Intelligent adjusting system and method for heat transfer of air conditioner

Abstract

According to the air conditioner heat transfer intelligent adjusting system and adjusting method, the heat transfer adjusting system achieving real-time monitoring, data driving and intelligent prediction is adopted, and precise, automatic and flexible application control over air conditioner heat transfer control is achieved. According to the method, the operation parameters of the next time point can be intelligently regulated and controlled in combination with the heat transfer related sequence features and the external features, the regulation and control mode is flexible, the control parameters related to air conditioner heat transfer can be automatically adjusted along with the environment features, and the air conditioner heat transfer strategy is intelligently and flexibly regulated and controlled along with the heat load change rate of the heat transfer system; the technical effects of energy conservation, high efficiency, intelligent regulation and control and data driving optimization are achieved, the user experience is improved, and the operation efficiency and the user satisfaction degree of the air conditioning system can be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent regulation technology, and in particular to an intelligent regulation system and method for heat transfer of an air conditioner, an electronic device and a computer-readable storage medium. Background Art

[0002] The heat transfer system of air conditioners mainly involves the circulation of refrigerants, the operation of compressors, condensers, and evaporators. The traditional heat transfer control method of air conditioners is mainly to implement PID control by the air conditioning system. However, traditional PID control also has the following application defects: Traditional PID control may not respond fast enough in dynamic environments, resulting in temperature fluctuations. PID control is a preset parameter control method that is mechanical and rigid. It cannot intelligently and flexibly control the air conditioning heat transfer strategy according to the heat load change rate of the heat transfer system, resulting in low efficiency. The sensor accuracy is insufficient, resulting in inaccurate control; In addition, when the ambient temperature changes greatly or the temperature difference between indoor and outdoor is large, the system may not be able to adjust in time, resulting in a decrease in energy efficiency; The system relies too much on the preset program parameters of the PID controller and cannot adapt to changes, resulting in low efficiency under different loads. It also cannot comprehensively consider the overall load conditions and external parameters of the heat transfer system to dynamically control the heat transfer, so the control method is relatively rigid. Summary of the invention

[0003] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions: In one aspect, a method for intelligently adjusting heat transfer of an air conditioner is provided, the method being implemented by an electronic device, the method comprising: S1. According to the preset air-conditioning working time point, monitor the operating power consumption parameters of the air-conditioning heat transfer system: the heat of the refrigerant heat cycle, the heat load change rate and the compressor power consumption, and the indoor three-dimensional image at the air-conditioning working time point and upload it to the air-conditioning backend server; S2, the air conditioning backend server calculates the indoor occupant density in the indoor three-dimensional image at the air conditioning working time point, and inputs the operating power consumption parameters and indoor occupant density at the air conditioning working time point into the pre-deployed heat transfer intelligent adjustment LSTM model; S3: The heat transfer intelligent adjustment LSTM model predicts the control sequence data features of the air conditioner working time point according to the operating power consumption parameters, and uses the control sequence data features and the indoor occupant density as the input sequence, and predicts again based on the feature correlation between the previous and next time points, and outputs the control sequence data features of the next air conditioner working time point. And the corresponding indoor occupancy density ; S4. Monitor the indoor occupancy density ρ(t+1) at the next air conditioning working time point and determine whether the following conditions are met: , If it is established, when the next air conditioning working time point is reached, the control sequence data feature is analyzed and executed The system heat transfer control parameters in the system are used to automatically adjust the heat transfer of the air conditioner; On the contrary, the operating power consumption parameters of the next air-conditioning working time point and the indoor occupant density ρ(t+1) are used as the input sequence, and the heat transfer intelligent adjustment LSTM model predicts the control sequence data characteristics of the next air-conditioning working time point, and parses and executes.

[0004] Preferably, the calculation method of the operating power consumption parameter includes: (1) Heat of refrigerant heat cycle Evaporator evap heat absorption: , Condenser cond heat release: , in, is the refrigerant mass flow rate, which is positively correlated with the opening of the air conditioner’s electronic expansion valve; is the specific enthalpy value, and its subscripts (evap, out), (evap, in), (cond, out), and (cond, in) represent the evaporator outlet, evaporator inlet, condenser outlet, and condenser inlet, respectively; (2) Heat load change rate set up: , in: , The heat load is calculated as: ; By introducing the time t, the heat load is calculated: The rate of change of heat load over time t: ; in, is the equivalent heat capacity of the room, where is the specific heat capacity of air (1.005 kJ / kg·K), is the room air conditioning outlet temperature, is the evaporator outlet temperature; is the air supply mass flow rate, which is positively correlated with the fan speed and air guide plate angle of the air conditioner; (3) Compressor power consumption set up: , in, The isentropic efficiency of the compressor is 0.65~0.85.

