Air conditioner carbon reduction control method, device and equipment and storage medium

Abstract

The invention belongs to the technical field of air conditioner carbon reduction control, and discloses an air conditioner carbon reduction control method, device and equipment and a storage medium, and the air conditioner carbon reduction control method, device and equipment based on encoder-decoder can effectively analyze the relation between regions and accurately predict building hourly carbon emission. Outdoor and indoor hourly temperature, humidity, holiday, time, partition carbon emission and hourly air volume information of each partition serve as input characteristics, and historical information of the characteristics is used for predicting the carbon emission intensity of each partition in the next one hour. In the research, the encoder-decoder model can accurately capture the interrelation between the partitions, and the high-precision prediction of the carbon emission is realized. The method can be used as an auxiliary means to quickly position the abnormal carbon emission region or time period. Meanwhile, the prediction result can be compared with a historical period or a reference value, and the problem that small-area carbon emission abnormity is difficult to detect is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of air-conditioning carbon reduction control, and in particular, relates to an air-conditioning carbon reduction control method, device, equipment and storage medium. Background Art

[0002] The construction industry accounts for about one-third of global carbon emissions, and excessive carbon emissions will cause a series of ecological problems. In view of the low degree of digitization and extensive management in the traditional construction industry, the present invention divides the target floor into three areas: west, south, and east, and provides an encoder-decoder model structure that integrates LSTM, GRU, SimpleRNN, CNN, and attention algorithms for real-time prediction of carbon emissions from building operation and maintenance. At the same time, the hourly air volume of air-conditioning equipment is optimized and controlled online. Current research focuses on the carbon reduction optimization of the entire building, mainly starting from a single building or a single-story building. Carbon emission prediction has not yet been carried out from the perspective of building zoning and considering the interconnection between each zone. At the same time, the current air volume control of the air-conditioning terminal is subjective, with little relevant basis, and dynamic control has not been achieved.

[0003] Through the above analysis, the problems and defects of the prior art are as follows:

[0004] (1) Previous studies focused on the overall carbon emissions of buildings and were unable to predict the hourly carbon emissions of high-value, high-energy density areas in buildings. It was difficult to identify abnormal carbon emission periods / areas from the carbon emissions of the entire building, and the level of refinement in building carbon reduction research was not high.

[0005] (2) In traditional carbon emission time series prediction research, it is impossible to simultaneously take into account information such as the correlation between carbon emissions at different times and the spatial characteristics of carbon emissions in different regions, resulting in low carbon emission prediction accuracy and inability to be used to assist decision-making on actual building carbon reduction.

[0006] (3) Traditional air-conditioning terminal air volume control is subjective and cannot be performed according to actual needs, resulting in long-term local discomfort and waste of resources. Summary of the invention

[0007] In view of the problems existing in the prior art, the present invention provides an air-conditioning carbon reduction control method, device, equipment and storage medium.

[0008] The present invention is implemented in this way: an air conditioning carbon reduction control method comprises:

[0009] Step 1: Acquisition of building parameters;

[0010] Step 2: Build an encoder-decoder time series prediction model for time series prediction;

[0011] Step three: Online optimization of air outlet control at the air conditioning terminal to achieve energy conservation and carbon reduction in buildings.

[0012] Further, the method for obtaining building parameters is as follows:

[0013] Acquisition of hourly energy consumption data of air-conditioning terminals on target floors during the cooling season, outdoor hourly environmental parameters, indoor hourly temperature and humidity, time, floor holiday calendar, and hourly air volume information of air-conditioning terminals;

[0014] The collected data is preprocessed as follows:

[0015]

[0016] Formula (1), t l is the time to be interpolated, t l-1 and t l+1 The time t l The moment before and the moment after ; y l-1 ,y l and l+1 The corresponding time t l-1 , t l and t l+1 carbon emissions;

[0017] CE elec =E elec ×EF elec (2)

[0018] E elec and EF elec are electricity consumption and carbon emission factor (EF elec =0.5703), CE elec It is the carbon emission corresponding to the electricity consumption.

[0019] Furthermore, the encoder-decoder time series prediction model is constructed for time series prediction:

[0020] (1) The Encoder module consists of SimpleRNN, GRU and CNN;

[0021] (2) The RNN model structure is as follows:

[0022] h t =σ(wx t +Uh t-1 +b)(1)

[0023] In the formula, x t and h t Corresponding to the input variables and hidden states at time t, w and U are the matrices corresponding to the current unit and input variables, respectively, and b is the bias of the neuron;

[0024] (3) The GRU model structure is as follows:

[0025] z t =σ(x t w xr +h t-1 w hr +b r )(2)

[0026] r t =σ(x t w xz +h t-1 w hz +b z )(3)

[0027]

[0028] In the formula, σ and tanh are sigmoid function and activation function respectively, w xr and w hr is the update gate weight, w xz and w hz is to reset the gate weight, w xh and w hh is a candidate hidden layer state; x t is the input at time t, h t-1 is the hidden layer t-1 matrix, r t is the reset gate, z t is the update gate, b r , b z and b h is the bias term;

[0029] (4) Bi-GRU hybrid structure and attention mechanism;

[0030]

[0031] In the formula, and are the hidden layers of the forward GRU and the reverse GRU respectively, and F is the combination of the bidirectional GRU;

[0032] (5) The calculation method of the attention mechanism is as follows:

[0033] e t =utanh(wh t +b)(7)

[0034]

[0035] In the formula, e t is the probability distribution, u and w are weight matrices, b and st They are the bias term and attention mechanism output respectively; a t is the distribution coefficient.

