Intelligent linkage control management system based on air conditioner application

Abstract

The invention discloses an intelligent linkage control management system based on air conditioner application, and relates to the technical field of intelligent control and energy management, and the system comprises a load prediction module which obtains an air conditioner load demand prediction value through an intelligent load prediction model based on preprocessed environment energy consumption data; the optimization scheduling module is used for optimizing the current air conditioner operation mode through the building energy management system based on the air conditioner load demand prediction value, and obtaining an optimized air conditioner operation scheduling instruction; the intelligent adjusting module is used for intelligently adjusting the operation parameters of the air conditioner system based on the optimized air conditioner operation scheduling instruction, adaptively adjusting the energy efficiency of the air conditioner and obtaining operation feedback data; and the fault management module is used for identifying potential fault risks of the air conditioner and formulating fault risk prevention measures through a time sequence analysis method based on the operation feedback data. According to the invention, through innovation of an intelligent load prediction and fault prevention mechanism, the energy utilization efficiency and the system stability are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control and energy management, and in particular to an intelligent linkage control and management system based on air conditioning application. Background Art

[0002] With the increasing attention paid by society to energy efficiency and environmental impact, air conditioning systems, as an important part of building energy consumption, have become a hot topic in energy-saving technology research. In recent years, intelligent management of central air conditioning systems has gradually developed into a trend. The control method of traditional air conditioning systems mainly relies on manual settings or automatic adjustment based on fixed parameters, which cannot respond to environmental changes and user needs in real time. In order to solve these problems, modern air conditioning systems have gradually introduced technologies such as big data analysis and artificial intelligence, and use environmental sensors and data analysis models for intelligent adjustment. In particular, load prediction based on deep learning and machine learning models such as long short-term memory networks (LSTM) and support vector machine regression (SVM) can effectively improve the adjustment accuracy and response speed of air conditioning systems, optimize energy consumption and improve the operating efficiency of the system. In addition, fault prediction and prevention are also important areas in air conditioning system management. Through real-time analysis of operation feedback data, potential fault risks can be identified, and maintenance and repair can be carried out in a timely manner, thereby extending equipment life and reducing energy waste.

[0003] However, the prior art still has certain shortcomings. First, although the existing load forecasting model can predict the load demand to a certain extent based on environmental data, the prediction accuracy and stability of the traditional model are still difficult to meet the needs of practical applications under highly complex environmental changes. Especially in the operation of the air-conditioning system, due to the complexity of environmental changes and personnel activity patterns, the traditional model fails to fully consider the real-time and dynamic nature of these variables, making it difficult to accurately predict future load demands. Secondly, the existing air-conditioning fault warning system usually relies on a single monitoring method and static rule setting, lacks in-depth analysis of operating data and adaptive optimization mechanism, and cannot identify hidden dangers in time and adjust the prediction model according to real-time feedback. Based on these shortcomings, the present invention proposes an intelligent linkage control and management system based on central air-conditioning applications, combining a variety of advanced technologies, aiming to achieve more accurate and dynamic load forecasting and fault prevention. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent linkage control management system based on air conditioning application to solve the problems of insufficient load prediction accuracy and lack of adaptive optimization of fault warning system.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: The present invention provides an intelligent linkage control and management system based on air-conditioning application, which includes: a data acquisition and preprocessing module, which acquires environmental energy consumption data and preprocesses the environmental energy consumption data; a model building module, which builds an intelligent load prediction model based on a long short-term memory network model and a support vector machine regression model; a load prediction module, which obtains an air-conditioning load demand prediction value through the intelligent load prediction model based on the preprocessed environmental energy consumption data; an optimization scheduling module, which optimizes the current air-conditioning operation mode through a building energy management system based on the air-conditioning load demand prediction value, and obtains an optimized air-conditioning operation scheduling instruction; an intelligent adjustment module, which intelligently adjusts the operation parameters of the air-conditioning system based on the optimized air-conditioning operation scheduling instruction, and adaptively adjusts the air-conditioning energy efficiency to obtain operation feedback data; a fault management module, which identifies potential fault risks of the air-conditioning through a time series analysis method based on the operation feedback data, and formulates fault risk prevention measures.

[0007] As a preferred solution of the intelligent linkage control and management system based on air conditioning application described in the present invention, wherein: the environmental energy consumption data includes temperature data, humidity data, CO2 concentration data, air quality data and air conditioning energy consumption data; The preprocessing of the environmental energy consumption data includes data cleaning, data standardization and data normalization.

[0008] As a preferred solution of the intelligent linkage control and management system based on air conditioning application described in the present invention, wherein: the intelligent load prediction model is constructed based on the long short-term memory network model and the support vector machine regression model, and the specific steps are as follows: Arrange the preprocessed historical environmental energy consumption data in chronological order, construct a windowed data set with timestamps, and use the historical environmental energy consumption data of each time step as input samples to form time series data; The model is based on the long short-term memory network model; The input layer receives time series data; The LSTM unit layer uses the input gate to determine which new information is stored in the unit state, the forget gate to determine which information should be forgotten from the unit state, and the output gate to determine which information is extracted from the unit state as the output of the current time step; LSTM output layer, outputs preliminary prediction values; The SVM regression layer minimizes the prediction error through the loss function and uses the kernel function to find the best regression hyperplane, handles the nonlinear relationship of the data, and optimizes the prediction value of the LSTM output; Output layer, outputs the optimized prediction results; Construct a preliminary intelligent load forecasting model; The preliminary intelligent load forecasting model is trained and evaluated through k-fold cross validation, and finally the intelligent load forecasting model is obtained.