[0005] Preferably, the method for generating the heat transfer intelligent adjustment LSTM model comprises: Collect historical control sequence big data of the heat transfer system of air conditioners: Data(T)=∏Data(t), Data(T+1)=∏Data(t+1), Data(T) and Data(T+1) are the historical regulation sequence data at time point (t) and time point (t+1), respectively; The historical control sequence data = {refrigerant heat cycle heat, heat load change rate, compressor power consumption, system heat transfer control parameters: fan speed, air guide plate angle, compressor frequency and electronic expansion valve}; Collect indoor three-dimensional images at time point (t) and time point (t+1) respectively, and calculate indoor occupant densities at time point (t) and time point (t+1) based on the indoor three-dimensional images: ρ(t) and ρ(t+1); Construct a regulatory sequence data set with X={[Data(Tn),ρ(tn)], ...[Data(T),ρ(t)]} as the input sequence and Y=[ų(T+1),ρ(t+1)] as the output result, where ų(T+1) represents the system thermal transfer control parameter at time point (t+1); Dividing the regulatory sequence data set into a training set and a validation set according to a preset ratio; The training set is input into a pre-built LSTM model, and the model is used to train and learn the regulatory sequence data features of X in the training set and the regulatory sequence data features of the correlation between time point (t) and time point (t+1), so as to build a heat transfer intelligent regulation LSTM model with X as the input sequence and Y as the output result; The prediction performance of the heat transfer intelligent regulation LSTM model is verified using the verification set: If the verification is successful, the heat transfer intelligent adjustment LSTM model is deployed and applied to the air conditioning backend server; If the verification fails, the above steps will be repeated to rebuild the heat transfer intelligent adjustment LSTM model.

[0006] Preferably, the LSTM model adopts a double-layer LSTM architecture, including: Input layer, used for round input X or (X, Y); The first layer of bidirectional LSTM is used to train and learn the regulatory sequence data features of X; The second layer of bidirectional LSTM is used to train and learn the regulatory sequence data features of the previous and next associations in (X, Y); The output layer is used to output the prediction results.

[0007] On the other hand, an air conditioner heat transfer intelligent adjustment system is provided, which is applied to an air conditioner heat transfer intelligent adjustment method, and the system includes: The data acquisition module is used to monitor the operating power consumption parameters of the air conditioning heat transfer system according to the preset air conditioning working time point: the refrigerant heat cycle heat, the heat load change rate and the compressor power consumption, and the indoor three-dimensional image at the air conditioning working time point and upload it to the air conditioning backend server; A data processing module, configured to calculate, by the air conditioning backend server, the indoor occupant density in the indoor three-dimensional image at the air conditioning working time point, and input the operating power consumption parameters and indoor occupant density at the air conditioning working time point into a pre-deployed heat transfer intelligent adjustment LSTM model; The heat transfer intelligent prediction module is used to predict the control sequence data features of the air conditioner working time point according to the operating power consumption parameters through the heat transfer intelligent adjustment LSTM model, and use the control sequence data features and the indoor occupant density as the input sequence, and predict again based on the feature correlation between the previous and next time points, and output the control sequence data features of the next air conditioner working time point. And the corresponding indoor occupancy density ; The heat transfer response module is used to monitor the indoor occupancy density ρ(t+1) at the next air conditioning working time point and determine whether the following conditions are met: , If it is established, when the next air conditioning working time point is reached, the control sequence data feature is analyzed and executed The system heat transfer control parameters in the system are used to automatically adjust the heat transfer of the air conditioner; On the contrary, the operating power consumption parameters of the next air-conditioning working time point and the indoor occupant density ρ(t+1) are used as the input sequence, and the heat transfer intelligent adjustment LSTM model predicts the control sequence data characteristics of the next air-conditioning working time point, and parses and executes.

[0008] On the other hand, an electronic device is provided, comprising: a processor; a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned intelligent air conditioning heat transfer adjustment methods is implemented.

[0009] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned air conditioning heat transfer intelligent adjustment methods.

[0010] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: The present invention adopts a heat transfer regulation system through real-time monitoring, data drive and intelligent prediction to achieve precise, automated and flexible application control of air conditioning heat transfer control. It can combine the sequence characteristics and external characteristics related to heat transfer to intelligently regulate the operating parameters at the next time point. The regulation method is flexible, and it can automatically adjust the control parameters related to air conditioning heat transfer according to environmental characteristics, and intelligently and flexibly regulate the air conditioning heat transfer strategy according to the heat load change rate of the heat transfer system, so as to achieve energy-saving and high-efficiency, intelligent regulation, data-driven optimization and improve user experience, and can significantly improve the operating efficiency and user satisfaction of the air conditioning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 This is a flow chart of an intelligent adjustment method for heat transfer of an air conditioner provided by an embodiment of the present invention; Figure 2 This is a schematic diagram of the LSTM model structure; Figure 3 It is a schematic diagram of the architecture structure of a double-layer LSTM provided by an embodiment of the present invention; Figure 4 This is a block diagram of an air conditioning heat transfer intelligent adjustment system provided by an embodiment of the present invention; Figure 5 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0014] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0015] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0016] In the embodiments of the present invention, sometimes a subscript such as W1 may be mistakenly written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are consistent.

