Heating and ventilation air conditioner heat supply prediction method, system, equipment and medium

Abstract

The invention provides a heating ventilation air conditioner heat supply prediction method, system and device and a medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: collecting adjustable and controllable parameters of the heating ventilation heat supply system, external environment data, related data of the heating ventilation air-conditioning heat supply system and indoor environment data, and processing the collected data to generate a time sequence data set; constructing a time sequence model based on the Mama architecture and the state space model, and training the time sequence model by using the preprocessed time sequence data set, so that the time sequence model learns time sequence rule information in a normal operation mode of the heating ventilation air conditioning system; and dynamically calling related parameters of the heating ventilation air conditioner by using the time sequence model, predicting the heat supply capacity of the heating ventilation air conditioner system, and generating corresponding control parameters according to a prediction result. According to the method, the time sequence characteristics in the operation process of the heating ventilation air-conditioning system can be accurately captured, so that the heating capacity of the heating ventilation air-conditioning system is accurately controlled.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and more specifically relates to a heating prediction method, system, device and medium for heating ventilation and air conditioning. Background Art

[0002] With the development of modern buildings and the increasing demand for indoor environmental comfort, efficient operation and precise control of HVAC systems have become essential. Traditional HVAC control methods are often based on simple sensor feedback and preset rules, such as starting or stopping heating when the temperature sensor detects that the temperature is higher or lower than the set value. However, this method has many limitations. On the one hand, the indoor environment is a complex dynamic system that is affected by many factors, including outdoor weather changes, indoor human activities, and air humidity. Simple rules are difficult to adapt to these complex changes, which may lead to waste of thermal energy and poor comfort for personnel. On the other hand, traditional control methods lack the ability to effectively utilize historical data and predict future trends, and cannot achieve forward-looking control.

[0003] In recent years, artificial intelligence technology has made significant progress in various fields. Many studies have proposed methods that can capture effective features in time series, providing new ideas for intelligent control of HVAC. Although artificial intelligence technology has been applied to HVAC parameter reasoning, such as recurrent neural networks (RNNs) that model continuous time points based on the Markov hypothesis, convolutional neural networks (TCNs) that extract change information along the time dimension, and TimesNet that extracts multi-period features of time series.

[0004] However, the inference methods of related technologies still have some problems when processing time series data, such as high requirements for training data quality and limited ability to understand different types of long sequence data. In addition, when processing long sequence data, the current method not only has a complex training process, but also has the problem of poor inference speed when applied to the production environment, which seriously limits its application and development in the field of HVAC. Therefore, a new intelligent parameter prediction method is needed to overcome these problems and improve the heating efficiency of HVAC systems. Summary of the invention

[0005] In view of the above problems, the purpose of the present invention is to provide a HVAC heating prediction method, system, device and medium, which utilizes the advantages of the state space model (SSM) in the Mamba architecture in processing time series data, and can accurately capture the time series characteristics of the HVAC system during operation, thereby achieving precise control of the heating capacity of the HVAC system.

[0006] In order to achieve the above object, the present invention is implemented through the following technical solutions: In a first aspect, an embodiment of the present application provides a HVAC heating prediction method, comprising: Collect the adjustable parameters of the HVAC system, external environment data, HVAC system related data and indoor environment data, clean and normalize the collected data, and generate a time series data set; A time series model is built based on the Mamba architecture and state space model, and the preprocessed time series data set is used to train the time series model so that the time series model can learn the time series regularity information under the normal operation mode of the HVAC system. By using the trained time series model, the relevant parameters of HVAC are dynamically called for fast recursive reasoning to predict the heating capacity of the HVAC system and generate corresponding control parameters based on the prediction results.

[0007] In an optional embodiment, the method comprises: Monitor the execution results of control parameters, record relevant energy consumption information, and feed the energy consumption information back to the time series model to optimize the control parameters.

[0008] In an optional embodiment, the adjustable parameters include the opening of the primary side of the plate heat exchanger, the frequency of the water pump, and the frequency of the terminal fan coil; the external environment data includes outdoor temperature and humidity sensor data, floor infrared radiation data, and camera-based number monitoring data; the HVAC heating system related data includes municipal heating calorimeter data, plate heat exchanger primary and secondary side supply and return water pressure transmitters, flow meters and temperature sensors collected corresponding data; the indoor environment data includes temperature data at different locations sensed by the temperature sensor, air humidity data monitored by the humidity sensor, and the number of people and activity range information detected by a combination of radar modules, infrared sensors and high-resolution cameras.