[0009] As a preferred solution of the intelligent linkage control management system based on air conditioning application described in the present invention, wherein: a preliminary intelligent load forecasting model is trained and evaluated by k-fold cross validation, and finally an intelligent load forecasting model is obtained. The specific steps are as follows: The preprocessed historical environmental energy consumption data is divided into k subsets through k-fold cross validation; In each iteration, a subset is randomly selected as the validation set, and the rest is used as the training set; Use the training set to train the LSTM model and adjust the parameters of the LSTM through the back-propagation algorithm; Take the output of LSTM as input and train the SVM regression model; Use the validation set to evaluate the trained preliminary intelligent load forecasting model and obtain the error between the predicted value and the actual value; Repeat the above steps k times, adjust the hyperparameters of the preliminary intelligent load forecasting model in each iteration, and calculate the average error of cross-validation. Select the optimal hyperparameters by comparing the average errors under different hyperparameter settings; Based on the optimal hyperparameters, the preliminary intelligent load forecasting model is finally trained using the preprocessed historical environmental energy consumption dataset to obtain the final intelligent load forecasting model.

[0010] As a preferred solution of the intelligent linkage control and management system based on air conditioning application described in the present invention, wherein: based on the pre-processed environmental energy consumption data, the air conditioning load demand forecast value is obtained through the intelligent load forecasting model, and the specific steps are as follows: Based on the pre-processed environmental energy consumption data, the long short-term memory network model in the intelligent load forecasting model partially processes the time series data, and the support vector machine regression model partially optimizes the nonlinear characteristics to calculate the air conditioning load demand forecast value, which is expressed as: ; in, is at the time step The predicted value of air conditioning load demand is is the weighted fusion coefficient of LSTM and SVM output, is the activation function of LSTM, is the weight matrix of the LSTM network, is the LSTM unit at time step The hidden state of is the bias term of LSTM, It is the first Lagrange multipliers of the support vectors, yes and The similarity value of is at the time step Input pre-processed environmental energy consumption data, It is the first support vectors, is the bias term of the SVM regression model, is the index variable of the support vector, is the number of support vectors, is the index variable of the time step.

[0011] As a preferred solution of the intelligent linkage control management system based on air conditioning application described in the present invention, wherein: based on the predicted value of air conditioning load demand, the current air conditioning operation mode is optimized through the building energy management system to obtain the optimized air conditioning operation scheduling instruction, the specific steps are as follows: Based on the building energy management system, real-time monitoring of building energy efficiency data; According to the predicted value of air conditioning load demand and combined with building energy efficiency data, the current air conditioning operation strategy is adjusted through the building energy management system to obtain the adjusted operation strategy; Based on the adjusted operation strategy, the building energy management system optimizes the current air-conditioning operation mode and obtains the optimized air-conditioning operation mode; The building energy management system generates optimized air conditioning operation scheduling instructions based on the optimized air conditioning operation mode.

[0012] As a preferred solution of the intelligent linkage control and management system based on air conditioning application described in the present invention, wherein: based on optimizing the air conditioning operation scheduling instruction, the air conditioning system operation parameters are intelligently adjusted, and the specific steps are as follows: The control unit based on the air conditioning system collects real-time environmental data through environmental sensors; Based on the optimized air conditioning operation scheduling instructions and combined with real-time environmental data, the air conditioning system operating parameters are adjusted. The expression is: ; in, is at the time step The adjusted air conditioning system operating parameters, is at the time step Air conditioning system default operating parameters, is at the time step Optimized air conditioning operation scheduling instructions, is at the time step Real-time environmental data, is at the time step Real-time environmental data thresholds, It is the adjustment coefficient of the air conditioning system operating parameters to control the changes in environmental data.

[0013] As a preferred solution of the intelligent linkage control and management system based on air conditioning application described in the present invention, wherein: adaptively adjusting the air conditioning energy efficiency and obtaining operation feedback data, the specific steps are as follows: Based on the adjusted air conditioning system operating parameters, the default air conditioning energy efficiency is obtained through a linear conversion relationship; Based on the control unit of the air conditioning system, the high-definition camera collects personnel activity data in real time to obtain personnel activity patterns; The default air conditioning energy efficiency is adaptively adjusted based on the personnel activity pattern, and the expression is: ; in, is the energy efficiency of the air conditioner after adaptive adjustment, is the target air conditioning energy efficiency, is the adjustment coefficient of the air conditioning energy efficiency adjustment by controlling the activity mode of personnel. It is the personnel activity mode. is the personnel activity mode threshold, is at the time step The default air conditioning energy efficiency, is the adjustment coefficient of the personnel control activity mode on energy efficiency adjustment; The air conditioning system starts to operate based on the adjusted operating parameters and the adaptively adjusted air conditioning energy efficiency, and generates operation feedback data.

[0014] As a preferred solution of the intelligent linkage control and management system based on air conditioning application described in the present invention, wherein: based on the operation feedback data, through the time series analysis method, the specific steps are as follows: Based on the operation feedback data, the change amount of the operation feedback data is obtained by comparing it with the preset normal working value; Through the time series analysis method, the fault prediction value is calculated based on the change in operation feedback data. The expression is: ; in, is at the time step No. Fault prediction value of operation feedback data, is the total weight factor for fault prediction, is at the time step Time The change of running feedback data, It is The normal value of the running feedback data, It is The nonlinear adjustment coefficient of the running feedback data, is the time decay factor, is the weight factor of the integral term, It's a historical moment Operation feedback data, is the maximum value of the running feedback data, It's a historical moment The tiny change in the passage of time, is the index variable of the historical moment, is the index variable of the run feedback data.