[0017] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0018] The embodiment of the present invention provides an intelligent adjustment method for heat transfer of an air conditioner, which can be implemented by an electronic device, which can be a terminal or a server. Figure 1 The process flow of the method for intelligently adjusting heat transfer of air conditioner shown in the figure may include the following steps: S1. According to the preset air-conditioning working time point, monitor the operating power consumption parameters of the air-conditioning heat transfer system: the heat of the refrigerant heat cycle, the heat load change rate and the compressor power consumption, and the indoor three-dimensional image at the air-conditioning working time point and upload it to the air-conditioning backend server; S2, the air conditioning backend server calculates the indoor occupant density in the indoor three-dimensional image at the air conditioning working time point, and inputs the operating power consumption parameters and indoor occupant density at the air conditioning working time point into the pre-deployed heat transfer intelligent adjustment LSTM model; S3: The heat transfer intelligent adjustment LSTM model predicts the control sequence data features of the air conditioner working time point according to the operating power consumption parameters, and uses the control sequence data features and the indoor occupant density as the input sequence, and predicts again based on the feature correlation between the previous and next time points, and outputs the control sequence data features of the next air conditioner working time point. And the corresponding indoor occupancy density ; S4. Monitor the indoor occupancy density ρ(t+1) at the next air conditioning working time point and determine whether the following conditions are met: , If it is established, when the next air conditioning working time point is reached, the control sequence data feature is analyzed and executed The system heat transfer control parameters in the system are used to automatically adjust the heat transfer of the air conditioner; On the contrary, the operating power consumption parameters of the next air-conditioning working time point and the indoor occupant density ρ(t+1) are used as the input sequence, and the heat transfer intelligent adjustment LSTM model predicts the control sequence data characteristics of the next air-conditioning working time point, and parses and executes.

[0019] The present invention adopts a heat transfer regulation system through real-time monitoring, data drive and intelligent prediction to achieve precise, automated and flexible application control of air conditioning heat transfer control. It can combine the sequence characteristics and external characteristics related to heat transfer to intelligently regulate the operating parameters at the next time point. The regulation method is flexible, and it can automatically adjust the control parameters related to air conditioning heat transfer according to environmental characteristics, and intelligently and flexibly regulate the air conditioning heat transfer strategy according to the heat load change rate of the heat transfer system, so as to achieve energy-saving and high-efficiency, intelligent regulation, data-driven optimization and improve user experience, and can significantly improve the operating efficiency and user satisfaction of the air conditioning system.

[0020] The present invention mainly comprises the following steps: S1. Monitor the operating parameters of the air conditioning heat transfer system 1) Data collection: Operating power consumption parameters, including: Refrigerant heat cycle heat; Heat load change rate; Compressor power consumption.

[0021] The above data is collected by system sensors and uploaded to the host computer, such as the temperature of the condenser inlet and outlet or the amount of heat released / absorbed (each device, such as the condenser, can collect the data in advance and upload it to the host computer after calculation by its own MCU processor). The data interaction and communication control between each collection terminal and the host computer (air conditioning background server) can be understood in combination with the communication mode of the existing air conditioning system.

[0022] The default parameter values ​​of the system air conditioner, such as specific enthalpy values, can be found in the air conditioner manual, etc. The system generally saves the corresponding original factory data by default.

[0023] As for indoor 3D images: Use a 3D camera or depth sensor (such as Kinect) to capture indoor 3D images in real time.

[0024] Time point: Collect data according to the preset air conditioner working time point (such as every 15 minutes). The air conditioner working time point is set by the user or factory set.

[0025] 2) Data upload: The collected operating power consumption parameters and three-dimensional images are uploaded to the air conditioner backend server.

[0026] Use efficient communication protocols such as MQTT or HTTP to ensure real-time and reliable data transmission.

[0027] S2. Calculate the indoor occupancy density and input it into the LSTM model 1) Calculation of indoor personnel density: Use computer vision algorithms (such as YOLO or OpenCV) to analyze 3D images and detect the number and location of people in the room.

[0028] Calculate the density of people (number of people per unit area): number of people detected / indoor area (pre-entered by the system).

[0029] 2) Data preprocessing: The operating power consumption parameters and personnel density are standardized to ensure that the input data are in the same dimension.

[0030] Organize the data in time series to form an input feature matrix.

[0031] 3) Input LSTM model: The operating power consumption parameters and personnel density are input into the pre-deployed heat transfer intelligent adjustment LSTM model.

[0032] The input features include: Operation power consumption parameters: [refrigerant heat cycle heat, heat load change rate, compressor power consumption], personnel density ρ(t); S3. Predicting regulatory sequence data features 1) LSTM model prediction: The LSTM model is used to predict the characteristics of the control sequence data based on the operating power consumption parameters at the current time point.

[0033] The control sequence data features include: target temperature, compressor frequency, fan speed, electronic expansion valve opening, etc.

[0034] 2) Combined with population density prediction: The predicted regulatory sequence data features and personnel density are used as input sequences and input into the LSTM model again.

[0035] Based on the feature correlation of previous and next time points, predict the regulatory sequence data features at the next time point and population density .

[0036] S4. Monitoring and execution control 1) Monitor the density of people at the next time point: At the next air conditioning working time, real-time monitoring of indoor occupancy density .

[0037] 2) Conditional judgment: Determine whether the conditions are met: , If established: Parse and implement predicted regulatory sequence data features , to automatically adjust the heat transfer of air conditioners; If not true: the operating power consumption parameters and actual personnel density at the next time point As input, re-predict regulatory sequence data features and execute.

