in combination with the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility.

[0162] The present invention determines the environmental optimization combination equipment in the environmental adjustment equipment by combining the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility, and then obtains the best equipment combination, which is convenient for improving the efficiency of the environmental optimization processing in the future. The environmental adjustment sequence is the order of environmental parameter adjustment operations performed in stages and according to priority in the control area, which is used to systematically solve environmental needs. The environmental optimization combination equipment is the environmental adjustment equipment combined with the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility after multi-dimensional evaluation, and the equipment combination that can efficiently and coordinately meet the environmental control goals is screened out. Furthermore, based on the degree of satisfaction of the requirements, the equipment combination is set through priority sorting. The environmental adjustment sequence of the control area; combined with the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility, the environmental optimization combination equipment in the environmental adjustment equipment is determined through a multi-objective optimization model, and the environmental adjustment sequence, the environmental adjustment efficiency, and the equipment collaborative compatibility are used as objective functions. With equipment operating energy consumption, equipment life loss, budget cost, etc. as constraints, the particle swarm optimization algorithm is used to solve and screen out the environmental adjustment equipment combination with the best comprehensive performance. In addition, a device control matrix can be constructed based on the MQTT protocol to achieve millisecond-level automatic response of heterogeneous equipment such as air conditioners, air purifiers, fresh air, and humidifiers. The indoor environment intelligent body forms a closed loop through perception-decision-execution.

[0163] S5. Based on the environmental optimization combination device, perform environmental optimization processing on the controlled area, and monitor the energy consumption of the equipment corresponding to the environmental optimization combination device to optimize the energy scheduling of the environmental optimization combination device until the demand satisfaction is higher than the preset threshold, stop the environmental optimization processing, and obtain the optimization result.

[0164] The present invention monitors the energy consumption of the equipment corresponding to the environmental optimization combination equipment to perform energy scheduling processing on the environmental optimization combination equipment, thereby dynamically adjusting the equipment operation status and energy distribution under the premise of ensuring the environmental optimization effect, reducing the overall energy consumption cost, and improving energy utilization efficiency. Until the demand satisfaction degree exceeds the preset threshold, the environmental optimization processing is stopped to obtain the optimization result, thereby improving the real-time optimization efficiency of the dynamic environment. The equipment energy consumption is the total amount and unit time consumption of various energy sources such as electrical energy, thermal energy, and mechanical energy consumed by the environmental optimization combination equipment during operation. Furthermore, based on the environmental optimization combination equipment, the environmental optimization processing of the control area is performed. The specific process can be referred to. Figure 2 As shown in FIG, it is a schematic diagram of the environment optimization process in the real-time optimization method of a dynamic environment based on an intelligent agent provided by the present invention. It should be noted that, in the present invention, Figure 2The presented environment optimization processing flow of the agent-based dynamic environment real-time optimization method is not limited to the environment optimization processing of the agent-based dynamic environment real-time optimization method in actual different application scenarios.

[0165] As an embodiment of the present invention, the monitoring of the energy consumption of the equipment corresponding to the environmental optimization combination equipment to optimize the energy scheduling of the environmental optimization combination equipment further includes:

[0166] Performing filtering on the device energy consumption to obtain filtered device energy consumption;

[0167] Constructing an energy consumption scatter plot corresponding to the energy consumption of the filtering device, and performing fitting processing on the energy consumption scatter plot to obtain an energy consumption curve graph;

[0168] Calculating the slope of an energy consumption curve corresponding to the energy consumption curve graph, and identifying abnormal energy consumption of devices in the environmental optimization combination device based on the slope of the energy consumption curve;

[0169] Analyzing the energy consumption trend of the environmental optimization combined equipment based on the slope of the energy consumption curve;

[0170] The energy scheduling of the environmental optimization combination equipment is optimized based on the energy consumption trend and the abnormal equipment energy consumption.