[0015] As a preferred solution of the intelligent linkage control and management system based on air conditioning application described in the present invention, the specific steps of identifying the potential failure risk of air conditioning and formulating failure risk prevention measures are as follows: Analyze historical operation data and historical fault records through data mining technology to define fault risk thresholds ; Identify potential air conditioner failure risks based on failure prediction values ​​and failure risk thresholds; when , indicating that the risk of potential failure of the air conditioner is low; when , indicating that the potential failure risk of the air conditioner is in a critical state; when , indicating that the air conditioner has a high risk of potential failure; Based on the potential failure risk identification results of the air conditioner, formulate specific failure risk prevention measures.

[0016] The beneficial effects of the present invention are as follows: through the innovation of intelligent load prediction and fault prevention mechanism, the energy efficiency and operation stability of the air-conditioning system are significantly improved. Through accurate load prediction based on long short-term memory network (LSTM) and support vector machine regression (SVM) model, the air-conditioning system can dynamically optimize the operation mode, reduce energy waste, and improve the response speed and comfort of the system. At the same time, the fault analysis and prevention module identifies potential fault risks in advance through time series analysis and real-time feedback data monitoring, and formulates targeted preventive measures, thereby extending the service life of the equipment, reducing maintenance costs, and ensuring the high efficiency and reliability of the air-conditioning system in long-term operation. The present invention realizes the intelligent management of the air-conditioning system and effectively improves the energy utilization efficiency and system stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 This is a schematic diagram of an intelligent linkage control and management system based on air conditioning application in Example 1.

[0019] Figure 2 This is a flow chart for obtaining optimized air conditioning operation scheduling instructions in Example 1. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and which provides an intelligent linkage control and management system based on air conditioning application, comprising the following steps: The data collection and preprocessing module collects environmental energy consumption data and preprocesses the environmental energy consumption data.

[0024] Environmental energy consumption data includes temperature data, humidity data, CO2 concentration data, air quality data and air conditioning energy consumption data.

[0025] Preprocessing of environmental energy consumption data includes data cleaning, data standardization and data normalization.

[0026] Data cleaning specifically involves identifying and deleting unreasonable values ​​(such as temperature, humidity or CO2 concentration data that are significantly out of the normal range) through algorithms, and then filling in missing values ​​through interpolation or other methods to ensure the integrity and consistency of environmental energy consumption data and the accuracy of subsequent analysis and modeling; Data standardization is to subtract the mean of environmental energy consumption data and divide it by the standard deviation, so that the mean of each feature is 0 and the variance is 1; standardization can eliminate the dimensional differences between different data sources, so that each environmental energy consumption data feature can be compared and analyzed on the same scale; Data normalization specifically involves scaling the environmental energy consumption data to a specified interval (usually between 0 and 1), and mapping the minimum value of each feature to 0 and the maximum value to 1 through linear transformation.

[0027] The model building module builds an intelligent load forecasting model based on the long short-term memory network model and the support vector machine regression model.

[0028] The preprocessed historical environmental energy consumption data are arranged in chronological order to construct a windowed dataset with timestamps. The historical environmental energy consumption data at each time step is used as an input sample to form time series data.

[0029] Specifically, the preprocessed historical environmental energy consumption data is arranged in chronological order, and the preprocessed historical environmental energy consumption data is divided into multiple time periods according to the set time window size. The preprocessed historical environmental energy consumption data in each time period is used as an input sample. Each sample contains the preprocessed historical environmental energy consumption data of one time step. These preprocessed historical environmental energy consumption data include environmental parameters such as temperature, humidity, and CO2 concentration, and each sample has a timestamp, indicating the specific position of the preprocessed historical environmental energy consumption data point in the time series. In this way, the generated time series data set can be used to capture the trend and law of data changes over time, providing a basis for the subsequent intelligent load forecasting model.

[0030] The model is based on the long short-term memory network model; The input layer receives time series data; The LSTM unit layer uses the input gate to determine which new information is stored in the unit state, the forget gate to determine which information should be forgotten from the unit state, and the output gate to determine which information is extracted from the unit state as the output of the current time step; LSTM output layer, outputs preliminary prediction values; The SVM regression layer minimizes the prediction error through the loss function and uses the kernel function to find the best regression hyperplane, handles the nonlinear relationship of the data, and optimizes the prediction value of the LSTM output; Output layer, outputs the optimized prediction results; A preliminary intelligent load forecasting model was constructed.

[0031] It should be noted that the combination of the long short-term memory network (LSTM) model and the support vector machine regression (SVM) model can give full play to the advantages of both. The LSTM model is good at capturing long-term dependencies in time series data and can effectively handle the time dynamic changes of air conditioning load, while the SVM regression model can handle the nonlinear relationship of data and improve the prediction accuracy by optimizing the regression hyperplane. After the two are combined, the time series features provided by LSTM can be further optimized by SVM regression, thereby improving the accuracy and stability of load forecasting, especially when the environmental energy consumption data changes complexly.

[0032] The preliminary intelligent load forecasting model is trained and evaluated through k-fold cross validation, and finally the intelligent load forecasting model is obtained.

[0033] The preliminary intelligent load forecasting model is trained and evaluated through k-fold cross validation, and finally the intelligent load forecasting model is obtained.

[0034] The preprocessed historical environmental energy consumption data are divided into k subsets through k-fold cross validation.

[0035] Specifically, the entire preprocessed historical environmental energy consumption dataset is randomly divided into k subsets of equal size, each of which contains a different part of the preprocessed historical environmental energy consumption dataset. Ensure that the sample distribution of each subset is as consistent as possible with the overall preprocessed historical environmental energy consumption dataset, thereby improving the generalization ability of the preliminary intelligent load forecasting model.