[0009] In an optional implementation, the constructing of a time series model based on the Mamba architecture and the state space model includes: Configure the input layer of the time series model and pass the input sequence It is expressed as:

[0010] in, Defined as specific data 1,17]: ; The hidden layer of the time series model is constructed according to the Mamba architecture, and the expressiveness of the model is enhanced by using the residual links, linear layers, and activation functions of the Transformer.

[0011] Construct the output layer of the time series model and set the primary side opening of the plate heat exchanger , Pump frequency , fan coil frequency , Indoor temperature prediction value , heat supply forecast value is the output parameter; the output prediction sequence is expressed as: , , where H is the process hidden layer.

[0012] In an optional embodiment, the hidden layer of the time series model constructed according to the Mamba architecture includes: The state of the hidden layer at time step t It is expressed as:

[0013]

[0014] in, , , is a learnable weight matrix; Set up the Mamba module, which includes the linear layer Linear, the convolution layer Conv, the gating mechanism layer σ and the state space model SSM layer. When the HVAC system and the environment state are When , the Mamba module is represented as:

[0015]

[0016]

[0017] Use 12 layers of Mamba module stacking to fully explore the effective features in time series data, and hide the process layer It is expressed as: .

[0018] In an optional embodiment, the method utilizes the trained time series model to dynamically call the relevant parameters of the HVAC system for fast recursive reasoning, predicts the heating capacity of the HVAC system, and generates corresponding control parameters according to the prediction results, including: During the operation of the HVAC system, the adjustable parameters of the HVAC system, external environmental data, HVAC system related data and indoor environmental data are collected in real time and input into the time series model at set time intervals; The corresponding control parameters are generated according to the prediction results through the time series model.

[0019] In an optional embodiment, the monitoring of the execution results of the control parameters, recording the relevant energy consumption information, and feeding the energy consumption information back to the time series model to optimize the control parameters includes: Control the HVAC system based on control parameters, monitor indoor environmental comfort indicators, and record the energy consumption and energy consumption trends of the HVAC system; The indoor environmental comfort index, energy consumption of the HVAC system and energy consumption trend are fed back to the time series model to generate optimized control parameters.

[0020] In a second aspect, the embodiment of the present application further provides a HVAC heating prediction system, comprising: The data acquisition and preprocessing module is used to collect the adjustable parameters of the HVAC system, external environment data, HVAC system related data and indoor environment data, clean and normalize the collected data, and generate a time series data set; The model building and training module is used to build a time series model based on the Mamba architecture and state space model, and use the preprocessed time series data set to train the time series model so that the time series model can learn the time series regularity information under the normal operation mode of the HVAC system; The inference and prediction module is used to use the trained time series model to dynamically call the relevant parameters of the HVAC for fast recursive reasoning, predict the heating capacity of the HVAC system, and generate corresponding control parameters based on the prediction results.

[0021] In a third aspect, an embodiment of the present application further provides an electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the steps of the HVAC heating prediction method as described in any one of the above items are implemented.

[0022] In a fourth aspect, an embodiment of the present application further provides a storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the HVAC heating prediction method as described in any one of the above items are implemented.

[0023] It can be seen from the above technical solutions that the present invention has the following advantages: In the HVAC heating prediction method provided in this application, the advantages of the state space model (SSM) in the Mamba architecture in processing time series data can be used to accurately capture the time series characteristics of the HVAC system during operation, thereby achieving precise control of the heating capacity of the HVAC system. This method combines the advantages of parallel training and recursive reasoning of recurrent neural networks. During model training, SSM can be used in parallel to utilize historical HVAC operation data, real-time sensor data, and external environment data to greatly improve the training speed. During reasoning and prediction, this method can dynamically call the relevant parameters of the HVAC and use SSM for fast recursive reasoning. At the same time, thanks to the richness of training data, the present invention can achieve the best balance between comfort and heating flow.

[0024] This application uses Mamba to mine the complex correlation features in HVAC-related time series data to achieve energy-saving operation of HVAC based on multi-source data.

[0025] This application integrates different time scale characteristics and external environmental data such as heating flow, pipeline pressure, inlet and outlet water temperature, indoor and outdoor temperature and humidity, and outdoor light, and uses the Mamba architecture to establish a HVAC energy-saving operation state space model to monitor and analyze multi-category time series data in real time. By building a system based on the state space model and gating mechanism, the optimal heating flow of the system at the target room temperature can be obtained.

