Self-adaptive energy-saving control method and system for air conditioning unit

Abstract

The invention relates to the technical field of air conditioner control, in particular to a self-adaptive energy-saving control method and system for an air conditioning unit, and the method comprises the following steps: monitoring indoor and outdoor environment temperature, humidity and illumination intensity in real time; identifying user behavior modes including rest, motion and leaving states; monitoring the operation states of the air conditioning unit in real time, wherein the operation states comprise compressor operation frequency, fan rotation speed and refrigerant pressure parameters; predicting an environment change trend and a user use demand in a period of time in the future; applying a multi-objective optimization method to generate an optimal control strategy comprehensively considering the energy-saving effect, the user comfort and the equipment stability; and according to the optimal control strategy, the operation parameters of the air conditioning unit are adjusted in real time, the environment change trend and the user use requirements in a period of time in the future can be accurately predicted, intelligent monitoring and prediction are achieved, and the energy-saving effect of the air conditioner is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning control, and in particular to an adaptive energy-saving control method and system for an air conditioning unit. Background Art

[0002] With the rapid development of modern society, air conditioners have become an indispensable device in people's daily life and work. However, the widespread use of air conditioners has brought about huge energy consumption problems, especially in hot summer and cold winter, the large-scale use of air conditioners leads to a huge waste of electricity resources. Therefore, improving the energy efficiency of air conditioners and achieving energy saving and consumption reduction have become technical problems that need to be solved urgently.

[0003] At present, most air conditioning units on the market use fixed temperature settings and operating modes, ignoring the dynamic changes of various factors such as ambient temperature, humidity, light intensity, user activities, etc. In addition, the existing air conditioning control system lacks intelligent analysis of user behavior patterns and cannot make adaptive adjustments based on the specific needs of users, resulting in energy waste and reduced user comfort.

[0004] Traditional energy-saving control methods mainly focus on hardware optimization and simple logic control, such as adjusting compressor frequency and fan speed through frequency conversion technology, but these methods are inadequate in dealing with complex and changing environments and user needs. The lack of systematic intelligent monitoring, prediction and optimization methods is a major pain point in current air-conditioning energy-saving technology. Summary of the invention

[0005] Based on the above purpose, the present invention provides an adaptive energy-saving control method and system for an air-conditioning unit.

[0006] An adaptive energy-saving control method for an air-conditioning unit comprises the following steps:

[0007] S100, dynamic environmental perception step: real-time monitoring of indoor and outdoor environmental temperature, humidity and light intensity through high-precision infrared temperature sensors, humidity sensors and light intensity sensors;

[0008] S200, user behavior recognition: Through image recognition technology and deep learning algorithms, it analyzes video data collected by indoor cameras and identifies user behavior patterns, including resting, exercising, and leaving.

[0009] S300, equipment status monitoring: real-time monitoring of the operating status of the air-conditioning unit, including compressor operating frequency, fan speed, and refrigerant pressure parameters;

[0010] S400, demand forecasting: Based on dynamic environmental perception data, user behavior patterns and device status data, using time series analysis, predict environmental change trends and user usage needs in the future;

[0011] S500, multi-objective optimization control: Based on the demand forecast results, multi-objective optimization methods are applied to generate the optimal control strategy that comprehensively considers energy saving effects, user comfort and equipment stability;

[0012] S600, intelligent regulation: According to the optimal control strategy, the operating parameters of the air-conditioning unit, including temperature setting value, operating mode, and wind speed, are adjusted in real time, and the operation of the compressor and fan is precisely controlled through frequency conversion technology.

[0013] Optionally, the user behavior identification in S200 specifically includes:

[0014] S210, arranging cameras indoors to ensure that the cameras cover indoor activity areas, including the living room and bedroom, and the cameras collect indoor video data in real time and transmit the data to a preset data processing unit via a wireless network;

[0015] S220, the data processing unit pre-processes the raw video data transmitted by the camera, including video frame extraction and image denoising;

[0016] S230, user behavior recognition model training: based on the labeled user behavior video data, the user behavior recognition model is trained using a deep learning algorithm so that the behavior recognition model can recognize different behavior patterns;

[0017] S240, real-time behavior recognition:

[0018] Resting state recognition: Identify the user's resting state by detecting the user's posture (such as sitting, lying) and activity frequency (such as not moving for a long time) in the video;

[0019] Motion state recognition: Identify the user's motion state by detecting the user's actions (such as walking, exercising) and activity frequency (such as continuous movement) in the video;

[0020] Away status recognition: Determine whether the user has left the room by detecting the user's appearance and disappearance in the video, and confirm it by combining the situation of no activity for a long time;

[0021] S250, behavior pattern analysis and recording: the data processing unit records the user's behavior pattern in different time periods according to the recognition result, and stores it in the behavior database for subsequent analysis and demand prediction.