[0036] (6) Coefficient of determination R 2 , root mean square error RMSE and mean absolute error MAE;

[0037]

[0038]

[0039] In the formula, y i and represents the predicted value, observed value and predicted mean, and n is the sample size.

[0040] Furthermore, the online optimization of air outlet volume control at the air conditioner terminal can achieve energy saving and carbon reduction in buildings:

[0041] (1) Calculation of comfort level in target area;

[0042] The hourly PMV value can be calculated based on the hourly temperature and humidity information detected by the indoor sensor;

[0043]

[0044] Where PMV is the comfort value, and is a constant coefficient, T and RH are temperature and relative humidity respectively.

[0045] (2) Accurate prediction of carbon emissions considering regional connections;

[0046] (3) Real-time online optimization of air volume at the air conditioning terminal; multi-objective optimization.

[0047] Furthermore, the carbon emissions are accurately predicted taking into account the partition relationship; the data collected in the early stage is directly input into the model after preprocessing, and the carbon emission intensity in the next hour is predicted through a sliding window method; at the same time, the constructed proxy model is used to predict the hourly PMV value.

[0048] Further, the multi-objective optimization process is as follows:

[0049] min.f(x)={f1(x),f2(x),f3(x),f4(x),f5(x),f6(x)}(14)

[0050] stg(x)={g1(x),g2(x),g3(x),g4(x)}<g req (15)

[0051] In the formula, f i(x),g(x)andg req They are fitness function, constant function and preset threshold respectively;

[0052] Parent population and progeny population is randomly generated; non-dominant solutions are grouped and distinguished by crowding distance; crowding distance is calculated as follows:

[0053]

[0054] Where d(x) i is the crowding distance, and M represents the number of optimization objects. m (i+1) and p m (i-1) are the target values ​​of the next individual and the previous individual of individual i on target m.

[0055] Another object of the present invention is to provide an air conditioning carbon reduction control system comprising:

[0056] The building parameter acquisition module is used to obtain the hourly energy consumption data of the air-conditioning terminal on the target floor during the cooling season, the outdoor hourly environmental parameters, the indoor hourly temperature and humidity, time, floor holiday calendar, and the hourly air volume information of the air-conditioning terminal;

[0057] The prediction module is used to build an encoder-decoder time series prediction model for time series prediction;

[0058] The optimization module is used to optimize the air volume control of air conditioner terminals online to achieve energy saving and carbon reduction in buildings.

[0059] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the air conditioning carbon reduction control method.

[0060] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to perform the steps of the air conditioning carbon reduction control method.

[0061] Another object of the present invention is to provide an information data processing terminal, which is used to implement the air-conditioning carbon reduction control system as described above.

[0062] In combination with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0063] First, the purpose of this invention is to propose an encoder-decoder model based on LSTM, GRU, SimpleRNN, CNN, and attention algorithms, taking into account the real-time high-precision prediction of carbon emissions in building partitions, and online optimization of the air supply volume of the air conditioner terminal in a single partition. It mainly solves the following problems:

[0064] (1) In order to address the problem that traditional building carbon reduction research is not sophisticated enough and cannot identify and locate abnormal carbon emission areas and time periods, the target floor (the ninth floor) is divided into three areas: west, south, and east according to the coverage of the power well. Considering the mutual connection between individual partitions, the real-time carbon emission data of each partition is used as the model input for time series prediction.

[0065] (2) In view of the fact that traditional deep learning models are unable to handle time series prediction and that there are problems such as gradient vanishing and gradient exploding in the process of processing long time series prediction, this study constructs an encoder-decoder algorithm framework that integrates LSTM, GRU, SimpleRNN, CNN, and attention modules to improve the accuracy and robustness of time series prediction.

[0066] (3) In order to address the problem that the air volume at the terminal of traditional air conditioners is fixed and the setting is subjective, an online optimization of the hourly air volume at the terminal of the air conditioner is proposed to minimize the building's carbon emissions while ensuring the comfort of indoor occupants.

[0067] The theoretical framework for reducing carbon emissions from building operation and maintenance provided by the present invention has the following advantages:

[0068] (1) The encoder-decoder based method proposed in this study can effectively analyze the relationship between regions and accurately predict the hourly carbon emissions of buildings. The outdoor and indoor hourly temperature, humidity, holidays, time, zone carbon emissions, and hourly wind volume information of each zone are used as input features, and the historical information of the above features is used to predict the carbon emission intensity of each zone in the next hour. The encoder-decoder model in this study can accurately capture the mutual connection between zones and achieve high-precision prediction of carbon emissions. This method can be used as an auxiliary means to quickly locate abnormal carbon emission areas or time periods. At the same time, the prediction results can be compared with the historical period or reference values, which solves the problem of difficult detection of abnormal carbon emissions in small areas.