[0026] This application realizes parameter analysis and control of the HVAC heating system, effectively improving the thermal energy utilization rate of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the description will be briefly introduced below. Obviously, the accompanying drawings in the following description 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.

[0028] Figure 1 A schematic diagram of the flow chart of the HVAC heating prediction method provided in this application.

[0029] Figure 2 A sequential Mamba network structure diagram provided for this application.

[0030] Figure 3 A schematic diagram of the structure of the HVAC heating prediction system provided in this application.

[0031] Figure 4 A schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0032] In the specific steps of the building cooling and heating load adaptive calculation method described in detail below, various embodiments of the present disclosure will be described more comprehensively. The present disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but the present disclosure should be understood to cover all adjustments, equivalents and / or alternatives that fall within the spirit and scope of the various embodiments of the present disclosure.

[0033] Hereinafter, the terms "include" or "may include" used in various embodiments of the present disclosure indicate the presence of disclosed functions, operations, or elements, and do not limit the addition of one or more functions, operations, or elements. In addition, as used in various embodiments of the present disclosure, the terms "include", "have", and their cognates are intended only to indicate specific features, numbers, steps, operations, elements, components, or a combination of the foregoing, and should not be understood as first excluding the presence of one or more other features, numbers, steps, operations, elements, components, or a combination of the foregoing or the possibility of adding one or more features, numbers, steps, operations, elements, components, or a combination of the foregoing.

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] See also Figure 1 The figure is a flow chart of a method for predicting heating supply of HVAC in a specific embodiment, the method comprising: S1: Collect the adjustable parameters of the HVAC system, external environment data, HVAC system related data and indoor environment data, clean and normalize the collected data, and generate a time series data set.

[0036] In this step, data acquisition and preprocessing are mainly responsible for collecting and processing four types of data, namely, the adjustable parameters of the HVAC heating system ( )、External environment data( )、HVAC heating system related data( )、Indoor environment data( ), where i represents a data type under this category of data, and t represents the data type at a certain time point under this category of data.

[0037] As an example, the adjustable parameters include the opening of the primary side of the plate heat exchanger, the frequency of the water pump, and the frequency of the terminal fan coil; the external environment data includes outdoor temperature and humidity sensor data, floor infrared radiation data, and camera-based number monitoring data; the HVAC heating system related data includes municipal heating calorimeter data, plate heat exchanger primary and secondary side supply and return water pressure transmitters, flow meters and temperature sensors collected corresponding data; indoor environment data includes temperature data at different locations sensed by temperature sensors, air humidity data monitored by humidity sensors, and information on the number of people and activity range detected by a combination of radar modules, infrared sensors and high-resolution cameras.

[0038] In this step, the collected data is preprocessed, including data cleaning and normalization. Outliers and erroneous data are removed to ensure data quality and consistency. Different types of data are unified into the same numerical range for subsequent model processing. The processed data is input into the Mamba time series algorithm. The input parameters can be expressed as .

[0039] The specific parameters are shown in Table 1: Table 1: HVAC system parameter reasoning core parameter table

[0040] The ultimate goal is to achieve the minimum municipal heating supply while meeting the user's comfort requirements. The influencing factors include external environmental data ( )、HVAC heating system related data( )、Indoor environment data( ), municipal heating It is expressed as: .

[0041] S2: A time series model is constructed based on the Mamba architecture and state space model, and the time series model is trained using the preprocessed time series dataset so that the time series model can learn the time series regularity information under the normal operating mode of the HVAC system.

[0042] It should be noted that the Mamba architecture is a model structure optimized for processing time series data. Building a time series model based on the time series Mamaba architecture to realize HVAC energy-saving parameter reasoning mainly includes two steps: time series model design stage and time series model training stage. Among them, the time series model stage refers to designing a time series model based on the Mamba architecture, and designing each hidden layer structure so that it can effectively capture the dependencies in long series data. The time series model stage refers to using the time series data set obtained in the data collection and preprocessing stage to train the time series model, so that the time series model can learn the time series regularity information under the normal operation mode of the HVAC system.

[0043] S3: Using the trained time series model, dynamically call the relevant parameters of HVAC for fast recursive reasoning, predict the heating capacity of the HVAC system, and generate corresponding control parameters based on the prediction results.