[0171] Among them, the energy consumption of the filtering device is more accurate energy consumption data obtained after filtering the energy consumption of the device, removing noise interference and smoothing the fluctuating data; the energy consumption scatter plot is a visual chart corresponding to the energy consumption of the filtering device, which intuitively presents the distribution and change relationship of energy consumption values ​​in the form of data points; the energy consumption curve chart is a continuous curve formed by connecting data points through mathematical methods after fitting the energy consumption scatter plot, which is used to show the trend of energy consumption changes; the slope of the energy consumption curve is a quantitative indicator corresponding to the energy consumption curve chart that reflects the rate of change of energy consumption per unit time, reflecting the speed of energy consumption increase or decrease; the energy consumption of the abnormal equipment is the energy consumption data in the environmental optimization combination equipment that deviates from the normal operating energy consumption range and may indicate equipment failure or inefficient operation; the energy consumption trend is the overall reflection of the change trend and law of energy consumption data of the environmental optimization combination equipment within a certain time span.

[0172] Furthermore, the energy consumption of the equipment can be filtered using a Kalman filter algorithm or a median filter algorithm to obtain the energy consumption of the filtered equipment; an energy consumption scatter plot corresponding to the energy consumption of the filtered equipment can be constructed using an Excel chart function; the energy consumption scatter plot can be fitted using a least squares fitting tool or a curve fitting module of Scikit-learn to obtain an energy consumption curve graph; the slope of the energy consumption curve corresponding to the energy consumption curve graph can be calculated using a numerical differential calculation method; based on the slope of the energy consumption curve, the energy consumption of abnormal equipment in the environmental optimization combination equipment can be identified using an isolation forest algorithm; based on the slope of the energy consumption curve, the energy consumption trend of the environmental optimization combination equipment can be analyzed using a time series analysis model; based on the energy consumption trend and the energy consumption of the abnormal equipment, a reinforcement learning algorithm can be used to dynamically adjust the equipment operating parameters, and equipment with abnormal energy consumption can be shut down or repaired first. The equipment start-stop timing and power distribution can be optimized based on the energy consumption trend prediction. At the same time, a genetic algorithm can be used to iteratively search for the optimal energy consumption solution under the premise of meeting environmental control requirements to achieve energy scheduling optimization of the environmental optimization combination equipment.

[0173] Furthermore, in the present invention, the environmental optimization process is stopped until the demand satisfaction is higher than a preset threshold, and an optimization result is obtained, such as maintaining the PMV thermal comfort index within the range of ±0.5. It has been proven that the satisfaction of hotel customers with personalized configurations that meet the PMV thermal comfort index will be greatly improved.

[0174] Compared with the problems described in the background technology, the present invention can fully understand the environmental conditions of the area by collecting the regional environmental data corresponding to the control area, analyze the breathing comfort of the control area based on the carbon dioxide concentration data, and then understand the breathing comfort of the people corresponding to the control area in the environment. Furthermore, the present invention can understand the air cleanliness of the control area by analyzing the air cleanliness of the control area based on the multi-source environmental data, and timely discover the pollution source, providing a basis for the intelligent control of air purification equipment, ventilation system optimization and environmental risk warning. It should be understood that when the demand satisfaction is not higher than the preset threshold, it means that the comfort, health and safety of the people in the control area for the environment (such as air quality, etc.) are not effectively met. Then the present invention retrieves the environmental control data of the intelligent body in the control area. Energy-saving equipment, thereby facilitating the subsequent improvement of the environmental quality in the control area through equipment linkage, improving personnel comfort and health and safety protection, further, the present invention determines the environmental optimization combination equipment in the environmental adjustment equipment by combining the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility, and then obtains the best equipment combination, which is convenient for subsequent improvement of the efficiency of environmental optimization processing, further, the present invention monitors the energy consumption of the equipment corresponding to the environmental optimization combination equipment to perform energy scheduling processing on the environmental optimization combination equipment, thereby achieving dynamic adjustment of equipment operation status and energy distribution under the premise of ensuring environmental optimization effect, reducing overall energy consumption costs, and improving energy utilization efficiency, until the demand satisfaction is higher than the preset threshold, stopping the environmental optimization processing, and obtaining the optimization result, thereby achieving improvement in the real-time optimization efficiency of the dynamic environment. Therefore, the real-time optimization method and system for dynamic environment based on intelligent agent provided in the embodiment of the present invention can improve the real-time optimization efficiency of dynamic environment based on intelligent agent.

[0175] Example 2:

[0176] like Figure 3 The figure shows a functional module diagram of a dynamic environment real-time optimization system based on intelligent agents according to the present invention.