[0069] (2) The encoder-decoder framework proposed in this study, which integrates SimpleRNN, GRU, CNN, Bi-GRU, and attention modules, can accurately extract key information from time series. SimpleRNN and CNN extract local features of time series, GRU extracts temporal information, attention focuses on key information in the sequence, and Bi-GRU can solve the problem of long-term memory forgetting. The designed encoding-decoding model structure can ensure the scientific nature of time series information capture and improve the accuracy of carbon emission prediction.

[0070] (3) The online optimization of air conditioning terminal air volume proposed in this study can effectively reduce the carbon emissions of buildings during the cooling season. Based on the constructed encoder-decoder agent model, the carbon emissions and comfort information of the next moment are predicted. At the same time, the range of the air supply volume of the air conditioning terminal is set, and the NSGA-II is used to search for the optimal air volume parameters at each moment to replace the initial air volume setting parameters of the air conditioning terminal. This online optimization process can realize the adaptive control of the air volume of the air conditioning terminal and achieve carbon reduction of the building while ensuring the comfort of indoor personnel.

[0071] The key technical points of the present invention are:

[0072] A real-time prediction of multi-zone carbon emissions based on the connection between building zone intervals is proposed, which can quickly locate abnormal carbon emission areas / periods and realize refined management of building carbon reduction.

[0073] An encoder-decoder hybrid model for time series prediction was constructed to make up for the problems that a single deep learning model cannot predict time series and that long time series information is forgotten.

[0074] A theoretical method for online optimization of the air supply volume at the air-conditioning terminal is proposed, which can dynamically adjust the air-conditioning air volume according to the actual comfort needs of the personnel, thereby significantly saving building energy.

[0075] Second, as auxiliary evidence of the inventiveness of the claims of the present invention, it is also reflected in the following important aspects:

[0076] (1) The expected benefits and commercial value of the technical solution of the present invention after transformation are:

[0077] The proposed online prediction and optimization scheme for high-rise building air-conditioning systems can improve the intelligence level of buildings, realize refined building management, and greatly improve the energy efficiency of buildings.

[0078] (2) The technical solution of the present invention fills the technical gap in the industry at home and abroad:

[0079] Through the prospect of existing literature and patents, we found that the research on carbon emission time series prediction and carbon reduction optimization has achieved remarkable success in the field of building energy conservation and emission reduction. Technologies such as the Internet of Things and big data provide new ideas for building operation and maintenance carbon reduction. Although there are many studies in this field, the existing research is not refined enough in terms of equipment optimization and control, and it is difficult to directly apply it to the dynamic response and carbon reduction optimization of building air conditioning systems. In view of this, this study proposes a theoretical framework for real-time prediction and online optimization of carbon emissions based on building zoning, which greatly reduces the carbon emission intensity of buildings.

[0080] The solution divides the target floor into three areas: west, south, and east, and divides them based on the coverage of the power well, while taking into account the interconnections between the various areas. This solves the problem of insufficient refinement in traditional building carbon reduction research, enabling the research to accurately identify and locate abnormal carbon emission areas and time periods. This refined division method greatly improves the accuracy of carbon emission monitoring and management, and provides more reliable data support for building energy optimization.

[0081] In view of the limitations of traditional deep learning models in time series prediction, such as the difficulty in dealing with the gradient vanishing and gradient exploding problems when predicting long time series, the solution proposes an encoder-decoder algorithm framework that integrates LSTM, GRU, SimpleRNN, CNN and attention modules. This framework combines the advantages of different models, not only can effectively capture the long-term dependencies in the time series, but also enhances the robustness and prediction accuracy of the model. Compared with a single model, the integrated framework performs well in the complex and changeable building carbon emission time series prediction, which helps to achieve more accurate carbon emission control.

[0082] The solution proposes a strategy for online optimization of the hourly air volume at the air conditioning terminal, which makes a significant improvement on the defects of the fixed and subjective air volume at the traditional air conditioning terminal. By adjusting the air volume of the air conditioner in real time, the solution can minimize the carbon emissions of the building while ensuring the comfort of indoor occupants. This optimization strategy not only improves the energy efficiency of the air conditioning system, but also provides an intelligent and dynamically adjusted solution for building energy management.

[0083] (3) The technical solution of the present invention solves the technical problems that people have been eager to solve but have never been able to solve successfully:

[0084] Building energy conservation and carbon reduction still mainly focuses on the entire building as the research target, facing problems such as low building refinement, low resource utilization, and poor comfort for personnel. The correlation between carbon emissions between various building partitions has rarely been explored, and the existing deep learning models have problems such as low accuracy when predicting carbon emission time series. In addition, the control of building air-conditioning systems tends to be subjective, and it is impossible to achieve simultaneous optimization and improvement of building carbon emissions and comfort. Traditional building optimization design focuses on the design stage, but most buildings have entered the operation and maintenance period, and related parameters are difficult to adjust. For this reason, the academic community has been committed to innovative research on building energy conservation and carbon reduction. Despite unremitting efforts, no major breakthroughs have been made so far.