[0022] Optionally, the deep learning algorithm includes a convolutional neural network (CNN) and a convolutional neural network (CNN). Based on the labeled user behavior video data, the convolutional neural network (CNN) is used to extract spatial features, and then combined with the recurrent neural network (RNN) to process time series data, and the user behavior recognition model is trained to enable the model to recognize different behavior patterns;

[0023] Convolutional neural network CNN is used to extract spatial features in images. The calculation of the convolution layer is:

[0024] (f*g)(i,j)=Σ m Σ n (fm,n) g(im,jn), where f is the convolution kernel, g is the input image, and (i,j) is the position of the output image;

[0025] Recurrent neural network RNN ​​is used to process time series data and analyze the temporal relationship between frames. RNN is represented as:

[0026] h t =tanh(W ih x t +W hh h t-1 +b h );

[0027] y t =W ho h t +b o ;

[0028] Among them, h t is the hidden state, x t is the input, y t is the output, W and b are the weight and bias.

[0029] Optionally, the demand forecasting in S400 specifically includes:

[0030] S410, data preprocessing: normalizing the collected environment dynamic perception data, user behavior pattern data, and device status data to eliminate dimensional differences in the data;

[0031] S420, time series feature extraction: extracting features from the environment dynamic perception data, user behavior pattern data, and device status data, as follows:

[0032] Moving average: Use the moving average method to smooth time series data, eliminate short-term fluctuations, and extract long-term trends;

[0033] Autoregressive model: Use autoregressive model to model time series data and identify autocorrelation in the data;

[0034] Seasonal decomposition: Use seasonal decomposition methods to decompose the time series into trend, seasonal and residual components;

[0035] S430, perform demand forecasting:

[0036] Trend prediction: Use regression model to predict the extracted trend components:

[0037] T t+k =β 0 +β 1 t+β 2 t 2 +…+β m t m , where T t+k is the trend forecast value at the next k moments, β i is the regression coefficient, t is the time variable, and m is the order of the regression model;

[0038] Seasonal forecasting: Based on the seasonal pattern of historical data, the seasonal decomposition results are used to predict the future seasonal components. The seasonal components are cyclical, and the data of the same historical period are used for prediction: S t+k =S t+k-n , where S t+k is the seasonal forecast value at k moments in the future, and n is the cycle length;

[0039] Residual prediction: Apply the mean method to the residual components for prediction. The residuals fluctuate randomly and have a mean of zero: R t+k =0, where R t+k is the residual prediction value at the future k time points;

[0040] S440, comprehensive forecast: Combine the forecast results of trend, seasonality and residual components to obtain the comprehensive forecast value of future environmental change trends and user usage needs:

[0041] Y t+k =T t+k +S t+k +R t+k ,According to the prediction results, the control strategy of the air-conditioning unit is dynamically adjusted.

[0042] Optionally, the multi-objective optimization control in S500 specifically includes:

[0043] S510, objective function definition: define the objective function of multi-objective optimization, including energy saving effect f 1 、User comfort 2 and equipment stability 3 ;

[0044] S520, weight coefficient setting: according to user preferences and actual needs, set the weight coefficients α, β and γ of each objective function: Objective = αf 1 +βf 2 +γf 3 , where α, β and γ are the weight coefficients of energy saving effect, user comfort and equipment stability, respectively, and their sum is 1;

[0045] S530, constraint definition: define the constraint conditions for multi-objective optimization, including the operating range of the air-conditioning equipment and the temperature setting range of the user:

[0046] Compressor operating frequency range: f min ≤f compressor ≤f max ;

[0047] Indoor temperature range: T min ≤T indoor ≤T max ;

[0048] S540, solving the optimal control strategy based on the particle swarm optimization algorithm;

[0049] S550 uses the particle swarm optimization algorithm to solve the optimal control strategy that comprehensively considers energy-saving effects, user comfort and equipment stability. The optimal control strategy is applied to the air-conditioning unit, and the control unit adjusts the operating parameters of the air-conditioning in real time to achieve adaptive energy-saving control.