[0177] The agent-based dynamic environment real-time optimization system 300 described in the present invention can be installed in an electronic device. Depending on the functionality implemented, the agent-based dynamic environment real-time optimization system can include a breathing comfort analysis module 301, a demand satisfaction analysis module 302, a collaborative compatibility analysis module 303, a device combination module 304, and an environment optimization module 305. The modules described in the present invention, also referred to as units, are a series of computer program segments that can be executed by an electronic device processor and perform fixed functions, and are stored in the electronic device's memory.

[0178] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0179] The breathing comfort analysis module 301 is used to obtain the control area where the intelligent agent is located, collect regional environmental data corresponding to the control area, and analyze the breathing comfort of the control area based on the carbon dioxide concentration data.

[0180] The demand satisfaction analysis module 302 is configured to analyze the air cleanliness of the controlled area based on the multi-source environmental data, calculate the air quality index corresponding to the controlled area by combining the breathing comfort and the air cleanliness, identify the behavior patterns corresponding to the environmental personnel data, and analyze the environmental demand satisfaction of the personnel in the controlled area based on the behavior patterns and the air quality index;

[0181] The collaborative compatibility analysis module 303 is configured to, when the demand satisfaction is not higher than a preset threshold, call the agent's environmental conditioning equipment in the control area, collect the equipment operating parameters corresponding to the environmental conditioning equipment in real time, evaluate the environmental conditioning performance of the environmental conditioning equipment based on the equipment operating parameters, and analyze the collaborative compatibility between the environmental conditioning equipment;

[0182] The device combination module 304 is configured to set an environment adjustment sequence for the control area based on the demand satisfaction, and determine an environment optimization combination device among the environment adjustment devices in combination with the environment adjustment sequence, the environment adjustment efficiency, and the collaborative compatibility;

[0183] The environmental optimization module 305 is used to perform environmental optimization processing on the control area based on the environmental optimization combination device, and monitor the equipment energy consumption corresponding to the environmental optimization combination device to perform energy scheduling processing on the environmental optimization combination device until the demand satisfaction is higher than the preset threshold, stop the environmental optimization processing, and obtain the optimization result.

[0184] In detail, the modules in the agent-based dynamic environment real-time optimization system 300 in the embodiment of the present invention are used in the same manner as above. Figure 1 The same technical means are used as the real-time optimization method for dynamic environments based on intelligent agents described in , and can produce the same technical effects, so they will not be repeated here.

[0185] 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 present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0186] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A real-time optimization method for dynamic environments based on an intelligent agent, characterized in that: The method comprises: Obtaining the controlled area where the agent is located, collecting regional environmental data corresponding to the controlled area, the regional environmental data including: carbon dioxide concentration data, multi-source environmental data, and environmental personnel data; and analyzing the breathing comfort of the controlled area based on the carbon dioxide concentration data; Based on the multi-source environmental data, analyze the air cleanliness of the controlled area, combine the breathing comfort and the air cleanliness to calculate the air quality index corresponding to the controlled area, identify the behavior pattern corresponding to the environmental personnel data, and analyze the satisfaction of the environmental needs of the personnel in the controlled area based on the behavior pattern and the air quality index; When the demand satisfaction is not higher than a preset threshold, the agent retrieves the environmental conditioning equipment in the control area, collects the equipment operating parameters corresponding to the environmental conditioning equipment in real time, evaluates the environmental conditioning performance corresponding to the environmental conditioning equipment based on the equipment operating parameters, and analyzes the collaborative compatibility between the environmental conditioning equipment; Based on the demand satisfaction, setting the environmental adjustment sequence of the control area, and determining the environmental optimization combination device in the environmental adjustment device in combination with the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility; Based on the environmental optimization combination device, environmental optimization processing of the control area is performed, and the energy consumption of the equipment corresponding to the environmental optimization combination device is monitored to perform energy scheduling processing on the environmental optimization combination device until the demand satisfaction is higher than the preset threshold, stop the environmental optimization processing, and obtain the optimization result.