[0036] In each iteration, a subset is randomly selected as the validation set and the rest is used as the training set.

[0037] The LSTM model is trained using the training set, and the parameters of the LSTM are adjusted through the back-propagation algorithm.

[0038] The output of LSTM is used as input to train the SVM regression model.

[0039] The trained preliminary intelligent load forecasting model is evaluated using the validation set to obtain the error between the predicted value and the actual value.

[0040] Repeat the above steps k times, adjust the hyperparameters of the preliminary intelligent load forecasting model in each iteration, calculate the average error of cross-validation, and select the optimal hyperparameters by comparing the average errors under different hyperparameter settings.

[0041] Specifically, first, for each set of hyperparameter configurations, training and validation are performed on k subsets, and the error of each validation set is calculated (usually the difference between the predicted value and the actual value, such as the mean square error). Then, the errors on these k validation sets are averaged to obtain the average error of the hyperparameter configuration. Next, the average errors under different hyperparameter configurations are compared, and the hyperparameter combination that minimizes the average error is selected as the optimal hyperparameter.

[0042] Based on the optimal hyperparameters, the preliminary intelligent load forecasting model is finally trained using the preprocessed historical environmental energy consumption dataset to obtain the final intelligent load forecasting model.

[0043] The load forecasting module obtains the predicted value of air conditioning load demand based on the pre-processed environmental energy consumption data through the intelligent load forecasting model.

[0044] Based on the pre-processed environmental energy consumption data, the long short-term memory network model in the intelligent load forecasting model partially processes the time series data, and the support vector machine regression model partially optimizes the nonlinear characteristics to calculate the air conditioning load demand forecast value, which is expressed as: ; in, is at the time step The predicted value of air conditioning load demand is is the weighted fusion coefficient of LSTM and SVM output, is the activation function of LSTM, is the weight matrix of the LSTM network, is the LSTM unit at time step The hidden state of is the bias term of LSTM, It is the first Lagrange multipliers of the support vectors, yes and The similarity value of is at the time step Input pre-processed environmental energy consumption data, It is the first support vectors, is the bias term of the SVM regression model, is the index variable of the support vector, is the number of support vectors, is the index variable of the time step.

[0045] It should be noted that the similarity value is calculated The expression is: ; in, are the parameters of the kernel function.

[0046] The optimization scheduling module optimizes the current air-conditioning operation mode based on the predicted value of air-conditioning load demand through the building energy management system and obtains the optimized air-conditioning operation scheduling instructions.

[0047] Based on the building energy management system, real-time monitoring of building energy efficiency data.

[0048] It should be noted that building energy efficiency data includes external climate data, electricity prices, and activity data within the building.

[0049] According to the predicted value of air-conditioning load demand and combined with building energy efficiency data, the current air-conditioning operation strategy is adjusted through the building energy management system to obtain the adjusted operation strategy.

[0050] It should be noted that the current air conditioning operation strategy is the initial operation strategy generated by BEMS based on historical data, preset operating parameters and current environmental conditions (such as temperature, humidity, etc.), as well as the user's comfort requirements; It should also be noted that the building energy management system (BEMS) first analyzes the relationship between the current air-conditioning load demand and the internal and external environment of the building, evaluates the effectiveness of the current air-conditioning operation strategy, and combines the real-time collected building energy efficiency data (such as temperature, humidity, electricity prices, personnel activities, etc.). The building energy management system (BEMS) compares the current energy efficiency and predicted load demand to identify whether there is excessive or insufficient air-conditioning load. Then, through optimization algorithms (such as linear programming, dynamic scheduling algorithms, etc.), the building energy management system (BEMS) adjusts the operating parameters of the air conditioner, including the operating time, set temperature, wind speed and other parameters of the air conditioner, to ensure that the air-conditioning system can save energy to the maximum extent while meeting the load demand, and finally generates an adjusted air-conditioning operation strategy.

[0051] Based on the adjusted operation strategy, the building energy management system optimizes the current air-conditioning operation mode to obtain the optimized air-conditioning operation mode.

[0052] Specifically, the building energy management system (BEMS) first conducts a comprehensive analysis of factors such as the air-conditioning load demand forecast, building energy efficiency data, external climate data and electricity prices. Then, the building energy management system (BEMS) evaluates the gap between the current air-conditioning operation mode and load demand through an optimization algorithm, and adjusts the parameters of the air-conditioning system, such as temperature setting, wind speed, operating time and working mode, based on the activity data in the building. During the optimization process, the system will gradually calculate the optimal combination of various parameters to minimize energy consumption while ensuring comfort. Finally, the building energy management system (BEMS) generates a new air-conditioning operation mode based on these adjusted parameters.

[0053] It should be noted that the building energy management system (BEMS) first conducts a comprehensive analysis of factors such as air conditioning load demand forecast, building energy efficiency data, external climate data and electricity prices. The specific contents are as follows: The building energy management system (BEMS) integrates information such as air conditioning load demand forecast, building energy efficiency data, external climate data and electricity prices, and uses a multi-dimensional data analysis method to first calculate the impact of each factor on the operation of the air conditioning system. For example, the energy demand of air conditioning is evaluated through the air conditioning load demand forecast, the energy efficiency level of the existing system is analyzed using building energy efficiency data, and the impact of weather changes on air conditioning load is predicted in combination with external climate data, and then the timing of energy use is optimized according to electricity price fluctuations.

[0054] The building energy management system generates optimized air conditioning operation scheduling instructions based on the optimized air conditioning operation mode.