[0038] Therefore, the auxiliary prediction technology of the air-conditioning control strategy for the next time point based on the LSTM model of the above-mentioned present invention can make intelligent control based on the user's air-conditioning usage habits and home environmental parameters, thereby avoiding the user's manual activation of the air-conditioning and time-consuming adjustment of the corresponding air-conditioning operating parameters, greatly improving its control flexibility.

[0039] The present invention has the following application advantages: 1. Precise control Dynamic Adaptation: By real-time monitoring of operating parameters and occupancy density, the air conditioning operation strategy is dynamically adjusted to ensure optimal heat transfer efficiency.

[0040] Personalized adjustment: Combined with the prediction of occupant density, precise control based on actual indoor needs can be achieved to improve comfort.

[0041] 2. Energy saving and high efficiency Optimize energy consumption: Avoid unnecessary energy waste and reduce compressor power consumption and thermal cycle heat by predicting and regulating sequence data characteristics.

[0042] Intelligent adjustment: Automatically adjust air conditioning operating parameters according to the heat load change rate and occupancy density to improve the overall energy efficiency of the system.

[0043] 3. Intelligence and Automation Intelligent prediction: Use the LSTM model to capture the complex dependencies of time series data and achieve high-precision prediction.

[0044] Automated execution: Automatically execute control strategies based on prediction results, reduce manual intervention, and improve the intelligence level of the system.

[0045] 4. Data-driven Multi-source data fusion: Combines operating power consumption parameters and 3D image data to fully reflect the operating status of the air-conditioning system and the indoor environment.

[0046] Continuous optimization: By continuously accumulating data, we can optimize the prediction accuracy of the LSTM model and improve system performance.

[0047] 5. Improve user experience Quick response: Real-time monitoring and prediction ensure that the air conditioning system responds quickly to user needs and improves the user experience.

[0048] Comfortable environment: A control strategy based on occupancy density ensures that the indoor environment is always comfortable.

[0049] Preferably, the calculation method of the operating power consumption parameter includes: (1) Heat of refrigerant heat cycle Evaporator evap heat absorption: , Condenser cond heat release: , in, is the refrigerant mass flow rate, which is positively correlated with the opening of the air conditioner’s electronic expansion valve; is the specific enthalpy value, and its subscripts (evap, out), (evap, in), (cond, out), and (cond, in) represent the evaporator outlet, evaporator inlet, condenser outlet, and condenser inlet, respectively; The values ​​of each inlet and outlet can be determined based on the original factory data. The above heat release and heat absorption calculation formulas can be calculated in combination with the corresponding parameters.

[0050] Here, we need to adjust the opening of the electronic expansion valve of the air conditioner in direct proportion to the calculation results, because it is related to the mass flow rate of the refrigerant. In the air conditioning system, there is a certain positive correlation between the mass flow rate of the refrigerant and the opening of the electronic expansion valve EEV. This relationship can be described by a proportional coefficient, usually called the flow coefficient or the opening-flow coefficient. The following is a detailed description and mathematical expression of this relationship: 1) Relationship description Electronic expansion valve opening: The opening of the electronic expansion valve (usually expressed as a percentage, 0% is fully closed and 100% is fully open) determines the flow area of ​​the refrigerant through the valve.

[0051] Refrigerant mass flow rate: The mass flow rate of the refrigerant (unit: kg / s) is positively correlated with the opening of the electronic expansion valve, that is, the larger the opening, the greater the flow rate.

[0052] 2) Mathematical expression The relationship between the refrigerant mass flow rate and the opening of the electronic expansion valve can be expressed as: , The meanings of the letters / characters from left to right are: refrigerant mass flow rate (kg / s), electronic expansion valve opening (%), proportional coefficient (kg / (s·%)) (indicates the refrigerant mass flow rate corresponding to unit opening).

[0053] The determination of the proportionality coefficient k is: The specific value of the proportionality coefficient k depends on the following factors: Refrigerant type: Different refrigerants have different densities and flow characteristics, which affect the flow coefficient; Valve characteristics: The design of the electronic expansion valve (such as flow area, valve core shape) will affect the relationship between flow and opening; System operating conditions: System parameters such as evaporation temperature, condensation temperature, and pressure difference will also affect the flow coefficient.

[0054] The value of k can be fitted by experimentally measuring the refrigerant mass flow rate at different openings. For example: In opening =50%, the measured flow rate .

[0055] but .

[0056] For example: In actual applications, the relationship between flow and opening may not be completely linear, especially when the opening is small or large. Therefore, the following correction method can be used: Piecewise linear model: The opening range is divided into multiple intervals, and a different k value is used for each interval.

[0057] .

[0058] For example: Assume that the relationship between the opening of the electronic expansion valve and the refrigerant mass flow rate of an air-conditioning system is as follows:

[0059] Then the proportionality coefficient is: .