2. The agent-based real-time optimization method for a dynamic environment according to claim 1, wherein: The analyzing the breathing comfort of the controlled area based on the carbon dioxide concentration data includes: performing data cleaning on the carbon dioxide concentration data to obtain cleaned concentration data; Based on a preset health threshold, the cleaning concentration data is marked as exceeding the concentration limit to obtain an exceeding-limit state marking matrix; Calculating the exposure risk index corresponding to the exceeding-standard state marking matrix, and performing spatial interpolation processing on the exposure risk index to obtain a respiratory risk heat map; Calculating the respiratory pressure coefficient of the controlled area based on the respiratory risk heat map; Based on the respiratory pressure coefficient, the respiratory comfort of the controlled area is analyzed.

3. The agent-based real-time optimization method for a dynamic environment according to claim 2, wherein: The calculating of the respiratory pressure coefficient of the control area based on the respiratory risk heat map includes: Performing color decoupling processing on the respiratory risk heat map to obtain risk mapping color components; Binarizing the color components of the risk map to obtain a binary map of the risk area; Counting the number of regional pixels corresponding to the risk area binary map, and calculating the risk coverage in the respiratory risk heat map based on the number of regional pixels; Based on the risk coverage, the respiratory pressure coefficient of the control area is calculated using the following formula: β=A base ×(1+α×γ) Among them, β represents the respiratory pressure coefficient of the control area, A base represents the pressure coefficient baseline value, α represents the risk response coefficient, and γ represents the risk coverage.

4. The agent-based real-time optimization method for a dynamic environment according to claim 1, wherein: Analyzing the air cleanliness of the controlled area based on the multi-source environmental data includes: Performing data noise reduction processing on the multi-source environmental data to obtain noise-reduced multi-source environmental data; Performing clustering processing on the noise-reduced multi-source environmental data to obtain clustered environmental data; Extracting environment characterization parameters corresponding to the clustered environment data, and analyzing parameter semantics corresponding to the environment characterization parameters; Based on the parameter semantics, filtering out air purity-related parameters from the environmental characterization parameters; Calculating a time series variation value corresponding to the air purity associated parameter, and calculating an environmental fluctuation entropy corresponding to the control area based on the time series variation value; Performing parameter scoring processing on the air purity related parameters to obtain parameter scoring values; Calculate the air purity index corresponding to the control area by combining the parameter score value and the environmental fluctuation entropy; Based on the air purity index, the air cleanliness of the controlled area is analyzed.

5. The agent-based real-time optimization method for a dynamic environment according to claim 1, wherein: The identifying the behavior pattern corresponding to the environmental personnel data includes: Identify data tags corresponding to the environmental personnel data, and determine behavior tags in the data tags; Extracting personnel behavior data from the environmental personnel data based on the behavior tags; Extracting behavioral features from the personnel behavior data to obtain key behavioral features; Analyzing the dynamic interaction characteristics between the key features of the behavior; Based on the dynamic interaction characteristics, the behavior pattern corresponding to the key behavior feature is analyzed.

6. The agent-based real-time optimization method for a dynamic environment according to claim 5, wherein: The analyzing the dynamic interaction characteristics between the key behavior features includes: Performing vectorization processing on the key features of the behavior to obtain a behavior feature vector; Based on the behavior feature vector, the feature correlation coefficient between the key behavior features is calculated using the following formula: Among them, E represents the feature correlation coefficient between key behavioral features, F b represents the bth behavior feature vector in the Fth behavior key feature, H d represents the dth behavior feature vector in the Hth behavior key feature, b represents the serial number of the behavior feature vector of the Fth behavior key feature, d represents the serial number of the behavior feature vector of the Hth behavior key feature, m and n represent the number of behavior feature vectors of the Fth behavior key feature and the Hth behavior key feature, respectively. The squared norm of the behavioral feature vector representing the Fth behavioral key feature, The squared norm of the behavior feature vector representing the Hth behavior key feature; Based on the feature correlation coefficients, constructing a correlation coefficient matrix between the key features of the behavior; Analyzing the feature correlation relationship between the key features of the behaviors based on the correlation coefficient matrix; Calculating the feature time series entropy corresponding to the key feature of the behavior, and analyzing the feature change trend corresponding to the key feature of the behavior based on the feature time series entropy; The dynamic interaction characteristics between the key behavior features are analyzed in combination with the feature association relationship and the feature change trend.