[0055] Specifically, the building energy management system (BEMS) determines the optimal operating parameters of the air conditioner, such as temperature setting, wind speed, and operating time, based on the optimized air conditioner operating mode; then the building energy management system (BEMS) evaluates the air conditioner efficiency and energy consumption under different load demands through simulation and calculation, and adjusts the air conditioner operating mode in combination with real-time external environmental changes. Finally, the building energy management system (BEMS) generates specific air conditioner operation scheduling instructions based on these optimization results, instructing the air conditioner system on how to adjust its operation in various time periods to achieve the lowest energy consumption.

[0056] The intelligent adjustment module intelligently adjusts the operating parameters of the air conditioning system based on the optimized air conditioning operation scheduling instructions.

[0057] Based on the optimized air conditioning operation scheduling instructions, the air conditioning system operating parameters are intelligently adjusted.

[0058] The control unit based on the air conditioning system collects real-time environmental data through environmental sensors.

[0059] Based on the optimized air conditioning operation scheduling instructions and combined with real-time environmental data, the air conditioning system operating parameters are adjusted. The expression is: ; in, is at the time step The operating parameters of the air conditioning system after adjustment, is at the time step Air conditioning system default operating parameters, is at the time step Optimized air conditioning operation scheduling instructions, is at the time step Real-time environmental data, is at the time step Real-time environmental data thresholds, It is the adjustment coefficient of the air conditioning system operating parameters to control the changes in environmental data.

[0060] It should be noted that the real-time environmental data threshold is defined through the analysis of historical environmental data; Specifically, the building energy management system (BEMS) analyzes environmental data over a period of time to identify the fluctuation range and trend, and then determines the appropriate threshold range based on the statistical characteristics of these historical environmental data (such as mean, standard deviation, maximum and minimum values, etc.). These thresholds can reflect the normal environmental fluctuation range, and then set an appropriate threshold by eliminating and adjusting abnormal data to ensure that changes in real-time environmental data do not exceed the predetermined range.

[0061] Intelligent adjustment module, adaptively adjusts air conditioning energy efficiency and obtains operation feedback data.

[0062] Based on the adjusted air-conditioning system operating parameters, the default air-conditioning energy efficiency is obtained through a linear conversion relationship.

[0063] The control unit based on the air-conditioning system collects personnel activity data in real time through high-definition cameras to obtain personnel activity patterns.

[0064] It should be noted that high-definition cameras monitor indoor areas in real time and capture the movement and distribution of people; the control unit of the air-conditioning system extracts and parses personnel activity information from the video stream through image processing and analysis algorithms, identifies the existence, number, location and activity mode (such as stillness, walking, etc.) of people, and thus derives the activity pattern of people.

[0065] The default air conditioning energy efficiency is adaptively adjusted based on the personnel activity pattern, and the expression is: ; in, is the energy efficiency of the air conditioner after adaptive adjustment, is the target air conditioning energy efficiency, is the adjustment coefficient of the air conditioning energy efficiency adjustment by controlling the activity mode of personnel. It is the personnel activity mode. is the personnel activity mode threshold, is at the time step The default air conditioning energy efficiency, It is the adjustment coefficient of energy efficiency adjustment by personnel control activity mode.

[0066] It should be noted that the target air conditioning energy efficiency It is usually determined based on a comprehensive consideration of multiple factors, including but not limited to the building's energy management policy, user comfort requirements, historical energy consumption data, current environmental conditions (such as temperature, humidity, etc.), and possible external energy price signals or energy-saving indicators. These factors are analyzed through optimization algorithms (such as linear programming, genetic algorithms) to calculate an air-conditioning operating state that achieves optimal energy efficiency while satisfying user comfort. The energy efficiency level corresponding to this ideal operating state is determined as the target air-conditioning energy efficiency; It should also be noted that the personnel activity pattern threshold is defined through the analysis of historical personnel activity patterns; Specifically, the building energy management system (BEMS) collects and analyzes personnel activity data over a period of time, identifies the relationship between different activity modes (such as standing, sitting, walking, etc.) and air-conditioning energy efficiency, and then determines the typical characteristics and variation range of various activity modes based on statistical analysis methods (such as cluster analysis, time series analysis, etc.); the building energy management system (BEMS) classifies different activity modes, and calculates the frequency of occurrence of various activities in different time periods. Combined with the relationship between activity modes and time, it identifies the degree of influence of each activity mode on air-conditioning energy efficiency in a specific period of time, and then sets reasonable thresholds for each activity mode.

[0067] The air conditioning system starts to operate based on the adjusted operating parameters and the adaptively adjusted air conditioning energy efficiency, and generates operation feedback data.

[0068] Specifically, first, the air-conditioning system starts to operate according to the optimized operating parameters (such as temperature setting, wind speed, etc.) and the adjusted energy efficiency values, and then continuously monitors and collects real-time environmental data, such as indoor temperature, humidity and CO2 concentration, to ensure that the air-conditioning maintains comfort and energy efficiency according to the set goals. Then the air-conditioning system fine-tunes the operating status according to the real-time data, and collects feedback information through sensors and control units to generate operating feedback data.

[0069] The fault management module is based on the operation feedback data through time series analysis method.

[0070] Based on operation feedback data, through time series analysis method.

[0071] Based on the operation feedback data, the change in the operation feedback data is obtained by comparing it with the preset normal working value.

[0072] It should be noted that the real-time operation feedback data of the air conditioner at different time steps are collected, including parameters such as temperature, humidity, and energy efficiency. Then, according to the air-conditioning system design standards, historical operation data, and current environmental conditions, a preset normal working value is set as a reference. These normal working values ​​are the parameter values ​​that the air conditioner should achieve under ideal working conditions. Then, the real-time collected feedback data is compared with these preset normal working values, and the change of each feedback data (that is, the deviation between the feedback data and the normal working value) is calculated.