[0060] (2) Heat load change rate set up: , in: , The heat load is calculated as: ; By introducing the time t, the heat load is calculated: The rate of change of heat load over time t: ; in, is the equivalent heat capacity of the room, where is the specific heat capacity of air (1.005 kJ / kg·K), is the room air conditioning outlet temperature, is the evaporator outlet temperature; is the air supply mass flow rate, which is positively correlated with the fan speed and air guide plate angle of the air conditioner; The above-mentioned inlet and outlet temperatures can also be monitored and collected by sensors deployed in various places, and then transmitted to the host computer (backend) after processing. Other parameters such as air supply quality can refer to the factory settings.

[0061] The air supply mass flow rate is positively correlated with the fan speed and air guide plate angle of the air conditioner. The air supply mass flow rate can be regulated by controlling the fan speed or air guide plate angle of the air conditioner.

[0062] In the air conditioning heat transfer system, the air supply mass flow rate is one of the key parameters that affect the heat load change rate. The air supply mass flow rate is positively correlated with the fan speed and the air guide angle. By adjusting the fan speed or the air guide angle, the air supply mass flow rate can be controlled, thereby affecting the heat load change rate.

[0063] 1. Factors affecting air supply mass flow 1.1 Fan speed Relationship description: The higher the fan speed, the greater the air mass flow rate.

[0064] 1.2 Wind deflector angle Relationship description: The larger the angle of the air guide plate, the larger the flow area of ​​the air supply channel and the greater the air supply mass flow rate.

[0065] Air supply mass flow It can be expressed as a function of the fan speed N and the deflector angle θ: , k is the proportionality coefficient, which is related to the fan characteristics, duct design and air density; The mathematical expression of heat load change rate and air supply mass flow rate is constructed as follows: , It is the difference between the room air conditioning outlet temperature and the evaporator outlet temperature.

[0066] By adjusting the fan speed or the angle of the air guide plate, the air mass flow rate can be controlled, thereby adjusting the heat load change rate.

[0067] Adjust the fan speed.

[0068] Method: Adjust the fan motor speed through the frequency converter.

[0069] advantage: Precisely control the air supply mass flow rate.

[0070] The energy saving effect is significant, especially under partial load conditions.

[0071] Adjust the angle of the air deflector.

[0072] Method: Adjust the angle of the air guide plate by a stepper motor or a servo motor.

[0073] advantage: Rapidly change air supply direction and mass flow.

[0074] Suitable for local heat load regulation.

[0075] (3) Compressor power consumption set up: , in, The isentropic efficiency of the compressor is 0.65~0.85.

[0076] The power consumption of the compressor can be calculated by the system, combining the above-mentioned factory setting parameters.

[0077] The present invention adopts a bidirectional LSTM LSTM model for model training.

[0078] The LSTM model adopts a double-layer LSTM architecture, including: Input layer, used for round input X or (X, Y); The first layer of bidirectional LSTM is used to train and learn the regulatory sequence data features of X; The second layer of bidirectional LSTM is used to train and learn the regulatory sequence data features of the previous and next associations in (X, Y); The output layer is used to output the prediction results.

[0079] The model structure of bidirectional LSTM is as follows Figure 2As shown, using the LSTM model, it is possible to predict and output the sequence data features of future related time points based on the environmental data sequence data of past time points. Therefore, the present invention uses the historical control sequence big data of the air conditioning heat transfer system to perform LSTM training, thereby constructing a heat transfer intelligent regulation LSTM model that can identify the heat transfer control sequence data features of the current point and predict future time points.

[0080] The structure of the LSTM model and the principles of model training and application are well-known technologies and will not be described in detail here.

[0081] like Figure 3 As shown, the present invention proposes a double-layer LSTM architecture, which includes 2 layers of LSTM: The first layer of bidirectional LSTM is used to train and learn the regulatory sequence data features of X; it mainly memorizes and learns the heat transfer regulatory sequence data features (including the corresponding system control parameters and the matching environmental features) at various time points in history; The second layer of bidirectional LSTM is used to train and learn the related control sequence data features in (X, Y); it mainly memorizes and learns the dependency relationship between the heat transfer control sequence data features at the previous and next time points in the historical control, that is, memorizes and learns the dependency relationship between the control strategies at the two time points (t) and (t+1), so as to facilitate the comprehensive prediction of the air conditioning heat transfer strategy at the next time point based on the heat transfer control sequence data features and / or environmental characteristics (indoor population density) at the previous time (t).

[0082] Therefore, the bidirectional structure captures the dependencies between historical control data at the same time. The bidirectional LSTM model can efficiently capture the time dependencies in the air conditioning heat transfer control sequence, and combine historical data to predict future control parameters, thereby optimizing the air conditioning operation strategy. In practical applications, the model structure and hyperparameters need to be adjusted according to the characteristics of the data, and the prediction accuracy and robustness need to be improved through continuous iteration.

