Air conditioning equipment

Abstract

The invention provides air conditioning equipment which comprises a detection device and a processing device, and the detection device is configured to obtain air parameters of an air-conditioned room so as to generate a detection signal; the processing device is provided with an analysis unit, a prediction unit, and a generation unit, and the analysis unit is configured to perform set wavelet transform analysis based on a time sequence of the detection signal and decompose the detection signal into an approximate component signal and a detail component signal; the prediction unit is configured to predict the change trend of the approximate component signal to obtain a predicted approximate component signal and predict the change trend of the detail component signal to obtain a predicted detail component signal; the generation unit is configured to generate a prediction detection signal based on the prediction approximation component signal and the prediction detail component signal. Through real-time detection of the air parameters and prediction of the change trend, intelligent prediction and control of the future air state of the air-conditioned room are achieved, the intelligent level of an air conditioning system is improved, and better use experience is provided.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning, and in particular to an air conditioning device. Background Art

[0002] As people's demands for air conditioning equipment increase, they are no longer satisfied with simply regulating indoor temperature. They are increasingly demanding higher standards for air quality parameters such as humidity, carbon dioxide, and inhalable particulate matter. Excessively high or low humidity can cause discomfort, while high carbon dioxide concentrations can cause fatigue, affect sleep quality, and reduce learning and work efficiency. Inhalable particulate matter is a major cause of respiratory diseases. Air quality parameters are closely linked to people's comfort and health.

[0003] Existing air conditioners, air purifiers, and fresh air products have also been gradually equipped with air quality sensor modules to monitor indoor air conditions. However, the energy-saving and comfort functions in these products, based on air quality parameters, often rely on the current air quality parameter values, resulting in delayed control and unsatisfactory functional results. Summary of the Invention

[0004] In response to the problem that the energy-saving and comfort function logic based on air quality parameters in current air-conditioning, fresh air and other products often depends on the current air quality parameter values, the control is relatively lagging, and the functional effects achieved are not satisfactory, the first aspect of the present invention designs and provides an air conditioning device.

[0005] An air conditioning device includes a detection device and a processing device, wherein the detection device is configured to obtain air parameters of an air-conditioned room to generate a detection signal; the processing device has an analysis unit, a prediction unit and a generation unit, wherein the analysis unit is configured to perform a wavelet transform analysis set based on the time series of the detection signal to decompose the detection signal into an approximate component signal and a detail component signal; the prediction unit is configured to respectively predict the changing trend of the approximate component signal to obtain a predicted approximate component signal and predict the changing trend of the detail component signal to obtain a predicted detail component signal; the generation unit is configured to generate a predicted detection signal based on the predicted approximate component signal and the predicted detail component signal.

[0006] In one or more embodiments of the present application, the parsing unit is configured to perform a wavelet transform analysis set based on the time series of the detection signal to perform multi-level decomposition: in each level of decomposition, the approximate component signal of the previous level is further decomposed into an approximate component signal and a detail component signal; the approximate component signal of the last level and the detail component signal of each level are output to the prediction unit; multiple detail component signals represent the noise of the detection device.

[0007] In one or more embodiments of the present application, the parsing unit is configured to perform a wavelet transform analysis set based on the time series of the detection signal to perform at least two levels of decomposition: in each level of decomposition, the approximate component signal of the previous level is further decomposed into an approximate component signal and a detail component signal; the approximate component signal of the last level and the detail component signal of each level are output to the prediction unit; wherein the detail component signal outputted from the first level represents the precision noise of the detection device, and the detail component signals outputted from other levels represent the interference noise acting on the detection device.

[0008] In one or more embodiments of the present application, the processing device also includes: a filtering unit, which is configured to receive the approximate component signal output by the last stage and the detail component signal output by each stage from the analysis unit, and filter the detail component signal corresponding to the precision noise, and then output the other detail component signals and the approximate component signal output by the last stage to the prediction unit.