[0073] Through the time series analysis method, the fault prediction value is calculated based on the change in operation feedback data. The expression is: ; in, is at the time step No. Fault prediction value of operation feedback data, is the total weight factor for fault prediction, is at the time step Time The change of running feedback data, It is The normal value of the running feedback data, It is The nonlinear adjustment coefficient of the running feedback data, is the time decay factor, is the weight factor of the integral term, It's a historical moment Operation feedback data, is the maximum value of the running feedback data, It's a historical moment The tiny change in the passage of time, is the index variable of the historical moment, is the index variable of the run feedback data.

[0074] The fault management module identifies potential air conditioner failure risks and formulates failure risk prevention measures.

[0075] Analyze historical operation data and historical fault records through data mining technology to define fault risk thresholds .

[0076] Furthermore, the historical operation data and historical fault records of the air conditioner are collected, including information such as the performance of the air conditioner in different time periods, the occurrence of faults, and the operating environment. Then, data mining techniques, such as cluster analysis, regression analysis, or anomaly detection methods, are used to analyze the failure modes and performance fluctuations in the historical operation data and historical fault records. By identifying and extracting key factors that are highly correlated with the occurrence of failures, such as temperature fluctuations, energy efficiency changes, and equipment aging, these factors are used as the main features for judging the failure risk. Then, based on these features, a failure risk threshold is defined as the standard for judging the failure risk of the air conditioner.

[0077] Identify potential air conditioner failure risks based on failure prediction values ​​and failure risk thresholds; when , indicating that the risk of potential failure of the air conditioner is low; when , indicating that the potential failure risk of the air conditioner is in a critical state; when , indicating that the air conditioner has a high risk of potential failure.

[0078] Based on the potential failure risk identification results of the air conditioner, formulate specific failure risk prevention measures.

[0079] It should be noted that potential air conditioner failure risks include compressor failure risk, circuit failure risk, and condenser / evaporator failure risk; Based on the above potential air conditioner failure risks, failure risk prevention measures are formulated respectively. Specifically, compressor failure risk prevention measures: regularly check the current and temperature of the compressor to ensure sufficient refrigerant, and install protection devices and intelligent monitoring systems to detect failures in advance; Circuit failure risk prevention measures: Check circuit connections, use current protectors and overvoltage protection devices, ensure stable power supply, and use intelligent power monitoring systems to detect current and voltage fluctuations in real time; Condenser / evaporator failure risk prevention measures: Clean the condenser and evaporator regularly, check the condensate discharge system, and ensure smooth heat exchange and refrigerant circulation.

[0080] In summary, the present invention significantly improves the energy efficiency and operational stability of the air-conditioning system through the innovation of intelligent load prediction and fault prevention mechanism. Through accurate load prediction based on long short-term memory network (LSTM) and support vector machine regression (SVM) model, the air-conditioning system can dynamically optimize the operation mode, reduce energy waste, and improve the response speed and comfort of the system. At the same time, the fault analysis and prevention module identifies potential fault risks in advance through time series analysis and real-time feedback data monitoring, and formulates targeted preventive measures, thereby extending the service life of the equipment, reducing maintenance costs, and ensuring the efficiency and reliability of the air-conditioning system in long-term operation. The present invention realizes the intelligent management of the air-conditioning system and effectively improves the energy utilization efficiency and system stability.

[0081] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of an intelligent linkage control and management system based on air conditioning application is provided.

[0082] In order to verify the innovation and advantages of the intelligent air conditioning management system in the present invention in optimizing air conditioning operation, improving energy efficiency and identifying fault risks, an experimental scenario was designed to test the performance of the system in an actual environment. The experiment took an office building as the experimental object, selected the air conditioning system and environmental energy consumption data of the building for real-time collection, and compared them with the traditional air conditioning system.

[0083] First, in the experimental preparation stage, historical environmental energy consumption data, including temperature, humidity, CO2 concentration, air quality, and air conditioning energy consumption data, are collected and preprocessed. Data preprocessing includes cleaning unreasonable historical environmental energy consumption data, filling missing values, standardization, and normalization to ensure the quality and consistency of historical environmental energy consumption data, thereby providing an accurate basis for subsequent modeling and prediction. Then, the long short-term memory network (LSTM) and support vector machine regression (SVM) models are used to combine the preprocessed historical environmental energy consumption data to build an intelligent load forecasting model. The historical environmental energy consumption data is processed through time windowing. LSTM is responsible for capturing the long-term dependency of the time series, while SVM optimizes the nonlinear relationship of the historical environmental energy consumption data, and finally forms a preliminary intelligent load forecasting model.

[0084] Secondly, the experiment trained and evaluated the preliminary intelligent load forecasting model through the k-fold cross-validation method. The historical environmental energy consumption dataset was divided into k subsets, one subset was used as the validation set each time, and the remaining subsets were used as the training set for iterative training to ensure the stability and generalization ability of the preliminary intelligent load forecasting model. The hyperparameters of the preliminary intelligent load forecasting model were repeatedly adjusted, and the optimal configuration was selected based on the average error of cross-validation. Finally, based on the optimal hyperparameters, the entire historical environmental energy consumption dataset was used for final training to obtain a high-precision intelligent load forecasting model.

[0085] Next, based on the trained intelligent load forecasting model, the air conditioning load demand forecast value is obtained by analyzing the pre-processed environmental data, and the air conditioning operation mode is optimized by combining building energy efficiency data, external climate information, electricity prices and other factors. The building energy management system (BEMS) adjusts the air conditioning operating parameters, such as temperature setting and wind speed, according to the predicted load demand and optimization algorithm to ensure that comfort requirements are met while maximizing energy savings, and generates optimized air conditioning operation scheduling instructions.