[0050] Optionally, the objective function is expressed as follows:

[0051] Energy saving effect, maximize the energy efficiency ratio EER of the air conditioner, the energy efficiency ratio is expressed as:

[0052] The objective function is expressed as:

[0053] Maximizef 1 = EER;

[0054] User comfort: Minimize the difference between indoor temperature and set temperature. Comfort is expressed as: Discomfort = |T indoor -T set |, the objective function is expressed as:

[0055] Minimize 2 =|T indoor -T set |;

[0056] Equipment stability: Minimize the frequent start and stop times and operation fluctuations of the equipment. Stability is expressed as: The objective function is expressed as:

[0057]

[0058] Optionally, the step of solving the optimal control strategy based on the particle swarm optimization algorithm in S540 specifically includes:

[0059] Initialize the particle swarm, where each particle represents a control strategy, including temperature setting value, operation mode and wind speed adjustment value;

[0060] The optimal solution is searched by iteratively updating the position and velocity of the particles.

[0061] Optionally, the particle swarm optimization algorithm further includes evaluating the fitness of each particle and calculating the comprehensive fitness according to the weight coefficient of the objective function: Fitness(i)=αf 1 (i)+βf 2 (i)+γf 3 (i), where f 1 (i) f 2 (i) and f 3 (i) are the energy saving effect, user comfort and equipment stability index of particle i respectively. For each particle, update its historical optimal position p i (t) and the global optimal position g(t) to ensure that the particle swarm converges to the optimal solution.

[0062] An air conditioning unit adaptive energy-saving control system, used to implement the above-mentioned air conditioning unit adaptive energy-saving control method, includes the following modules:

[0063] Environmental monitoring module: including infrared temperature sensor, humidity sensor and light intensity sensor, real-time monitoring of indoor and outdoor environmental temperature, humidity and light intensity;

[0064] User behavior recognition module: It includes indoor cameras and image recognition units. It uses image recognition and deep learning algorithms to analyze video data collected by cameras and recognize user behavior patterns, including rest, exercise, and away status.

[0065] Equipment status monitoring module: used to monitor the operating status of the air-conditioning unit in real time, including compressor operating frequency, fan speed and refrigerant pressure, and record the historical data of air-conditioning operation;

[0066] Data processing module: pre-processes, extracts features and performs time series analysis on environmental monitoring data, user behavior data and device status data to predict environmental change trends and user usage needs in the future;

[0067] Multi-objective optimization module: Based on the demand forecast results, the multi-objective optimization algorithm is applied to generate the optimal control strategy that comprehensively considers energy saving effect, user comfort and equipment stability;

[0068] Control execution module: According to the optimal control strategy, adjust the operating parameters of the air-conditioning unit in real time and control the operation of the compressor and fan.

[0069] Beneficial effects of the present invention:

[0070] The present invention uses high-precision infrared temperature sensors, humidity sensors and light intensity sensors to monitor indoor and outdoor environmental temperature, humidity and light intensity in real time, combines image recognition technology and deep learning algorithms to analyze user behavior patterns, including rest, exercise and away states, thereby achieving comprehensive perception of the environment and user needs. Through time series analysis, feature extraction and demand prediction are performed on environmental dynamic perception data, user behavior patterns and device status data, and it can accurately predict environmental change trends and user usage needs in the future. Intelligent monitoring and prediction improve the energy-saving effect of air conditioning and reduce unnecessary energy consumption.

[0071] The present invention, based on the demand forecast result, applies a multi-objective optimization method to generate an optimal control strategy that comprehensively considers energy-saving effects, user comfort and equipment stability, reduces the energy consumption of the air conditioner by maximizing the energy efficiency ratio (EER); ensures the user's comfort by minimizing the difference between the indoor temperature and the set temperature; extends the service life of the equipment by minimizing the number of frequent starts and stops of the equipment and operating fluctuations; utilizes multi-objective optimization, combines the user's preference to set the weight coefficient, generates the optimal control strategy, and adjusts the temperature setting value, operating mode and wind speed of the air conditioner in real time to achieve a balance between energy saving, comfort and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0073] Figure 1 A schematic diagram of a flow chart of an energy-saving control method according to an embodiment of the present invention;

[0074] Figure 2 The figure is a schematic diagram of the demand forecasting process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0075] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0076] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0077] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0078] like Figure 1-2 As shown, an adaptive energy-saving control method for an air-conditioning unit comprises the following steps:

[0079] S100, dynamic environmental perception step: real-time monitoring of indoor and outdoor environmental temperature, humidity and light intensity through high-precision infrared temperature sensors, humidity sensors and light intensity (light intensity) sensors;

[0080] Multiple high-precision infrared temperature sensors are arranged indoors and outdoors respectively. The infrared temperature sensors can obtain temperature data at different locations indoors and outdoors in real time through non-contact measurement.