[0044] In a specific implementation, this step implements inference prediction using a time series model. Specifically, the relevant data obtained during the operation of the production system is input into the trained time series model to implement inference prediction of the system's energy-saving parameters. The time series model based on the time series Mamba architecture designed by the present invention has a stronger ability to process time series data to adapt to the energy-saving control needs of HVAC in different usage scenarios. Moreover, as time goes by and data continues to accumulate, the model can automatically learn and adapt to new environmental change patterns, and the control effect will continue to be optimized without frequent manual intervention. The inference and prediction stage consists of two major parts: real-time data acquisition and control parameter generation, feedback and continuous optimization.

[0045] S4: Monitor the execution results of the control parameters, record the relevant energy consumption information, and feed the energy consumption information back to the time series model to optimize the control parameters.

[0046] In this embodiment, the method constructs a high-quality time series data set by comprehensively collecting adjustable parameters of the HVAC system, external environment data, HVAC system related data and indoor environment data, and after fine data cleaning and normalization. The time series model based on the Mamba architecture and state space model can make full use of the information in these data sets, learn the time series laws of the HVAC system in normal operating mode, and thus achieve accurate prediction of heating capacity.

[0047] Through the trained time series model, this method can dynamically call the relevant parameters of HVAC, perform fast recursive reasoning, and timely predict the heating capacity of the system. This not only improves the accuracy and real-time performance of the prediction, but also provides strong support for the subsequent generation of control parameters. According to the prediction results, the system can automatically generate the corresponding control parameters to achieve precise control of the HVAC system, thereby improving heating efficiency and reducing unnecessary energy waste.

[0048] In addition, this method also focuses on monitoring the execution results of control parameters and recording relevant energy consumption information. By feeding this information back to the time series model, the control parameters can be further optimized and the overall performance of the system can be improved. This closed-loop feedback mechanism not only helps to improve heating efficiency, but also effectively reduces energy consumption and realizes green and energy-saving HVAC system operation. In summary, this method has shown significant beneficial effects in improving heating prediction accuracy, optimizing control parameters, and reducing energy consumption.

[0049] In an embodiment of the present invention, based on step S2, a possible embodiment is given below to illustrate its specific implementation scheme in a non-limiting manner.

[0050] The time series model of the present invention is built on Mamba, which adopts a state space model (SSM) with selective sequence scanning capability, can capture long-term dependencies, and has linear time complexity. The method is combined with the Transformer decoder structure to integrate the sequence processing advantages of the state space model and the mask attention layer.

[0051] As an example, step S2 specifically includes a model design phase and a model training phase.

[0052] 1. Model design stage: The input layer of the time series model of the present invention can be based on the timestamp dimension of 17 types of collected data Custom configuration is used to receive pre-processed sensor data and environmental data, which are input into the model in the form of set time steps, input sequence It can be expressed as:

[0053] in, Defined as specific data 1,17]:

[0054] The hidden layer of the time series model is constructed according to the Mamba architecture design, and the model's expressiveness is enhanced by using the residual links, linear layers, and activation functions of the Transformer. The Mamba architecture SSM model, with its RNN-like recursive calculation mechanism, shows a unique advantage in inference speed when processing HVAC-related time series data. It can effectively capture the hidden dependencies in long sequence data through its complex internal structure. Assuming that the model is given a current time step Time series of HVAC system and environmental conditions ,in is the time step The training goal of the model is to predict the next time step HVAC and environment related status values , based on this data, the optimal heat-saving parameters of the relevant controllable equipment are inferred.

[0055] The state space model SSM assumes that the current time step The hidden layer state of the model can be obtained from the previous time step The hidden layer state is obtained, and the hidden layer is at time step It can be expressed as:

[0056]

[0057] in, , , is a learnable weight matrix.

[0058] like Figure 2 As shown, the Mamba module consists of a linear layer , convolutional layer , Gating mechanism layer and state space model SSM layer, when the HVAC system and environmental state are When , the Mamba module can be expressed as:

[0059]

[0060]

[0061] According to the richness of HVAC system related data in the dataset, this method uses 12 layers of Mamba module stacking to fully mine the effective features in time series data. It can be expressed as:

[0062] The relevant parameters of HVAC heating output at the output layer include the primary side opening of the plate heat exchanger , Pump frequency , fan coil frequency , Indoor temperature prediction value , heat supply forecast value There are five types of forecast data, and each output parameter is output according to 15 timestamp dimensions. Therefore, the forecast sequence output by the time series model is It can be expressed as:

[0063]

[0064] 2. Time series model training phase: (1) Preparation of training data set: A large amount of historical data is collected in the data acquisition and preprocessing stages as the training data set. This data covers the HVAC operation conditions and corresponding heat source flow data under various environmental conditions. This data set is used to train the time series model to ensure that the model can learn rich patterns and rules.