[0086] Finally, after obtaining the optimized air conditioning operation scheduling instructions, the building energy management system intelligently adjusts the operating parameters of the air conditioning system and adaptively adjusts the air conditioning energy efficiency. The system further optimizes the operation of the air conditioning system through real-time environmental data and personnel activity patterns to ensure a balance between energy efficiency and comfort. In addition, by collecting operation feedback data and performing time series analysis, the system can predict potential failures and take preventive measures, thereby improving the stability and operating efficiency of the air conditioning system.

[0087] The details are shown in Table 1 below: Table 1 Air conditioning system performance comparison test table Through the analysis of the data in the above table, it can be clearly seen that the intelligent air conditioning management system is superior to the traditional air conditioning system in many aspects. First, the intelligent system performs more accurately in environmental control. When the collection time is 12:00, the temperature is optimized from 23.1°C of the traditional air conditioning system to 22.3°C, the humidity is reduced from 47.5% to 45.8%, and the CO2 concentration is also reduced from 590 ppm to 570ppm, which effectively improves the indoor comfort and air quality. In terms of energy efficiency, the air conditioning energy consumption of the intelligent air conditioning system is 14.6kWh, which is significantly lower than the 17.8 kWh of the traditional air conditioning, and the energy saving effect reaches about 17%. At the same time, the running time is also shorter (8.4 hours vs. 9.5 hours). The adjusted energy efficiency of the intelligent air conditioning system is 2.2 kWh / ton, which is better than the 2.7 kWh / ton of the traditional system, further optimizing energy use. More importantly, the intelligent system also performs well in fault risk identification. Its fault risk identification time is 40 hours, which is 50 hours ahead of the traditional system. Early warning helps to avoid potential faults. Finally, the feedback data error of the intelligent air-conditioning system is 0.3 kWh, which is significantly lower than the 0.6 kWh of the traditional system, improving the accuracy and reliability of the system.

[0088] The present invention significantly improves the energy efficiency and comfort of air conditioning operation through the application of intelligent air conditioning management system. By real-time collection and analysis of environmental data, such as temperature, humidity and CO2 concentration, the operating parameters of the air conditioner are accurately adjusted to optimize energy consumption and reduce air conditioning energy consumption by about 17%. In addition, the air conditioner operation can be adaptively adjusted to reduce energy efficiency errors and provide higher comfort and air quality. Through advanced fault risk prediction and identification mechanisms, the system can warn of potential faults in advance, reducing downtime and maintenance costs.

[0089] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. 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, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent linkage control and management system based on air conditioning application, characterized in that: include, Data collection and preprocessing module, collects environmental energy consumption data and preprocesses the environmental energy consumption data; Model building module, which builds intelligent load forecasting model based on long short-term memory network model and support vector machine regression model; The load forecasting module obtains the predicted value of air conditioning load demand based on the pre-processed environmental energy consumption data through the intelligent load forecasting model; The optimization scheduling module optimizes the current air-conditioning operation mode through the building energy management system based on the predicted value of air-conditioning load demand and obtains the optimized air-conditioning operation scheduling instructions; Intelligent adjustment module, based on optimizing air conditioning operation scheduling instructions, intelligently adjusts air conditioning system operating parameters, adaptively adjusts air conditioning energy efficiency, and obtains operation feedback data; The fault management module, based on the operation feedback data, uses the time series analysis method to identify the potential failure risks of the air conditioner and formulate failure risk prevention measures.

2. The intelligent linkage control and management system based on air conditioning application according to claim 1, characterized in that: The environmental energy consumption data includes temperature data, humidity data, CO2 concentration data, air quality data and air conditioning energy consumption data; The preprocessing of the environmental energy consumption data includes data cleaning, data standardization and data normalization.

3. The intelligent linkage control and management system based on air conditioning application as claimed in claim 2, characterized in that: The intelligent load forecasting model is constructed based on the long short-term memory network model and the support vector machine regression model. The specific steps are as follows: Arrange the preprocessed historical environmental energy consumption data in chronological order, construct a windowed data set with timestamps, and use the historical environmental energy consumption data of each time step as input samples to form time series data; The model is based on the long short-term memory network model; The input layer receives time series data; The LSTM unit layer uses the input gate to determine which new information is stored in the unit state, the forget gate to determine which information should be forgotten from the unit state, and the output gate to determine which information is extracted from the unit state as the output of the current time step; LSTM output layer, outputs preliminary prediction values; The SVM regression layer minimizes the prediction error through the loss function and uses the kernel function to find the best regression hyperplane, handles the nonlinear relationship of the data, and optimizes the prediction value of the LSTM output; Output layer, outputs the optimized prediction results; Construct a preliminary intelligent load forecasting model; The preliminary intelligent load forecasting model is trained and evaluated through k-fold cross validation, and finally the intelligent load forecasting model is obtained.

4. The intelligent linkage control and management system based on air conditioning application as claimed in claim 3, characterized in that: The preliminary intelligent load forecasting model is trained and evaluated through k-fold cross validation, and finally the intelligent load forecasting model is obtained. The specific steps are as follows: The preprocessed historical environmental energy consumption data is divided into k subsets through k-fold cross validation; In each iteration, a subset is randomly selected as the validation set, and the rest is used as the training set; Use the training set to train the LSTM model and adjust the parameters of the LSTM through the back-propagation algorithm; Take the output of LSTM as input and train the SVM regression model; Use the validation set to evaluate the trained preliminary intelligent load forecasting model and obtain the error between the predicted value and the actual value; Repeat the above steps k times, adjust the hyperparameters of the preliminary intelligent load forecasting model in each iteration, and calculate the average error of cross-validation. Select the optimal hyperparameters by comparing the average errors under different hyperparameter settings; Based on the optimal hyperparameters, the preliminary intelligent load forecasting model is finally trained using the preprocessed historical environmental energy consumption dataset to obtain the final intelligent load forecasting model.