[0081] Humidity sensor monitoring: Install high-precision humidity sensors at the air inlet and exhaust outlet of the air-conditioning unit to monitor the humidity changes of the incoming and outgoing air in real time. Combined with humidity sensors arranged in different indoor and outdoor areas, humidity data of the indoor and outdoor environments can be obtained.

[0082] Light intensity sensor monitoring: Install light intensity sensors at key locations indoors and outdoors to monitor changes in ambient light intensity in real time, especially focusing on areas with large light changes such as windows and balconies.

[0083] The preset data processing unit receives real-time data from infrared temperature sensors, humidity sensors and light intensity sensors, integrates multi-source data, eliminates noise and redundant information, and ensures high accuracy and reliability of the data.

[0084] S200, user behavior recognition: Through image recognition technology and deep learning algorithms, it analyzes video data collected by indoor cameras and identifies user behavior patterns, including resting, exercising, and leaving.

[0085] S300, equipment status monitoring: real-time monitoring of the operating status of the air-conditioning unit, including compressor operating frequency, fan speed, and refrigerant pressure parameters;

[0086] Compressor operating frequency monitoring: Install current sensors and voltage sensors on the compressor motor of the air-conditioning unit. By detecting the current and voltage of the motor, the operating frequency of the compressor is calculated. The calculation formula is: Where f is the operating frequency of the compressor, P is the power of the motor, V is the voltage, and I is the current. is the power factor.

[0087] Fan speed monitoring: A speed sensor is installed on the fan motor. The fan speed is detected by a photoelectric sensor or a Hall effect sensor. The signal output by the speed sensor is recorded and monitored in real time by the data processing unit. The calculation formula is: Where N is the fan speed (rpm), f is the motor operating frequency, and p is the number of motor pole pairs.

[0088] Refrigerant pressure monitoring: Install high-precision pressure sensors at key points in the refrigeration circuit to monitor the refrigerant pressure in real time. The pressure sensor converts the detected pressure signal into an electrical signal and transmits it to the data processing unit for recording and analysis. The relationship between the output signal of the pressure sensor and the refrigerant pressure is expressed as: P = k·V s +b, where P is the refrigerant pressure, V s is the sensor output voltage, k and b are the calibration coefficients of the sensor.

[0089] S400, demand forecasting: Based on dynamic environmental perception data, user behavior patterns and device status data, using time series analysis, predict environmental change trends and user usage needs in the future;

[0090] S500, multi-objective optimization control: Based on the demand forecast results, multi-objective optimization methods are applied to generate the optimal control strategy that comprehensively considers energy saving effects, user comfort and equipment stability;

[0091] S600, intelligent regulation: According to the optimal control strategy, the operating parameters of the air-conditioning unit, including temperature setting value, operating mode, and wind speed, are adjusted in real time, and the operation of the compressor and fan is precisely controlled through frequency conversion technology.

[0092] The user behavior identification in S200 specifically includes:

[0093] S210, arranging cameras indoors to ensure that the cameras cover indoor activity areas, including the living room and bedroom, and the cameras collect indoor video data in real time and transmit the data to a preset data processing unit via a wireless network;

[0094] S220, the data processing unit pre-processes the raw video data transmitted by the camera, including video frame extraction and image denoising, so as to improve the accuracy and efficiency of image recognition, and uses a Gaussian filter to smooth the image and remove noise in the image. The Gaussian filter is expressed as: Where σ is the standard deviation of the Gaussian distribution, (x, y) is the coordinate of the pixel;

[0095] S230, user behavior recognition model training: based on the labeled user behavior video data, the user behavior recognition model is trained using a deep learning algorithm so that the behavior recognition model can recognize different behavior patterns;

[0096] S240, real-time behavior recognition:

[0097] Resting state recognition: Identify the user's resting state by detecting the user's posture (such as sitting, lying) and activity frequency (such as not moving for a long time) in the video;

[0098] Motion state recognition: Identify the user's motion state by detecting the user's actions (such as walking, exercising) and activity frequency (such as continuous movement) in the video;

[0099] Away status recognition: Determine whether the user has left the room by detecting the user's appearance and disappearance in the video, and confirm it by combining the situation of no activity for a long time;

[0100] S250, behavior pattern analysis and recording: the data processing unit records the user's behavior pattern in different time periods according to the recognition result, and stores it in the behavior database for subsequent analysis and demand prediction.

[0101] Deep learning algorithms include convolutional neural networks (CNNs) and convolutional neural networks (CNNs). Based on labeled user behavior video data, convolutional neural networks (CNNs) are used to extract spatial features, and then combined with recurrent neural networks (RNNs) to process time series data to train user behavior recognition models, so that the models can recognize different behavior patterns.