[0085] Traditional building carbon emission research usually focuses on overall or extensive management, and it is difficult to conduct detailed analysis of different areas inside the building. This makes it difficult to identify and locate specific high-emission areas and abnormal emission periods in carbon emission control, which in turn affects the overall energy efficiency optimization effect. The present invention breaks through this technical bottleneck by subdividing the target floors of the building into multiple areas and dividing them based on the coverage of the power well, thereby achieving accurate monitoring and management of carbon emissions in the internal areas of the building, and solving the long-standing problem of detailed monitoring.

[0086] In the field of time series prediction, traditional deep learning models generally have problems such as gradient vanishing and gradient explosion when processing long time series data, making it difficult to obtain high-precision prediction results. This technical challenge has always plagued researchers, especially in complex building energy management. The present invention successfully solves this problem by constructing an encoder-decoder algorithm framework that integrates LSTM, GRU, SimpleRNN, CNN, and attention modules, and achieves high-precision time series prediction of building carbon emissions, providing strong technical support for energy optimization.

[0087] The terminal air volume of traditional air conditioning systems is fixed and the settings are subjective, which makes it difficult to achieve dynamic adjustments in different time periods, resulting in low energy efficiency management. This problem has long existed in building carbon reduction and has never been successfully solved. The online optimization strategy for hourly air volume at the air conditioning terminal proposed in this invention has successfully broken through this technical barrier. By adjusting the air volume in real time, it not only ensures indoor comfort, but also significantly reduces carbon emissions.

[0088] Fourth, the present invention proposes a new control method to address the shortcomings of existing air-conditioning systems in terms of energy conservation and carbon emission reduction, solving many problems of traditional air-conditioning systems in terms of energy consumption and carbon emission management. Existing technologies usually rely on static and fixed control strategies, which cannot adapt to complex building environments and changing external conditions, resulting in low operating efficiency of air-conditioning systems, serious energy waste, and high carbon emissions. In addition, existing systems lack real-time monitoring and accurate prediction methods for carbon emissions, making it difficult to effectively implement energy-saving and carbon emission reduction strategies.

[0089] The present invention significantly improves the intelligence level of the air conditioning system by introducing a dynamic control method based on time series prediction. By obtaining the hourly energy consumption, environmental parameters and air conditioning operation data of the building, the system can build an accurate encoder-decoder time series prediction model to make high-precision predictions of future energy consumption and carbon emissions. Compared with traditional methods, the present invention can make dynamic adjustments based on real-time data, avoid the rigidity and lag of traditional control strategies, and significantly improve the energy-saving effect and operating efficiency of the air conditioning system.

[0090] On this basis, the present invention further introduces an online optimization mechanism to perform real-time control of the air volume at the air conditioning terminal. By calculating the comfort index of the indoor environment (such as the PMV value) and combining the carbon emission prediction of the partition connection, the system can accurately control the air volume of the air conditioning in each area, which not only ensures the comfort of the indoor environment, but also achieves low carbon emissions. This optimization strategy breaks the one-size-fits-all operation mode of the traditional air conditioning system, can flexibly respond to different environments and needs, and further reduces overall energy consumption and carbon emissions.

[0091] The present invention has made significant technical progress on the basis of the existing technology. Through the control strategy combining dynamic prediction and real-time optimization, the present invention effectively solves the shortcomings of traditional air-conditioning systems in energy saving and carbon reduction, and realizes more refined and intelligent air-conditioning system management. This technology not only improves the energy utilization efficiency of buildings and reduces operating costs, but also provides strong technical support for achieving green buildings and sustainable development goals. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 It is a flow chart of the air conditioning carbon reduction control method provided by an embodiment of the present invention.

[0093] Figure 2 It is a structural block diagram of an air conditioning carbon reduction control system provided by an embodiment of the present invention.

[0094] Figure 3 It is a functional zoning diagram of a target floor provided in an embodiment of the present invention.

[0095] Figure 4It is a schematic diagram of the structure of the Encoder module provided in an embodiment of the present invention.

[0096] Figure 5 It is a schematic diagram of the structure of a Decoder module provided in an embodiment of the present invention.

[0097] Figure 6 It is a structural diagram of the Encoder-decoder model provided by an embodiment of the present invention.

[0098] Figure 7 It is a diagram of the NSGA-II individual selection process provided by an embodiment of the present invention.

[0099] Figure 8 It is a schematic diagram of online optimization of air-conditioning terminals provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0100] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0101] The following are two specific examples of industrial applications, showing the application scenarios of the air conditioning carbon reduction control method of the present invention in different fields:

[0102] Example 1: Air conditioning system optimization in commercial office buildings

[0103] In large commercial office buildings, air conditioning systems are usually one of the most energy-consuming devices and the main source of carbon emissions. Through the air conditioning carbon reduction control method of the present invention, the energy-saving and carbon reduction goals of the air conditioning system can be achieved according to the actual use of office buildings. First, the system obtains the energy consumption data, indoor and outdoor environmental parameters, and air volume information of each floor of the office building from the building management system, and performs data preprocessing. Then, an encoder-decoder time series prediction model is constructed based on these data to predict future energy consumption and carbon emissions. On this basis, the system optimizes the air volume control of the air conditioning terminal online, adjusts the air volume according to real-time demand, and ensures that energy consumption and carbon emissions are minimized while meeting comfort. This method can significantly reduce the air conditioning operating costs of large office buildings, while improving energy utilization efficiency and achieving sustainable development goals.