[0083] Preferably, the method for generating the heat transfer intelligent adjustment LSTM model comprises: Collect the historical control sequence big data of the heat transfer system of the air conditioner (composed of the control sequence data recorded by the heat transfer system of the air conditioner at each time point): Data(T)=∏Data(t), Data(T+1)=∏Data(t+1), Data(T) and Data(T+1) are the historical regulation sequence data at time point (t) and time point (t+1), respectively; The historical control sequence data = {refrigerant heat cycle heat, heat load change rate, compressor power consumption, system heat transfer control parameters: fan speed, air guide plate angle, compressor frequency and electronic expansion valve}; Collect indoor three-dimensional images at time point (t) and time point (t+1) respectively, and calculate indoor occupant densities at time point (t) and time point (t+1) based on the indoor three-dimensional images: ρ(t) and ρ(t+1); Construct a regulatory sequence data set with X={[Data(Tn),ρ(tn)], ...[Data(T),ρ(t)]} as the input sequence and Y=[ų(T+1),ρ(t+1)] as the output result, where ų(T+1) represents the system thermal transfer control parameter at time point (t+1); Dividing the regulatory sequence data set into a training set and a validation set according to a preset ratio; The training set is input into a pre-built LSTM model, and the model is used to train and learn the regulatory sequence data features of X in the training set and the regulatory sequence data features of the correlation between time point (t) and time point (t+1), so as to build a heat transfer intelligent regulation LSTM model with X as the input sequence and Y as the output result; The prediction performance of the heat transfer intelligent regulation LSTM model is verified using the verification set: If the verification is successful, the heat transfer intelligent adjustment LSTM model is deployed and applied to the air conditioning backend server; If the verification fails, the above steps will be repeated to rebuild the heat transfer intelligent adjustment LSTM model.

[0084] For data collection at each time point, please combine the previous description to collect, process and upload to the background.

[0085] Construct a regulatory sequence dataset: Input sequence X: X= {[Data(T -n),ρ(t - n)],..., [Data(T),ρ(t)]} Where n is the time window length.

[0086] Output result Y: Y = [ų (T + 1), ρ (t + 1)] Among them, ų(T+1) is the system heat transfer control parameter at time point t+1.

[0087] Dataset division: The data set is divided according to the preset ratio (e.g. 80% training set, 20% validation set). Here, X and (X, Y) are divided into two data sets, which will be divided into training set and validation set respectively, and the two-layer model will be input and trained in turns.

[0088] Constructing a heat transfer intelligent adjustment LSTM model: 1) Model input and output Input: Time series data X, including historical control sequence data and personnel density, is first trained and learned by the first layer of bidirectional LSTM to identify the corresponding control sequence data features at each time and the adapted personnel density; then (X, Y) is input into the second layer of bidirectional LSTM to memorize and learn the dependency relationship between the heat transfer control sequence data features at the previous and next time points in the historical control, so as to train the model recognition characteristics of "automatically identifying and outputting the control sequence data features at the next time point with dependency / association relationship under the control sequence data features at the previous time point"; Output: Prediction result Y, including system heat transfer control parameters and personnel density at future time points.

[0089] The execution process can be simplified as shown in the following code: #Input layer model.add(Bidirectional) LSTM(units=64,return_sequences=True) #The first layer of bidirectional LSTM returns the complete sequence input_shape=(window_size, num_features) model.add(Dropout(8.2))#Prevent overfitting #The second layer of bidirectional LSTM model.add(Bidirectional(LSTM(units=32))) model.add(Dropout(8.2)) # Output layer (predict target parameters) model.add(Dense(units=2))#Assume output target temperature and compressor frequency # Compile the model model.compile(optimizer='adam',loss='msemetrics=['mae']).

[0090] And training data: Use training set X and (X, Y) for training.

[0091] 2) Training parameters Optimizer: Adam; Loss function: mean square error (MSE); Evaluation metric: mean absolute error (MAE); Early stopping method: prevent overfitting.

[0092] For example, using the following parameters: epochs=188, batch_size - 32.

[0093] 3) Verify model performance Validation data: Use validation sets of X and (X, Y) to evaluate model performance separately The following evaluation indicators can be used: Mean square error (MSE); Mean absolute error (MAE); Visual comparison of predicted value and true value. Specific verification can be done by the administrator.

[0094] If verification passes: The model performance meets the requirements and can be deployed directly; If verification fails: Adjust model hyperparameters (such as the number of LSTM units, learning rate, and Dropout rate), or increase the amount of training data or improve the data quality, and retrain the model.

[0095] The model can be integrated into the server, loaded using Python or C++, and a prediction interface can be provided.

[0096] First, the model predicts the characteristics of the regulatory sequence data at the next time point based on the characteristics of the regulatory sequence data and the density of personnel at the current time point. The predicted regulatory sequence data characteristics It contains the control parameters of each device in the air conditioning heat transfer system at the next time point, such as fan speed, etc. However, this part combines the external feature, that is, the number of people in the room at the next time point, to determine whether to apply the previously predicted and outputted If the indoor population density ρ(t+1) at time t+1 satisfies the predicted indoor population density , then the previous prediction output can be executed If the number of people exceeds the limit, it is necessary to combine the actual number of people at the current time t+1 and the operating power consumption parameters at the current time t+1, return to the previous step, and re-predict by the first layer (bidirectional) LSTM of the heat transfer intelligent adjustment LSTM model to avoid the situation where the system control strategy is not adjusted in time due to changes in environmental characteristics and the indoor air-conditioning temperature does not match the actual number of people.

[0097] Please understand the application of the above model in conjunction with the description of the previous steps S3 and S4, which will not be repeated here.