[0065] (2) Training process optimization: Benefiting from the SSM adopted by the time series model, the HVAC model can be given the ability to parallelize training, which greatly improves the training speed of the HVAC model.

[0066] The time series model is deeply trained using the prepared training data set. During the training process, the mean square error (MSE) and mean absolute error (MAE) are selected as loss functions to accurately measure the difference between the control parameters output by the model and the actual optimal control parameters, and the weight and bias parameters of the model are continuously adjusted through back propagation. In order to improve the training efficiency and accelerate the convergence of the model, the Adam optimizer is used. At the same time, regularization terms are added to the loss function to avoid the model being too complex and performing well on the training data but poor generalization ability in actual applications.

[0067] At the beginning of the model training phase, 80% of the data set is selected as the training set and 20% of the data set is selected as the validation set for model training. The parameter configuration of the time series model in the training phase is shown in Table 2: Table 2: Time series model training parameter configuration table

[0068] In an embodiment of the present invention, based on step S3, a possible embodiment is given below to illustrate its specific implementation scheme in a non-limiting manner.

[0069] During the operation of the HVAC system, sensor data and the latest environmental data are collected in real time, and input into the trained time series model at set time intervals. Based on its powerful time series analysis capabilities, the model generates corresponding control parameters according to the input real-time data, and the HVAC control system adjusts the operating status according to these parameters. For example, when the time series model predicts that the indoor temperature will rise due to the increase in outdoor temperature, increased solar radiation intensity and an increase in the number of people, it can accurately adjust the various parameters of the heating mode in advance, including lowering the indoor temperature set point, appropriately adjusting the fan exhaust duct wind speed, reducing the compressor frequency, etc., to maintain a comfortable indoor temperature and humidity.

[0070] In an embodiment of the present invention, based on step S4, a possible embodiment is given below to illustrate its specific implementation scheme in a non-limiting manner.

[0071] Continuously monitor the basic parameters of indoor environmental comfort indicators such as temperature and humidity, and accurately record the energy consumption of the HVAC system, including the power consumption of different equipment such as compressors and fans and the overall energy consumption trend. Feed the above data to the time series model, which further optimizes the control parameters based on the feedback information, realizes adaptive control, and continuously improves the control effect and energy-saving efficiency. For example, if users in a certain area of ​​the building report that the local temperature is too low, the model will analyze the factors that cause the problem, such as insufficient air speed at the air outlet in the area or too low temperature setting value, and adjust the control parameters accordingly. Through this adaptive control mechanism, the energy-saving efficiency is continuously improved while ensuring the comfort of personnel, and the long-term stable and optimized operation of the HVAC system is achieved.

[0072] In addition, it should be noted that after the execution effect of this method was verified by experiments, it was proved that the time series model can continuously learn effective features as the length of the input sequence increases, and longer time series data is more conducive to optimizing the model effect. The method of the present invention accurately senses the subtle changes in the indoor and outdoor environment, adjusts the operating parameters of the HVAC in advance and accurately, and effectively maintains a stable and comfortable indoor environment. Whether it is the temperature and humidity fluctuations caused by the frequent entry and exit of personnel, or the impact of drastic changes in outdoor weather, they can be properly dealt with.

[0073] like Figure 3 As shown, the following is an embodiment of a HVAC heating prediction system provided by an embodiment of the present disclosure. The system and the HVAC heating prediction methods of the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiments of the HVAC heating prediction system, reference can be made to the embodiments of the above-mentioned HVAC heating prediction methods.

[0074] A heating prediction system for HVAC includes: a data acquisition and preprocessing module 1, a model building and training module 2, a reasoning prediction module 3 and a feedback optimization module 4.

[0075] Data collection and preprocessing module 1, used to collect adjustable parameters of the HVAC heating system, external environment data, HVAC heating system related data and indoor environment data, perform data cleaning and normalization on the collected data, and generate a time series data set; Model building and training module 2 is used to build a time series model based on the Mamba architecture and state space model, and use the preprocessed time series data set to train the time series model so that the time series model can learn the time series regularity information under the normal operation mode of the HVAC system; The inference prediction module 3 is used to use the trained time series model to dynamically call the relevant parameters of the HVAC for fast recursive reasoning, predict the heating capacity of the HVAC system, and generate corresponding control parameters based on the prediction results.