[0102] Convolutional neural network CNN is used to extract spatial features in images. The calculation of the convolution layer is:

[0103] (f*g)(i,j)=Σ m Σ n (fm,n) g(im,jn), where f is the convolution kernel, g is the input image, and (i,j) is the position of the output image;

[0104] Recurrent neural network RNN ​​is used to process time series data and analyze the temporal relationship between frames. RNN is represented as:

[0105] h t=tanh(W ih x t +W hh h t-1 +b h );

[0106] y t =W ho h t +b o ;

[0107] Among them, h t is the hidden state, x t is the input, y t is the output, W and b are the weight and bias.

[0108] Real-time behavior recognition is as follows:

[0109] Video frame sequence input: The preprocessed continuous video frames are input into the user behavior recognition model. The model first extracts the spatial features of each frame through CNN, and then inputs the feature sequence into RNN.

[0110] Resting state recognition: RNN is used to analyze the user's posture changes and activity frequency. If it is detected that the user maintains the same posture for a long time, such as sitting or lying down, it is recognized as a resting state.

[0111] Motion state recognition: The user's movements and activity frequencies are analyzed through RNN. If the user is detected to have obvious movements or actions in multiple consecutive frames, such as walking or exercising, it is recognized as a motion state.

[0112] Away state recognition: RNN is used to analyze the appearance and disappearance of users in the video frame sequence. If it is detected that the user suddenly disappears within a certain period of time and does not return for a long time, it is identified as away state.

[0113] Demand forecasting in S400 specifically includes:

[0114] S410, data preprocessing: normalizing the collected environment dynamic perception data, user behavior pattern data, and device status data to eliminate dimensional differences in the data;

[0115] S420, time series feature extraction: extracting features from the environment dynamic perception data, user behavior pattern data, and device status data, as follows:

[0116] Moving average: Use the moving average method to smooth time series data, eliminate short-term fluctuations, and extract long-term trends. The moving average is expressed as:

[0117] Among them, MA t is the moving average at time t, xt-i is the data at time ti, and n is the window size of the moving average. For example, for the operating temperature data of the air conditioner, the daily temperature data can be smoothed by the moving average method to obtain a more stable temperature change trend. Then x t-i It is the temperature data at time ti;

[0118] Autoregressive model: Use the autoregressive model to model time series data and identify the autocorrelation relationship in the data. The autoregressive model is expressed as:

[0119] Among them, x t is the data at time t, c is a constant term, is the autoregressive coefficient, p is the model order, ∈ t is a white noise error term; for example, by analyzing humidity data over a period of time through an autoregressive model, the future humidity change trend is predicted, x t It is the humidity data at time t;

[0120] Seasonal decomposition: Use seasonal decomposition methods to decompose time series into trend, seasonal and residual components. For example, decompose one year's temperature data into trend (long-term temperature change), seasonality (such as temperature change patterns in summer and winter) and residual (short-term fluctuations). The decomposition is expressed as: Y(t) = T(t) + S(t) + R(t), where Y(t) is the time series data, T(t) is the trend component, S(t) is the seasonal component, and R(t) is the residual component.

[0121] S430, perform demand forecasting:

[0122] Trend prediction: Use regression models to predict the extracted trend components. For example, use linear regression models to predict the temperature change trend over a period of time in the future:

[0123] T t+k =β 0 +β 1 t+β 2 t 2 +…+β m t m , where T t+k is the trend forecast value at the next k moments, β i is the regression coefficient, t is the time variable, and m is the order of the regression model;

[0124] Seasonal forecasting: Based on the seasonal pattern of historical data, the seasonal decomposition results are used to predict the future seasonal components. The seasonal components are cyclical, and the data of the same historical period are used for prediction: S t+k =S t+k-n , where St+k is the seasonal forecast value at k moments in the future, and n is the cycle length;

[0125] Residual prediction: Apply the mean method to the residual components for prediction. The residuals fluctuate randomly and have a mean of zero: R t+k =0, where R t+k is the residual prediction value at the future k time points;

[0126] S440, comprehensive forecast: Combine the forecast results of trend, seasonality and residual components to obtain the comprehensive forecast value of future environmental change trends and user usage needs:

[0127] Y t+k =T t+k +S t+k +R t+k ,According to the prediction results, the control strategy of the air-conditioning unit is dynamically adjusted.