5. The intelligent linkage control and management system based on air conditioning application as claimed in claim 4, characterized in that: The air conditioning load demand forecast value is obtained based on the pre-processed environmental energy consumption data through the intelligent load forecasting model. The specific steps are as follows: Based on the pre-processed environmental energy consumption data, the long short-term memory network model in the intelligent load forecasting model partially processes the time series data, and the support vector machine regression model partially optimizes the nonlinear characteristics to calculate the air conditioning load demand forecast value, which is expressed as: ； in, is at the time step The predicted value of air conditioning load demand is is the weighted fusion coefficient of LSTM and SVM output, is the activation function of LSTM, is the weight matrix of the LSTM network, is the LSTM unit at time step The hidden state of is the bias term of LSTM, It is the first Lagrange multipliers of the support vectors, yes and The similarity value of is at the time step Input pre-processed environmental energy consumption data, is the first support vectors, is the bias term of the SVM regression model, is the index variable of the support vector, is the number of support vectors, is the index variable of the time step.

6. The intelligent linkage control and management system based on air conditioning application as claimed in claim 5, characterized in that: Based on the predicted value of air conditioning load demand, the current air conditioning operation mode is optimized through the building energy management system to obtain the optimized air conditioning operation scheduling instruction. The specific steps are as follows: Based on the building energy management system, real-time monitoring of building energy efficiency data; According to the predicted value of air conditioning load demand and combined with building energy efficiency data, the current air conditioning operation strategy is adjusted through the building energy management system to obtain the adjusted operation strategy; Based on the adjusted operation strategy, the building energy management system optimizes the current air-conditioning operation mode and obtains the optimized air-conditioning operation mode; The building energy management system generates optimized air conditioning operation scheduling instructions based on the optimized air conditioning operation mode.

7. The intelligent linkage control and management system based on air conditioning application according to claim 6, characterized in that: The specific steps of intelligently adjusting the operating parameters of the air conditioning system based on optimizing the air conditioning operation scheduling instructions are as follows: The control unit based on the air conditioning system collects real-time environmental data through environmental sensors; Based on the optimized air conditioning operation scheduling instructions and combined with real-time environmental data, the air conditioning system operating parameters are adjusted. The expression is: ； in, is at the time step The operating parameters of the air conditioning system after adjustment, is at the time step Air conditioning system default operating parameters, is at the time step Optimized air conditioning operation scheduling instructions, is at the time step Real-time environmental data, is at the time step Real-time environmental data thresholds, It is the adjustment coefficient of the air conditioning system operating parameters to control the changes in environmental data.

8. The intelligent linkage control and management system based on air conditioning application according to claim 7, characterized in that: Adaptively adjust the air conditioning energy efficiency and obtain operation feedback data. The specific steps are as follows: Based on the adjusted air conditioning system operating parameters, the default air conditioning energy efficiency is obtained through a linear conversion relationship; Based on the control unit of the air conditioning system, the high-definition camera collects personnel activity data in real time to obtain personnel activity patterns; The default air conditioning energy efficiency is adaptively adjusted based on the personnel activity pattern, and the expression is: ； in, is the energy efficiency of the air conditioner after adaptive adjustment, is the target air conditioning energy efficiency, is the adjustment coefficient of the air conditioning energy efficiency adjustment by controlling the activity mode of personnel. It is the personnel activity mode. is the personnel activity mode threshold, is at the time step The default air conditioning energy efficiency, is the adjustment coefficient of the personnel control activity mode on energy efficiency adjustment; The air conditioning system starts to operate based on the adjusted operating parameters and the adaptively adjusted air conditioning energy efficiency, and generates operation feedback data.

9. The intelligent linkage control and management system based on air conditioning application as claimed in claim 8, characterized in that: Based on the operation feedback data, the specific steps are as follows through the time series analysis method: Based on the operation feedback data, the change amount of the operation feedback data is obtained by comparing it with the preset normal working value; Through the time series analysis method, the fault prediction value is calculated based on the change in operation feedback data. The expression is: ； in, is at the time step No. Fault prediction value of operation feedback data, is the total weight factor for fault prediction, is at the time step Time The change of running feedback data, It is The normal value of the running feedback data, It is The nonlinear adjustment coefficient of the running feedback data, is the time decay factor, is the weight factor of the integral term, It's a historical moment Operation feedback data, is the maximum value of the running feedback data, It's a historical moment The tiny change in the passage of time, is the index variable of the historical moment, is the index variable of the run feedback data.

10. The intelligent linkage control and management system based on air conditioning application according to claim 9, characterized in that: The specific steps of identifying the potential failure risk of the air conditioner and formulating failure risk prevention measures are as follows: Analyze historical operation data and historical fault records through data mining technology to define fault risk thresholds ; Identify potential air conditioner failure risks based on failure prediction values ​​and failure risk thresholds; when , indicating that the risk of potential failure of the air conditioner is low; when , indicating that the potential failure risk of the air conditioner is in a critical state; when , indicating that the air conditioner has a high risk of potential failure; Based on the potential failure risk identification results of the air conditioner, formulate specific failure risk prevention measures.

Images