7. The agent-based real-time optimization method for a dynamic environment according to claim 1, wherein: The evaluating the environmental conditioning performance of the environmental conditioning device based on the device operating parameters includes: Identify the device operation identifier corresponding to the device operation parameter and its corresponding identifier association parameter; Querying the device operation characteristics corresponding to the environmental conditioning device, and calculating the matching degree between the device operation identifier and the device operation characteristics; Based on the matching degree, assigning an identification importance corresponding to the device operation identification, and querying a benchmark operation parameter corresponding to the device operation identification; Calculating a parameter deviation of the environmental conditioning device by combining the benchmark operating parameters and the identification-related parameters; Calculating the performance deviation value of the environmental conditioning device by combining the parameter deviation and the identification importance; Based on the performance deviation value, the environmental adjustment performance corresponding to the environmental adjustment device is evaluated.

8. The agent-based real-time optimization method for a dynamic environment according to claim 1, wherein: The analyzing the cooperative compatibility between the environmental adjustment devices includes: querying a device adjustment attribute corresponding to the environment adjustment device, performing semantic analysis on the device adjustment attribute, and obtaining an attribute semantic vector; Performing cluster analysis on the attribute semantic vectors to obtain the divided device functions corresponding to the environmental adjustment devices; Based on the divided device functions, constructing a device association map of the environmental conditioning device; Performing information flow simulation on the device association map to determine the data interaction path between the environmental adjustment devices; Analyzing the control timing relationship of the environmental conditioning device based on the data interaction path; Performing conflict optimization on the regulation timing relationship to obtain an optimized regulation timing relationship; The collaborative compatibility between the environmental conditioning devices is determined by combining the optimized control timing relationship with the device association map.

9. The agent-based real-time optimization method for a dynamic environment according to claim 1, wherein: The monitoring of the energy consumption of the equipment corresponding to the environmental optimization combination equipment to optimize the energy scheduling of the environmental optimization combination equipment further includes: Performing filtering on the device energy consumption to obtain filtered device energy consumption; Constructing an energy consumption scatter plot corresponding to the energy consumption of the filtering device, and performing fitting processing on the energy consumption scatter plot to obtain an energy consumption curve graph; Calculating the slope of an energy consumption curve corresponding to the energy consumption curve graph, and identifying abnormal energy consumption of devices in the environmental optimization combination device based on the slope of the energy consumption curve; Analyzing the energy consumption trend of the environmental optimization combined equipment based on the slope of the energy consumption curve; The energy scheduling of the environmental optimization combination equipment is optimized based on the energy consumption trend and the abnormal equipment energy consumption.

10. A dynamic environment real-time optimization system based on intelligent agents, characterized in that: The system comprises: A breathing comfort analysis module is used to obtain the control area where the intelligent agent is located, collect regional environmental data corresponding to the control area, and analyze the breathing comfort of the control area based on the carbon dioxide concentration data. a demand satisfaction analysis module for analyzing the air cleanliness of the controlled area based on the multi-source environmental data, calculating an air quality index corresponding to the controlled area in combination with the breathing comfort and the air cleanliness, identifying behavioral patterns corresponding to the environmental personnel data, and analyzing the environmental demand satisfaction of personnel in the controlled area based on the behavioral patterns and the air quality index; a collaborative compatibility analysis module configured to, when the degree of satisfaction of the demand is not higher than a preset threshold, retrieve the environmental conditioning equipment of the agent in the control area, collect the equipment operating parameters corresponding to the environmental conditioning equipment in real time, evaluate the environmental conditioning efficiency corresponding to the environmental conditioning equipment based on the equipment operating parameters, and analyze the collaborative compatibility between the environmental conditioning equipment; An equipment combination module is used to set the environmental adjustment sequence of the control area based on the demand satisfaction, and determine the environmental optimization combination equipment among the environmental adjustment equipment in combination with the environmental adjustment sequence, the environmental adjustment efficiency and the collaborative compatibility; The environmental optimization module is used to perform environmental optimization processing on the controlled area based on the environmental optimization combination device, and monitor the energy consumption of the equipment corresponding to the environmental optimization combination device, so as to perform energy scheduling processing on the environmental optimization combination device until the demand satisfaction is higher than the preset threshold, stop the environmental optimization processing, and obtain the optimization result.

Images