[0128] The multi-objective optimization control in S500 specifically includes:

[0129] S510, objective function definition: define the objective function of multi-objective optimization, including energy saving effect f 1 、User comfort 2 and equipment stability 3 ;

[0130] S520, weight coefficient setting: according to user preferences and actual needs, set the weight coefficients α, β and γ of each objective function: Objective = αf 1 +βf 2 +γf 3 , where α, β and γ are the weight coefficients of energy saving effect, user comfort and equipment stability, respectively, and their sum is 1;

[0131] S530, constraint definition: define the constraint conditions for multi-objective optimization, including the operating range of the air-conditioning equipment and the temperature setting range of the user:

[0132] Compressor operating frequency range: f min ≤f compressor ≤f max ;

[0133] Indoor temperature range: T min ≤T indoor ≤T max ;

[0134] S540, solving the optimal control strategy based on the particle swarm optimization algorithm;

[0135] S550 uses the particle swarm optimization algorithm to solve the optimal control strategy that comprehensively considers energy-saving effects, user comfort, and equipment stability. The optimal control strategy is applied to the air-conditioning unit, and the control unit adjusts the operating parameters of the air-conditioning in real time to achieve adaptive energy-saving control:

[0136] Temperature setting value adjustment: According to the optimal strategy, set the appropriate indoor temperature to meet user comfort and energy saving needs.

[0137] Operation mode adjustment: Based on demand forecast results, select cooling, heating, dehumidification or ventilation mode to optimize energy consumption and user experience.

[0138] Wind speed adjustment: Dynamically adjust wind speed based on indoor and outdoor temperatures and user activities to ensure even air flow and a suitable indoor environment.

[0139] The objective function is expressed as follows:

[0140] Energy saving effect, maximize the energy efficiency ratio EER of the air conditioner, the energy efficiency ratio is expressed as:

[0141] The objective function is expressed as:

[0142] Maximizef 1 = EER;

[0143] User comfort: Minimize the difference between indoor temperature and set temperature. Comfort is expressed as: Discomfort = |T indoor -T set |, the objective function is expressed as:

[0144] Minimize 2 =|T indoor -T set |;

[0145] Equipment stability: Minimize the frequent start and stop times and operation fluctuations of the equipment. Stability is expressed as: The objective function is expressed as:

[0146]

[0147] The particle swarm optimization algorithm in S540 is used to solve the optimal control strategy, including:

[0148] Initialize the particle swarm, where each particle represents a control strategy, including temperature setting value, operation mode and wind speed adjustment value;

[0149] Search for the optimal solution by iteratively updating the particle's position and velocity;

[0150] The formula for updating the particle's velocity and position is:

[0151] v i (t+1)=ωv i (t)+c 1 r 1 (p i (t)-x i (t))+c 2 r 2 (g(t)-x i (t));

[0152] x i (t+1)=x i (t)+v i (t+1);

[0153] Among them, v i (t) and x i (t) are the velocity and position of particle i at time t, p i (t) is the historical optimal position of particle i, g(t) is the global optimal position, ω is the property weight, c 1 and c 2 is the learning factor, r 1 and r 2 Is a random number.

[0154] The particle swarm optimization algorithm also includes evaluating the fitness of each particle and calculating the comprehensive fitness according to the weight coefficient of the objective function: Fitness(i) = a 1 g(i)+βf 2 (i)+γf 3 (i), where f 1 (i) f 2 (i) and f 3 (i) are the energy saving effect, user comfort and equipment stability index of particle i respectively. For each particle, update its historical optimal position p i (t) and the global optimal position g(t) to ensure that the particle swarm converges to the optimal solution.

[0155] An air conditioning unit adaptive energy-saving control system, used to implement the above-mentioned air conditioning unit adaptive energy-saving control method, includes the following modules:

[0156] Environmental monitoring module: including infrared temperature sensor, humidity sensor and light intensity sensor, real-time monitoring of indoor and outdoor environmental temperature, humidity and light intensity;

[0157] User behavior recognition module: It includes indoor cameras and image recognition units. It uses image recognition and deep learning algorithms to analyze video data collected by cameras and recognize user behavior patterns, including rest, exercise, and away status.

[0158] Equipment status monitoring module: used to monitor the operating status of the air-conditioning unit in real time, including compressor operating frequency, fan speed and refrigerant pressure, and record the historical data of air-conditioning operation;

[0159] Data processing module: pre-processes, extracts features and performs time series analysis on environmental monitoring data, user behavior data and device status data to predict environmental change trends and user usage needs in the future;

[0160] Multi-objective optimization module: Based on the demand forecast results, the multi-objective optimization algorithm is applied to generate the optimal control strategy that comprehensively considers energy saving effect, user comfort and equipment stability;

[0161] Control execution module: According to the optimal control strategy, adjust the operating parameters of the air-conditioning unit in real time and control the operation of the compressor and fan.