[0104] Example 2: Air conditioning system management in intelligent industrial plants

[0105] In intelligent industrial plants, the energy consumption and carbon emission management of air-conditioning systems are also crucial. Especially in some production workshops with strict requirements on temperature and humidity, the operating efficiency of the air-conditioning system directly affects the stability of the production process. The method of the present invention monitors the temperature, humidity and energy consumption data in the plant in real time, combines external environmental parameters, and constructs a time series prediction model to accurately predict the energy consumption and carbon emissions of the air-conditioning system. Subsequently, the system adjusts the air volume and air outlet temperature of the air-conditioning terminal of each workshop online according to the actual needs of different workshops through a multi-objective optimization algorithm. This method not only ensures the stability of the production environment, but also effectively reduces the overall energy consumption and carbon emissions of the plant, which helps enterprises achieve the goals of green manufacturing and low-carbon production.

[0106] The air conditioning carbon reduction control method of the present invention achieves the energy-saving and carbon reduction goals of the building air conditioning system through the coordinated work of multiple steps. First, the system obtains the necessary parameter data from the actual operation of the building, including the hourly energy consumption data of the air conditioning terminal, outdoor environmental parameters, indoor temperature and humidity, floor holiday calendar, etc. After preprocessing, these data can ensure that the model building and optimization processes in the subsequent steps have accurate and complete basic data support. During the preprocessing process, the data is interpolated to fill the data gaps in the time series, and the carbon emissions at each moment are calculated based on the relationship between power consumption and carbon emission factors, laying the foundation for subsequent carbon emission predictions.

[0107] Next, using these processed building parameter data, the system constructed a time series prediction model based on the encoder-decoder structure. The Encoder module of this model consists of SimpleRNN, GRU and CNN, which is responsible for encoding the input data in time series and capturing the temporal characteristics of the data. The GRU model effectively retains and updates the information in the hidden state by updating the gate and resetting the gate structure, thereby improving the model's ability to capture long sequence dependencies. The Bi-GRU hybrid structure combines the advantages of forward and reverse GRU, further enhancing the model's understanding of data temporal dependencies, while the introduction of the attention mechanism enables the model to focus on the most critical input data, improving the accuracy and efficiency of predictions.

[0108] After completing the model construction and training, the system enters the online optimization stage to control the air volume of the air conditioner terminal in real time to achieve the goal of building energy conservation and carbon reduction. Specifically, the system calculates the hourly PMV value based on the temperature and humidity data obtained by the indoor sensor to evaluate the comfort level of the target area, and accurately predicts carbon emissions based on the connection between different partitions. On this basis, the system uses a multi-objective optimization algorithm to adjust the air volume output of the air conditioner terminal in real time to ensure that carbon emissions are minimized while meeting the comfort requirements. This optimization process combines the prediction results of the agent model and the connection between the partitions, making the control more intelligent and precise.

[0109] Finally, in order to further optimize the performance of the system, the system introduced an algorithm based on the fitness function in the multi-objective optimization process, and combined the concept of crowding distance to group the solution space. By calculating the crowding distance of non-dominant solutions, the optimization objectives are distinguished, thereby effectively avoiding the local optimal problem in the optimization process and improving the overall optimization effect. This optimization strategy not only ensures the operating efficiency of the air-conditioning system, but also significantly reduces energy consumption and carbon emissions, fully realizing the technical progress of the present invention in building energy conservation and carbon reduction.

[0110] like Figure 1 As shown, an air conditioning carbon reduction control method provided by an embodiment of the present invention comprises the following steps:

[0111] S101: Acquisition of building parameters;

[0112] (1) Acquisition of hourly energy consumption data of air conditioner terminals on the target floor during the cooling season, outdoor hourly environmental parameters, indoor hourly temperature and humidity, time, floor holiday calendar, hourly air volume of air conditioner terminals, and other information. Building energy consumption data mainly refers to hourly energy consumption of air conditioner terminals, which is measured by a single partition meter on each floor. The partition results are as follows: Figure 3 As shown in the figure. Due to the tripping and failure of the meter, some of the collected data is missing, and the collected data needs to be preprocessed as follows:

[0113]

[0114] Formula (1), t l is the time to be interpolated, t l-1 and t l+1 The time t l The moment before and the moment after y l-1 ,y l and l+1 The corresponding time t l-1 , t l and t l+1 of carbon emissions.

[0115] CE elec =Eelec ×EF elec (2)

[0116] E elec and EF elec are electricity consumption and carbon emission factor (EF elec =0.5703), CE elec It is the carbon emission corresponding to the electricity consumption.

[0117] S102: Construct an encoder-decoder time series prediction model for time series prediction.