[0098] Figure 4 1 is a block diagram of an air conditioning heat transfer intelligent adjustment system according to an exemplary embodiment, the system is used for an air conditioning heat transfer intelligent adjustment method. Figure 4 The system includes a data acquisition module 401, a data processing module 402, a heat transfer intelligent prediction module 403, and a heat transfer response module 404. Among them: The data acquisition module 401 is used to monitor the operating power consumption parameters of the air conditioning heat transfer system according to the preset air conditioning working time point: the refrigerant heat cycle heat, the heat load change rate and the compressor power consumption, and the indoor three-dimensional image at the air conditioning working time point and upload it to the air conditioning backend server; The data processing module 402 is used to calculate the indoor occupant density in the indoor three-dimensional image at the air conditioner working time point by the air conditioner backend server, and input the operation power consumption parameter and indoor occupant density at the air conditioner working time point into the pre-deployed heat transfer intelligent adjustment LSTM model; The heat transfer intelligent prediction module 403 is used to predict the control sequence data features of the air conditioner working time point according to the operating power consumption parameters through the heat transfer intelligent adjustment LSTM model, and use the control sequence data features and the indoor occupant density as the input sequence, and predict again based on the feature correlation between the previous and next time points, and output the control sequence data features of the next air conditioner working time point. And the corresponding indoor occupancy density ; The heat transfer response module 404 is used to monitor the indoor occupant density ρ(t+1) at the next air conditioning working time point to determine whether the following conditions are met: , If it is established, when the next air conditioning working time point is reached, the control sequence data feature is analyzed and executed The system heat transfer control parameters in the system are used to automatically adjust the heat transfer of the air conditioner; On the contrary, the operating power consumption parameters of the next air-conditioning working time point and the indoor occupant density ρ(t+1) are used as the input sequence, and the heat transfer intelligent adjustment LSTM model predicts the control sequence data characteristics of the next air-conditioning working time point, and parses and executes.

[0099] Please understand the corresponding functions and interactions of the above modules in conjunction with the corresponding steps of the previous method.

[0100] Figure 5 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 5 As shown, the electronic device may include the above Figure 4 The air conditioner heat transfer intelligent adjustment system shown. Optionally, the electronic device 410 may include a first processor 2001.

[0101] Optionally, the electronic device 410 may further include a memory 2002 and a transceiver 2003 .

[0102] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0103] Combine the following Figure 5 The components of the electronic device 410 are described in detail: The first processor 2001 is the control center of the electronic device 410, and may be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPUs), or may be application specific integrated circuits (ASICs), or may be one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (field programmable gate arrays, FPGAs).

[0104] Optionally, the first processor 2001 can execute various functions of the electronic device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0105] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 5 CPU0 and CPU1 are shown in FIG.

[0106] In a specific implementation, as an embodiment, the electronic device 410 may also include multiple processors, such as Figure 5 The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0107] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled to be executed by the first processor 2001. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0108] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and access the first processor 2001 through the interface circuit ( Figure 5 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0109] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0110] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 5 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0111] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 5(not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0112] It should be noted that Figure 5 The structure of the electronic device 410 shown in the figure does not constitute a limitation on the router, and the actual knowledge structure recognition device may include more or fewer components than those shown in the figure, or combine certain components, or arrange the components differently.

[0113] In addition, the technical effects of the electronic device 410 can refer to the technical effects of the intelligent adjustment method for air conditioning heat transfer described in the above method embodiment, and will not be repeated here.

[0114] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0115] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0116] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0117] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0118] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0119] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0120] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0121] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0122] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of the system or unit, which can be electrical, mechanical or other forms.

[0123] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0124] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0125] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0126] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An intelligent adjustment method for heat transfer of air conditioner, characterized in that: The method comprises: S1. According to the preset air-conditioning working time point, monitor the operating power consumption parameters of the air-conditioning heat transfer system: the heat of the refrigerant heat cycle, the heat load change rate and the compressor power consumption, and the indoor three-dimensional image at the air-conditioning working time point and upload it to the air-conditioning backend server; S2, the air conditioning backend server calculates the indoor occupant density in the indoor three-dimensional image at the air conditioning working time point, and inputs the operating power consumption parameters and indoor occupant density at the air conditioning working time point into the pre-deployed heat transfer intelligent adjustment LSTM model; S3: The heat transfer intelligent adjustment LSTM model predicts the control sequence data features of the air conditioner working time point according to the operating power consumption parameters, and uses the control sequence data features and the indoor occupant density as the input sequence, and predicts again based on the feature correlation between the previous and next time points, and outputs the control sequence data features of the next air conditioner working time point. And the corresponding indoor occupancy density ; S4. Monitor the indoor occupancy density ρ(t+1) at the next air conditioning working time point and determine whether the following conditions are met: ， If it is established, when the next air conditioning working time point is reached, the control sequence data feature is analyzed and executed The system heat transfer control parameters in the system are used to automatically adjust the heat transfer of the air conditioner; On the contrary, the operating power consumption parameters of the next air-conditioning working time point and the indoor occupant density ρ(t+1) are used as the input sequence, and the heat transfer intelligent adjustment LSTM model predicts the control sequence data characteristics of the next air-conditioning working time point, and parses and executes.