[0162] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0163] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An adaptive energy-saving control method for an air-conditioning unit, characterized in that: The following steps are involved: S100, dynamic environment perception step: real-time monitoring of indoor and outdoor environment temperature, humidity and light intensity through infrared temperature sensors, humidity sensors and light intensity sensors; S200, user behavior recognition: Through image recognition technology and deep learning algorithms, it analyzes video data collected by indoor cameras and identifies user behavior patterns, including resting, exercising, and leaving. S300, equipment status monitoring: real-time monitoring of the operating status of the air-conditioning unit, including compressor operating frequency, fan speed, and refrigerant pressure parameters; S400, demand forecasting: Based on dynamic environmental perception data, user behavior patterns and device status data, using time series analysis, predict environmental change trends and user usage needs in the future; S500, multi-objective optimization control: Based on the demand forecast results, multi-objective optimization methods are applied to generate the optimal control strategy that comprehensively considers energy saving effects, user comfort and equipment stability; S600, intelligent regulation: According to the optimal control strategy, the operating parameters of the air-conditioning unit are adjusted in real time, including the temperature setting value, operating mode, and wind speed, and the operation of the compressor and fan is controlled through frequency conversion technology.

2. The adaptive energy-saving control method for an air-conditioning unit according to claim 1, characterized in that: The user behavior identification in S200 specifically includes: S210, arranging cameras indoors to ensure that the cameras cover indoor activity areas, including the living room and bedroom, and the cameras collect indoor video data in real time and transmit the data to a preset data processing unit via a wireless network; S220, the data processing unit pre-processes the raw video data transmitted by the camera, including video frame extraction and image denoising; S230, user behavior recognition model training: based on the labeled user behavior video data, the user behavior recognition model is trained using a deep learning algorithm so that the behavior recognition model can recognize different behavior patterns; S240, real-time behavior recognition: Resting state recognition: Identify the user's resting state by detecting the user's posture and activity frequency in the video; Motion state recognition: Identify the user in motion by detecting the user's movements and activity frequency in the video; Away status recognition: Determine whether the user has left the room by detecting the user's appearance and disappearance in the video, and confirm it by combining the situation of no activity for a long time; S250, behavior pattern analysis and recording: the data processing unit records the user's behavior pattern in different time periods according to the recognition result, and stores it in the behavior database for subsequent analysis and demand prediction.

3. The adaptive energy-saving control method for an air-conditioning unit according to claim 2, characterized in that: The deep learning algorithm includes convolutional neural network (CNN), convolutional neural network (CNN), based on the labeled user behavior video data, using convolutional neural network (CNN) to extract spatial features, and then combined with recurrent neural network (RNN) to process time series data, training user behavior recognition model, so that the model can recognize different behavior patterns; Convolutional neural network CNN is used to extract spatial features in images. The calculation of the convolution layer is: (f*g)(i,j)=∑ m ∑ n f(m,n) g(im,jn), where f is the convolution kernel, g is the input image, and (i,j) is the position of the output image; Recurrent neural network RNN ​​is used to process time series data and analyze the temporal relationship between frames. RNN is represented as: h t =tanh(W ih x t +W hh h t-1 +b h ); y t =W ho h t +b o ; Among them, h t is the hidden state, x t is the input, y t is the output, W and b are the weight and bias.

4. The adaptive energy-saving control method for an air-conditioning unit according to claim 1, characterized in that: The demand forecast in S400 specifically includes: S410, data preprocessing: normalizing the collected environment dynamic perception data, user behavior pattern data, and device status data to eliminate dimensional differences in the data; S420, time series feature extraction: extracting features from the environment dynamic perception data, user behavior pattern data, and device status data, as follows: Moving average: Use the moving average method to smooth time series data, eliminate short-term fluctuations, and extract long-term trends; Autoregressive model: Use autoregressive model to model time series data and identify autocorrelation in the data; Seasonal decomposition: Use seasonal decomposition methods to decompose the time series into trend, seasonal and residual components; S430, perform demand forecasting: Trend prediction: Use regression model to predict the extracted trend components: T t+k =β0+β1t+β2t 2 +…+β m t m , where T t+k is the trend forecast value at the next k moments, β i is the regression coefficient, t is the time variable, and m is the order of the regression model; Seasonal forecasting: Based on the seasonal pattern of historical data, the seasonal decomposition results are used to predict the future seasonal components. The seasonal components are cyclical, and the data of the same historical period are used for prediction: S t+k =S t+k-n , where S t+k is the seasonal forecast value at k moments in the future, and n is the cycle length; Residual prediction: Apply the mean method to the residual components for prediction. The residuals fluctuate randomly and have a mean of zero: R t+k =0, where R t+k is the residual prediction value at the future k time points; S440, comprehensive forecast: Combine the forecast results of trend, seasonality and residual components to obtain the comprehensive forecast value of future environmental change trends and user usage needs: Y t+k =T t+k +S t+k +R t+k ,According to the prediction results, the control strategy of the air-conditioning unit is dynamically adjusted.