[0118] (1) The Encoder module consists of SimpleRNN, GRU and CNN, and its structure is as follows. As a commonly used natural language processing algorithm, the Seq2seq model has received increasing attention in the field of civil engineering. The model generally contains encoder and decoder modules, which can convert time series into sequences that are easy for computers to understand and output the final sequence. Considering the special structure of time series, RNN usually plays an important role in capturing long-term dependencies in time series. However, when extracting valuable information from longer time series, RNN will have the problem of gradient disappearance or gradient explosion. Therefore, based on previous research, we further optimized the model structure to improve its performance in predicting building carbon emissions. The encoder consists of three neural network structures, namely SimpleRNN, GRU and CNN, and its framework is as follows Figure 4 shown.

[0119] (2) CNN can extract valuable information from time series. We use a one-dimensional convolutional neural network (1D-CNN) as a feature extraction module. This structure cannot be directly used for feature extraction of time series. Therefore, two other improved RNN algorithm structures (i.e., SimpleRNN and GRU) are used for time series prediction. SimpleRNN, as one of the basic structures of RNN, can link the current output with the output of the previous unit. More specifically, this model can memorize previous information and use it for time series prediction. These characteristics enable the traditional RNN model to learn an excellent structure that integrates historical data in specific situations where the latest prediction is divided into a small range. However, when dealing with long-term time series, SimpleRNN usually suffers from the problem of gradient disappearance or explosion. The RNN model structure is as follows:

[0120] h t =σ(wx t +Uh t-1 +b) (1)

[0121] In the formula, x t and h tCorresponding to the input variables and hidden states at time t, w and U are the matrices corresponding to the current unit and input variables, respectively, and b is the bias of the neuron.

[0122] (3) Considering the insufficient performance of SimpleRNN in achieving long-term time series prediction, scholars proposed LSTM and GRU to solve this problem. As an improved form of RNN, LSTM has three gates inside, namely input gate, forget gate and output gate, which are used to control block state and output. Like LSTM, GRU also uses gates to achieve long-term time series prediction. Since it merges the input gate and forget gate into one update gate, the training efficiency is higher. In this study, GRU is used for long-term time feature extraction. The GRU model structure is as follows:

[0123] z t =σ(x t w xr +h t-1 w hr +b r ) (2)

[0124] r t =σ(x t w xz +h t-1 w hz +b z ) (3)

[0125]

[0126] In the formula, σ and tanh are sigmoid function and activation function respectively, w xr and w hr is the update gate weight, w xz and w hz is to reset the gate weight. xh and w hh is a candidate hidden layer state; x t is the input at time t, h t-1 is the hidden layer t-1 matrix, r t is the reset gate, z t is the update gate, b r , b z and b h is the bias term.

[0127] (4) In the encoder, these initial data are encoded and stored in the hidden layer, which decodes them into a new sequence and outputs the prediction results to the decoder. We designed the encoder using GRU, 1D-CNN, and SimpleRNN. The output of this part includes the feature information and time series information extracted by GRU and SimpleRNN, respectively. As the length of the time series increases, the performance of GRU will decrease due to the generation of too much fuzzy information. Therefore, a Bi-GRU consisting of a forward GRU and a backward GRU is used to decode the sequence, thereby mining the valuable information between the input and the prediction variance, thereby improving the prediction performance of the carbon emission intensity of the air-conditioning terminal on the target floor. Figure 5 The proposed Bi-GRU hybrid structure and attention mechanism are demonstrated.

[0128]

[0129] In the formula, and are the hidden layers of the forward GRU and the reverse GRU respectively, and F is the combination of the bidirectional GRU.

[0130] (5) Building carbon emission intensity is affected by many uncertain factors, such as weather conditions, seasons, building calendar information, etc. We want to pay more attention to these key factors and exclude the interference of unimportant factors. Therefore, we introduced the attention mechanism to help us identify these factors by giving more weight to the main factors. The attention mechanism is designed based on the human brain behavior model, and it can automatically focus on the main features of the target we observe. The calculation method of the attention mechanism is as follows:

[0131] e t =utanh(wh t +b) (7)

[0132]

[0133] In the formula, e t is the probability distribution, u and w are weight matrices, b and s t are the bias term and the attention mechanism output, respectively, t is the distribution coefficient.

[0134] (6) Figure 6The encoder module based on SimpleRNN-CNN-GRU and the decoder module based on Bi-GRU-Attention constitute the structure of the encoder-decoder model in this study. The role of the encoder module is to convert the input variables into a fixed-length sequence according to its neural layer, while the decoder outputs our prediction results. The asymmetric design of the encoder and decoder structure enables our model to complete more challenging encoding tasks, capture more complex correlations, and improve the performance of the model.

[0135] In this study, three commonly used regression indicators, including the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE), were used to highlight the prediction performance of the constructed encoder-decoder model. R2 ranges between 0 and 1, and an indicator close to 1 indicates excellent model performance. On the contrary, when the other evaluation indicators related to the error are 0, it means that the model performance is excellent.

[0136]

[0137] In the formula, y i and represents the predicted value, observed value and predicted mean, and n is the sample size.

[0138] S103: Online optimization of air outlet volume control at air conditioner terminals to achieve energy saving and carbon reduction in buildings;

[0139] (1) Calculation of comfort in the target area. Usually, personnel comfort is a qualitative indicator and is difficult to describe quantitatively. Common comfort indicators include PMV, PPD, and uncomfortable hours (hrs). This study uses the PMV indicator to describe personnel comfort, and the empirical calculation formula is as follows. From formula (13), it can be seen that comfort is a binary linear relationship between indoor temperature and humidity. Therefore, the hourly PMV value can be calculated based on the hourly temperature and humidity information detected by the indoor sensor.