[0076] The feedback optimization module 4 is used to monitor the execution results of the control parameters, record the relevant energy consumption information, and feed back the energy consumption information to the time series model to optimize the control parameters.

[0077] The HVAC heating prediction system provided in this embodiment utilizes the advantages of the state space model (SSM) in the Mamba architecture in processing time series data, and can accurately capture the time series characteristics of the HVAC system during operation, thereby achieving precise control of the heating capacity of the HVAC system.

[0078] Figure 4 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0079] The HVAC heating prediction method provided in the embodiment of the present application can be applied to electronic devices. Those skilled in the art will appreciate that the electronic device structure involved in the embodiment of the present invention does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown, or combine certain components, or arrange different components. In an embodiment of the present invention, the electronic device includes but is not limited to a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0080] The electronic device may include a processor, an external memory interface, an internal memory, a universal serial bus (USB) interface, a charging management module, a power management module, a battery, a wireless communication module, an audio module, a speaker, a microphone, a sensor module, buttons, a camera, a display, and a SIM card interface, etc.

[0081] The processor may include one or more processing units, for example, the processor may include a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Among them, different processing units may be independent devices or integrated into one or more processors.

[0082] The processor can be the nerve center and command center of the electronic device. The controller can generate an operation control signal according to the instruction operation code and timing signal to complete the control of fetching and executing instructions.

[0083] A memory may also be provided in the processor for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. The memory may store instructions or data that the processor has just used or is cyclically used. If the processor needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor, and thus improves system efficiency.

[0084] The external memory interface can be used to connect an external memory card, such as a MicroSD card, to expand the storage capacity of the electronic device. The external memory card communicates with the processor through the external memory interface to implement data storage functions. For example, files such as music and videos can be saved in the external memory card.

[0085] The internal memory can be used to store computer executable program codes, which include instructions. The processor executes various functional applications and data processing of the electronic device by running the instructions stored in the internal memory. The internal memory may include a program storage area and a data storage area. The internal memory may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc.

[0086] The wireless communication function of an electronic device can be realized through an antenna, a wireless communication module, a modem processor, and a baseband processor.

[0087] The wireless communication module can provide wireless communication solutions for electronic devices, including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared technology (IR), etc.

[0088] Electronic devices can implement audio functions, etc. through audio modules, speakers, receivers, microphones, headphone jacks, and application processors.

[0089] Electronic devices can achieve shooting functions through ISP, camera, video codec, GPU, display and application processor.

[0090] Electronic devices can achieve display functions through GPU, display screen and application processor.

[0091] The GPU is a microprocessor for image processing that connects the display screen and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor may include one or more GPUs that execute program instructions to generate or change display information.

[0092] The display screen is used to display images, videos, etc. The display screen includes a display panel.

[0093] The above-mentioned electronic device implements the HVAC heating prediction method of the present application by utilizing the advantages of the state space model (SSM) in the Mamba architecture in processing time series data, and can accurately capture the time series characteristics of the HVAC system during operation, thereby achieving the beneficial effect of accurately controlling the heating capacity of the HVAC system.

[0094] The storage medium provided in the present application stores a program product capable of implementing a HVAC heating prediction method.

[0095] The HVAC heating prediction method includes: collecting adjustable parameters of the HVAC heating system, external environmental data, HVAC heating system related data and indoor environmental data, cleaning and normalizing the collected data, and generating a time series data set; building a time series model based on the Mamba architecture and state space model, and using the preprocessed time series data set to train the time series model so that the time series model can learn the time series regularity information under the normal operating mode of the HVAC system; using the trained time series model, dynamically calling the relevant parameters of the HVAC for fast recursive reasoning, predicting the heating capacity of the HVAC system, and generating corresponding control parameters according to the prediction results.

[0096] In some possible implementations, the HVAC heating prediction method of the present disclosure may be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps described in the above “Exemplary Method” section of this specification according to various exemplary implementations of the present disclosure.