5. The adaptive energy-saving control method for air-conditioning units according to claim 4, characterized in that: The multi-objective optimization control in S500 specifically includes: S510, objective function definition: define the objective function of multi-objective optimization, including energy saving effect f1, user comfort f2 and equipment stability f3; S520, weight coefficient setting: according to user preferences and actual needs, set the weight coefficients α, β and γ of each objective function: Objective = αf1 + βf2 + γf3, where α, β and γ are the weight coefficients of energy saving effect, user comfort and equipment stability respectively, and the sum is 1; S530, constraint definition: define the constraint conditions for multi-objective optimization, including the operating range of the air-conditioning equipment and the temperature setting range of the user: Compressor operating frequency range: f min ≤f compressor ≤f max ; Indoor temperature range: T min ≤T indoor ≤T max ; S540, solving the optimal control strategy based on the particle swarm optimization algorithm; S550 uses the particle swarm optimization algorithm to solve the optimal control strategy that comprehensively considers energy-saving effects, user comfort and equipment stability. The optimal control strategy is applied to the air-conditioning unit, and the control unit adjusts the operating parameters of the air-conditioning in real time to achieve adaptive energy-saving control.

6. The adaptive energy-saving control method for air-conditioning units according to claim 5, characterized in that: The objective function is expressed as follows: Energy saving effect, maximize the energy efficiency ratio EER of the air conditioner, the energy efficiency ratio is expressed as: The objective function is expressed as: MaximizeF1 = EER; User comfort: Minimize the difference between indoor temperature and set temperature. Comfort is expressed as: Discomfort=|T indoor -T set |, the objective function is expressed as: Minimizef2=|T indoor -T set |; Equipment stability: Minimize the frequent start and stop times and operation fluctuations of the equipment. Stability is expressed as: The objective function is expressed as:

7. The adaptive energy-saving control method for air-conditioning units according to claim 5, characterized in that: The step of solving the optimal control strategy based on the particle swarm optimization algorithm in S540 specifically includes: Initialize the particle swarm, where each particle represents a control strategy, including temperature setting value, operation mode and wind speed adjustment value; The optimal solution is searched by iteratively updating the position and velocity of the particles.

8. The adaptive energy-saving control method for an air-conditioning unit according to claim 7, characterized in that: The particle swarm optimization algorithm also includes evaluating the fitness of each particle and calculating the comprehensive fitness according to the weight coefficient of the objective function: Fitness(i) = αf1(i) + βf2(i) + γf3(i), where f1(i), f2(i and f3(i) are the energy saving effect, user comfort and equipment stability indicators of particle i respectively. For each particle, its historical optimal position p is updated. i (t) and the global optimal position g(t) to ensure that the particle swarm converges to the optimal solution.

9. An air conditioning unit adaptive energy-saving control system, used to implement an air conditioning unit adaptive energy-saving control method according to any one of claims 1 to 8, characterized in that: Includes the following modules: Environmental monitoring module: including infrared temperature sensor, humidity sensor and light intensity sensor, real-time monitoring of indoor and outdoor environmental temperature, humidity and light intensity; User behavior recognition module: It includes indoor cameras and image recognition units. It uses image recognition and deep learning algorithms to analyze video data collected by cameras and recognize user behavior patterns, including rest, exercise, and away status. Equipment status monitoring module: used to monitor the operating status of the air-conditioning unit in real time, including compressor operating frequency, fan speed and refrigerant pressure, and record the historical data of air-conditioning operation; Data processing module: pre-processes, extracts features and performs time series analysis on environmental monitoring data, user behavior data and device status data to predict environmental change trends and user usage needs in the future; Multi-objective optimization module: Based on the demand forecast results, the multi-objective optimization algorithm is applied to generate the optimal control strategy that comprehensively considers energy saving effect, user comfort and equipment stability; Control execution module: According to the optimal control strategy, adjust the operating parameters of the air-conditioning unit in real time and control the operation of the compressor and fan.

Images