[0140]

[0141] Where PMV is the comfort value, and is a constant coefficient, T and RH are temperature and relative humidity respectively.

[0142] (2) Accurate prediction of carbon emissions considering regional connections. The pre-collected data is directly input into the model after preprocessing, and the carbon emission intensity in the next hour is predicted through a sliding window method. At the same time, the constructed proxy model (encoder-decoder) is used to predict the hourly PMV value.

[0143] (3) Figure 7Real-time online optimization of air volume at the air conditioning terminal. Based on the encoder-decoder time series prediction agent model constructed in the early stage, the carbon emissions and comfort level at a certain moment are first predicted, and then the NSGA-Ⅱ algorithm is used to optimize the air volume corresponding to each moment. The optimized air volume replaces the initial air volume at the air conditioning terminal. And so on, the hourly air volume at the terminal is gradually optimized. NSGA-II is an optimization algorithm developed for multi-objective optimization problems. Compared with NSGA (non-dominated sorting genetic algorithm), the second-generation algorithm comprehensively considers the dominant solution and its dominated solutions. The crowding distance, which calculates the closeness of the solution to its neighboring solutions, is used to evaluate the fitness. The multi-objective optimization process is as follows:

[0144] min.f(x)={f1(x),f2(x),f3(x),f4(x),f5(x),f6(x)} (14)

[0145] stg(x)={g1(x),g2(x),g3(x),g4(x)}<g req (15)

[0146] In the formula, f i (x),g(x)andg req They are fitness function, constant function and preset threshold respectively.

[0147] Parent population and progeny population is randomly generated. The non-dominant solutions are grouped and distinguished by crowding distance. The crowding distance is calculated as follows:

[0148]

[0149] Where d(x) i is the crowding distance, M represents the number of optimization objects, and p m (i+1) and p m (i-1) are the target values ​​of the next individual and the previous individual of individual i on target m.

[0150] Figure 8 Schematic diagram of online optimization of air conditioning terminal

[0151] like Figure 2 As shown, another object of the present invention provided by the embodiment of the present invention is to provide an air conditioning carbon reduction control system comprising:

[0152] The building parameter acquisition module is used to obtain the hourly energy consumption data of the air-conditioning terminal on the target floor during the cooling season, the outdoor hourly environmental parameters, the indoor hourly temperature and humidity, time, floor holiday calendar, and the hourly air volume information of the air-conditioning terminal;

[0153] The prediction module is used to build an encoder-decoder time series prediction model for time series prediction;

[0154] The optimization module is used to optimize the air volume control of air conditioner terminals online to achieve energy saving and carbon reduction in buildings.

[0155] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the air conditioning carbon reduction control method.

[0156] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to perform the steps of the air conditioning carbon reduction control method.

[0157] Another object of the present invention is to provide an information data processing terminal, which is used to implement the air-conditioning carbon reduction control system as described above.

[0158] On the ninth floor of a commercial complex, we subdivided the floor into three areas: west, south, and east according to the coverage of the power well, and collected the carbon emission data of each area in real time and input it into the prediction model. Compared with the traditional overall monitoring method, our solution can accurately locate the carbon emission anomalies in a specific area, significantly improving the monitoring accuracy and response efficiency of carbon emissions. This refined division scheme verifies the creativity of the present invention in regional carbon emission management and can effectively improve the energy efficiency of buildings.

[0159] We used an encoder-decoder algorithm framework that integrates LSTM, GRU, SimpleRNN, CNN, and attention modules to predict the historical carbon emission data of the building floors. During the experiment, the model of the present invention not only showed better prediction accuracy than the traditional single model, but also overcame the common gradient vanishing and gradient exploding problems in time series prediction, further proving the technical advantages of the present invention in processing complex time series data.

[0160] In the building air conditioning system, we applied the online optimization technology of hourly air volume at the air conditioning terminal. By adjusting the air volume of the air conditioner in real time, we ensured that carbon emissions were minimized without reducing indoor comfort. Compared with the traditional fixed air volume control, this solution reduced energy consumption and significantly reduced carbon dioxide emissions. This optimization strategy verified the practical application value of this invention in improving the energy efficiency management of air conditioning systems.

[0161] Through the verification of the above embodiments, we have further confirmed the creativity and technical value of the technical solution of the present invention, which can not only solve long-standing technical problems, but also has broad application prospects.

[0162] The control method of traditional building air conditioning equipment is single and not refined enough, and the temperature of the area on the same floor is often too high or too low, as shown in Table 1. This phenomenon not only causes a huge waste of building energy, but also causes discomfort indoors. Therefore, this study divides the floors into zones. On the one hand, the air conditioning setting temperature in the zone is low, which is bound to increase the carbon emissions of the building. Predicting the carbon emissions of the zones can quickly predict the carbon emissions of the zones, and can quickly determine whether the carbon emissions are abnormal compared with the same period in history.