[0097] The storage medium of the present disclosure can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0098] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting heating supply of HVAC, characterized in that: include: Collect the adjustable parameters of the HVAC system, external environment data, HVAC system related data and indoor environment data, clean and normalize the collected data, and generate a time series data set; A time series model is built based on the Mamba architecture and state space model, and the preprocessed time series data set is used to train the time series model so that the time series model can learn the time series regularity information under the normal operation mode of the HVAC system. By using the trained time series model, the relevant parameters of HVAC are dynamically called for fast recursive reasoning to predict the heating capacity of the HVAC system and generate corresponding control parameters based on the prediction results.

2. The HVAC heating prediction method according to claim 1, characterized in that: The method comprises: Monitor the execution results of control parameters, record relevant energy consumption information, and feed the energy consumption information back to the time series model to optimize the control parameters.

3. The HVAC heating prediction method according to claim 1, characterized in that: The adjustable parameters include the primary side opening of the plate heat exchanger, the frequency of the water pump, and the frequency of the terminal fan coil; the external environment data includes the outdoor temperature and humidity sensor data, the floor infrared radiation data and the camera-based number monitoring data; the HVAC heating system related data includes the municipal heating calorimeter data, the plate heat exchanger primary and secondary side supply and return water pressure transmitters, flow meters and temperature sensors collected corresponding data; the indoor environment data includes the temperature data of different locations sensed by the temperature sensor, the air humidity data monitored by the humidity sensor, and the number of people and activity range information detected by the combination of radar modules, infrared sensors and high-resolution cameras.

4. The HVAC heating prediction method according to claim 1, characterized in that: The time series model is constructed based on the Mamba architecture and the state space model, including: Configure the input layer of the time series model and pass the input sequence It is expressed as: in, Defined as specific data 1,17]: ; The hidden layer of the time series model is constructed according to the Mamba architecture, and the expressiveness of the model is enhanced by using the residual links, linear layers, and activation functions of the Transformer. Construct the output layer of the time series model and set the primary side opening of the plate heat exchanger , Pump frequency , fan coil frequency , Indoor temperature prediction value , heat supply forecast value is the output parameter; the output prediction sequence is expressed as: , , where H is the process hidden layer.

5. The HVAC heating prediction method according to claim 4, characterized in that: The hidden layer of the time series model constructed according to the Mamba architecture includes: The state of the hidden layer at time step t It is expressed as: in, , , is a learnable weight matrix; Set up the Mamba module, which includes the linear layer Linear, the convolution layer Conv, the gating mechanism layer σ and the state space model SSM layer. When the HVAC system and the environment state are When , the Mamba module is represented as: Use 12 layers of Mamba module stacking to fully explore the effective features in time series data, and hide the process layer It is expressed as: .

6. The HVAC heating prediction method according to claim 5, characterized in that: The trained time series model is used to dynamically call the relevant parameters of the HVAC system for fast recursive reasoning, predict the heating capacity of the HVAC system, and generate corresponding control parameters according to the prediction results, including: During the operation of the HVAC system, the adjustable parameters of the HVAC system, external environmental data, HVAC system related data and indoor environmental data are collected in real time and input into the time series model at set time intervals; The corresponding control parameters are generated according to the prediction results through the time series model.

7. The HVAC heating prediction method according to claim 2, characterized in that: The monitoring control parameter execution results and recording related energy consumption information, and feeding the energy consumption information back to the time series model to optimize the control parameters, include: Control the HVAC system based on control parameters, monitor indoor environmental comfort indicators, and record the energy consumption and energy consumption trends of the HVAC system; The indoor environmental comfort index, energy consumption of the HVAC system and energy consumption trend are fed back to the time series model to generate optimized control parameters.

8. A heating and air conditioning heating prediction system, characterized in that: The system adopts the HVAC heating prediction method as claimed in any one of claims 1 to 7; The system comprises: The data acquisition and preprocessing module is used to collect the adjustable parameters of the HVAC system, external environment data, HVAC system related data and indoor environment data, clean and normalize the collected data, and generate a time series data set; The model building and training module is used to build a time series model based on the Mamba architecture and state space model, and use the preprocessed time series data set to train the time series model so that the time series model can learn the time series regularity information under the normal operation mode of the HVAC system; The inference prediction module is used to use the trained time series model to dynamically call the relevant parameters of the HVAC for fast recursive reasoning, predict the heating capacity of the HVAC system, and generate corresponding control parameters based on the prediction results.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the HVAC heating prediction method according to any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the HVAC heating prediction method according to any one of claims 1 to 7 are implemented.

Images