2. The air conditioning heat transfer intelligent adjustment method according to claim 1, characterized in that: The calculation method of the operating power consumption parameter includes: (1) Heat of refrigerant heat cycle Evaporator evap heat absorption: ， Condenser cond heat release: , in, is the refrigerant mass flow rate, which is positively correlated with the opening of the air conditioner’s electronic expansion valve; is the specific enthalpy value, and its subscripts (evap, out), (evap, in), (cond, out), and (cond, in) represent the evaporator outlet, evaporator inlet, condenser outlet, and condenser inlet, respectively; (2) Heat load change rate set up: ， in: , The heat load is calculated as: ; By introducing the time t, the heat load is calculated: The rate of change of heat load with time t: ； in, is the equivalent heat capacity of the room, where is the specific heat capacity of air (1.005 kJ / kg·K), is the room air conditioning outlet temperature, is the evaporator outlet temperature; is the air supply mass flow rate, which is positively correlated with the fan speed and air guide plate angle of the air conditioner; (3) Compressor power consumption set up: ， in, The isentropic efficiency of the compressor is 0.65~0.

85.

3. The air conditioning heat transfer intelligent adjustment method according to claim 1, characterized in that: The method for generating the heat transfer intelligent adjustment LSTM model comprises: Collect historical control sequence big data of air conditioning heat transfer system: Data(T)=∏Data(t), Data(T+1)=∏Data(t+1), Data(T) and Data(T+1) are the historical regulation sequence data at time point (t) and time point (t+1), respectively; The historical control sequence data = {refrigerant heat cycle heat, heat load change rate, compressor power consumption, system heat transfer control parameters: fan speed, air guide plate angle, compressor frequency and electronic expansion valve}; Collect indoor three-dimensional images at time point (t) and time point (t+1) respectively, and calculate indoor occupant densities at time point (t) and time point (t+1) based on the indoor three-dimensional images: ρ(t) and ρ(t+1); Construct a regulatory sequence data set with X={[Data(Tn),ρ(tn)], ...[Data(T),ρ(t)]} as the input sequence and Y=[ų(T+1),ρ(t+1)] as the output result, where ų(T+1) represents the system thermal transfer control parameter at time point (t+1); Dividing the regulatory sequence data set into a training set and a validation set according to a preset ratio; The training set is input into a pre-built LSTM model, and the model is used to train and learn the regulatory sequence data features of X in the training set and the regulatory sequence data features of the correlation between time point (t) and time point (t+1), so as to build a heat transfer intelligent regulation LSTM model with X as the input sequence and Y as the output result; The prediction performance of the heat transfer intelligent regulation LSTM model is verified using the verification set: If the verification is successful, the heat transfer intelligent adjustment LSTM model is deployed and applied to the air conditioning backend server; If the verification fails, the above steps will be repeated to rebuild the heat transfer intelligent adjustment LSTM model.

4. The air conditioning heat transfer intelligent adjustment method according to claim 3 is characterized in that: The LSTM model adopts a double-layer LSTM architecture, including: Input layer, used for round input X or (X, Y); The first layer of bidirectional LSTM is used to train and learn the regulatory sequence data features of X; The second layer of bidirectional LSTM is used to train and learn the regulatory sequence data features of the previous and next associations in (X, Y); The output layer is used to output the prediction results.

5. An air conditioning heat transfer intelligent adjustment system, the air conditioning heat transfer intelligent adjustment system is used to implement the air conditioning heat transfer intelligent adjustment method according to any one of claims 1 to 4, characterized in that: The system comprises: The data acquisition module is used to monitor the operating power consumption parameters of the air conditioning heat transfer system according to the preset air conditioning working time point: the refrigerant heat cycle heat, the heat load change rate and the compressor power consumption, and the indoor three-dimensional image at the air conditioning working time point and upload it to the air conditioning backend server; A data processing module, configured to calculate, by the air conditioning backend server, the indoor occupant density in the indoor three-dimensional image at the air conditioning working time point, and input the operating power consumption parameters and indoor occupant density at the air conditioning working time point into a pre-deployed heat transfer intelligent adjustment LSTM model; The heat transfer intelligent prediction module is used to predict the control sequence data features of the air conditioner working time point according to the operating power consumption parameters through the heat transfer intelligent adjustment LSTM model, and use the control sequence data features and the indoor occupant density as the input sequence, and predict again based on the feature correlation between the previous and next time points, and output the control sequence data features of the next air conditioner working time point. And the corresponding indoor occupancy density ; The heat transfer response module is used to monitor the indoor occupancy density ρ(t+1) at the next air conditioning working time point and determine whether the following conditions are met: ， If it is established, when the next air conditioning working time point is reached, the control sequence data feature is analyzed and executed The system heat transfer control parameters in the system are used to automatically adjust the heat transfer of the air conditioner; On the contrary, the operating power consumption parameters of the next air-conditioning working time point and the indoor occupant density ρ(t+1) are used as the input sequence, and the heat transfer intelligent adjustment LSTM model predicts the control sequence data characteristics of the next air-conditioning working time point, and parses and executes.

6. An electronic device, characterized in that: The electronic device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 4.

Images