[0163] Table 1 Indoor occupant comfort feedback on some floors

[0164]

[0165] The distribution of people in each zone is different, and the demand for cooling and resource consumption are also different. Optimizing the comfort and carbon emission targets of each zone can minimize carbon emissions. Traditional air conditioning experience control uses only one set temperature or air volume every day, but the hourly changes in people in each zone cause the demand for cooling to fluctuate. This study optimizes the comfort and carbon emissions of each zone every hour to obtain the optimal air volume at the next moment.

[0166] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0167] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. An air conditioning carbon reduction control method, characterized in that: The following steps are involved: Step 1: Acquisition of building parameters; Step 2: Build an encoder-decoder time series prediction model for time series prediction; Step three: Online optimization of air outlet control at the air conditioning terminal to achieve energy conservation and carbon reduction in buildings.

2. The air conditioning carbon reduction control method according to claim 1, characterized in that: The method for obtaining the building parameters is as follows: Acquisition of hourly energy consumption data of air-conditioning terminals on target floors during the cooling season, outdoor hourly environmental parameters, indoor hourly temperature and humidity, time, floor holiday calendar, and hourly air volume information of air-conditioning terminals; The collected data is preprocessed as follows: Formula (1), t l is the time to be interpolated, t l-1 and t l+1 The time t l The moment before and the moment after ; y l-1 ,y l and l+1 The corresponding time t l-1 , t l and t l+1 carbon emissions; WHAT elec =E elec ×EF elec (2) E elec and EF elec are electricity consumption and carbon emission factor (EF elec =0.5703), CE elec It is the carbon emission corresponding to the electricity consumption.

3. The air conditioning carbon reduction control method according to claim 1, characterized in that: The encoder-decoder time series prediction model is constructed for time series prediction: (1) The Encoder module consists of SimpleRNN, GRU and CNN; (2) The RNN model structure is as follows: h t =σ(wx t +Uh t-1 +b)(1) In the formula, x t and h t Corresponding to the input variables and hidden states at time t, w and U are the matrices corresponding to the current unit and input variables, respectively, and b is the bias of the neuron; (3) The GRU model structure is as follows: z t =σ(x t w xr +h t-1 w hr +b r )(2) r t =σ(x t w xz +h t-1 w hz +b z )(3) In the formula, σ and tanh are sigmoid function and activation function respectively, w xr and w hr is the update gate weight, w xz and w hz is to reset the gate weight; x t is the input at time t, h t-1 is the hidden layer t-1 matrix; (4) Bi-GRU hybrid structure and attention mechanism; In the formula, and are the hidden layers of the forward GRU and the reverse GRU respectively, and F is the combination of the bidirectional GRU; (5) The calculation method of the attention mechanism is as follows: e t =utanh(wh t +b)(7) In the formula, e t is the probability distribution, u and w are weight matrices, b and s t They are bias term and attention mechanism output respectively; (6) Coefficient of determination R2, root mean square error RMSE and mean absolute error MAE; In the formula, y i and represents the predicted value, observed value and predicted mean, and n is the sample size.

4. The air conditioning carbon reduction control method according to claim 1, characterized in that: The online optimization of air outlet volume control at the air conditioner terminal can achieve energy saving and carbon reduction in buildings: (1) Calculation of comfort level in target area; The hourly PMV value can be calculated based on the hourly temperature and humidity information detected by the indoor sensor; (2) Accurate prediction of carbon emissions considering regional connections; (3) Real-time online optimization of air volume at the air conditioning terminal; multi-objective optimization.

5. The air conditioning carbon reduction control method according to claim 4, characterized in that: The carbon emission is accurately predicted considering the partition relationship; the data collected in the early stage is directly input into the model after preprocessing, and the carbon emission intensity in the next hour is predicted through a sliding window method; at the same time, the constructed proxy model is used to predict the hourly PMV value.

6. The air conditioning carbon reduction control method according to claim 4, characterized in that: The multi-objective optimization process is as follows: min.f(x)={f1(x),f2(x),f3(x),f4(x),f5(x),f6(x)}(14) s.t.g(x)={g1(x),g2(x),g3(x),g4(x)}<g req (15) In the formula, f i (x),g(x)andg req They are fitness function, constant function and preset threshold respectively; Parent population and progeny population is randomly generated; non-dominant solutions are grouped and distinguished by crowding distance; crowding distance is calculated as follows: Where d(x) i is the crowding distance, and M represents the number of optimization objects.

7. An air conditioning carbon reduction control system implementing the air conditioning carbon reduction control method according to any one of claims 1 to 6, characterized in that: The air conditioning carbon reduction control system comprises: The building parameter acquisition module is used to obtain the hourly energy consumption data of the air-conditioning terminal on the target floor during the cooling season, the outdoor hourly environmental parameters, the indoor hourly temperature and humidity, time, floor holiday calendar, and the hourly air volume information of the air-conditioning terminal; The prediction module is used to build an encoder-decoder time series prediction model for time series prediction; The optimization module is used to optimize the air volume control of air conditioner terminals online to achieve energy saving and carbon reduction in buildings.

8. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the air conditioning carbon reduction control method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the air-conditioning carbon reduction control method according to any one of claims 1 to 6.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the air conditioning carbon reduction control system as described in claim 7.

Images