[0009] In one or more embodiments of the present application, the prediction unit is configured with: a first prediction model and a second prediction model, the first prediction model is configured to perform feature identification of the approximate component signal, and predict the changing trend of the approximate component signal to obtain a predicted approximate component signal; the second prediction model is configured to perform feature identification of the detail component signal, and predict the changing trend of the detail component signal to obtain a predicted detail component signal.

[0010] In one or more embodiments of the present application, the first prediction model is a long short-term memory model.

[0011] In one or more embodiments of the present application, the second prediction model is an ARMA model.

[0012] In one or more embodiments of the present application, the processing device further includes: a storage unit, a query unit and a time period configuration unit; the storage unit is configured to operably store the detection signal time series, the query unit is configured to query the length of the time period covered by the detection signal time series stored in the storage unit according to operation information specified by the user, and the time period configuration unit is configured to generate future predictable time period options based on the time period length, and the future predictable time period options are available for user operation and selection; the number of predictable time periods corresponding to the future predictable time period options is positively correlated with the length of the time period.

[0013] In one or more embodiments of the present application, the prediction unit predicts the change trend of the approximate component signal to obtain a predicted approximate component signal and predicts the change trend of the detail component signal to obtain a predicted detail component signal according to the prediction period selected by the user in the future predictable period option.

[0014] In one or more embodiments of the present application, the generating unit is configured to generate a predicted detection signal based on the predicted approximation component signal and the predicted detail component signal.

[0015] In one or more embodiments of the present application, the processing device further includes: a function configuration unit; the function configuration unit is configured to generate available function options based on the length of the time period, and the available function options are available for user operation and selection; the number of available functions corresponding to the available function options is positively correlated with the length of the time period.

[0016] In one or more embodiments of the present application, the detection device includes a temperature sensor, a humidity sensor, a particulate matter sensor, a gas pollutant sensor and / or a volatile organic compound sensor.

[0017] The present invention realizes intelligent prediction and control of the future air status of the air-conditioned room through real-time detection of air parameters and prediction of change trends, thereby improving the intelligence level of the air-conditioning system and providing a better user experience.

[0018] Other features and advantages of the present invention will become more apparent after reading the detailed description of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 is a schematic block diagram of the structure of an air conditioning device provided by one or more embodiments of the present invention;

[0021] Figure 2 is a schematic block diagram of the structure of a detection device in an air conditioning device provided by one or more embodiments of the present invention;

[0022] Figure 3 is a schematic block diagram of the structure of a processing device in an air conditioning device provided by one or more embodiments of the present invention;

[0023] Figure 4 is a schematic block diagram of the structure of a processing device in an air conditioning device provided by one or more embodiments of the present invention;

[0024] Figure 5is a schematic block diagram of the structure of a gas pollutant sensor in an air conditioning device provided by one or more embodiments of the present invention;

[0025] Figure 6 is a signal flow chart of an air conditioning device provided by one or more embodiments of the present invention;

[0026] Figure 7 is a signal flow chart of an air conditioning device provided by one or more embodiments of the present invention;

[0027] Figure 8 This is an example of a time-frequency distribution diagram of a detection signal in an air conditioning device provided by one or more embodiments of the present invention;

[0028] Figure 9 This is an example of a time-frequency distribution diagram of the approximate component signal CA3 in the air conditioning device provided by one or more embodiments of the present invention;

[0029] Figure 10 This is an example of a time-frequency distribution diagram of a detail component signal CD 1 in an air conditioning device provided by one or more embodiments of the present invention;

[0030] Figure 11 This is an example of a time-frequency distribution diagram of a detail component signal CD2 in an air conditioning device provided by one or more embodiments of the present invention;

[0031] Figure 12 This is an example of a time-frequency distribution diagram of a detail component signal CD3 in an air conditioning device provided by one or more embodiments of the present invention;

[0032] Figure 13 is a schematic block diagram of the structure of a processing device in an air conditioning device provided by one or more embodiments of the present invention;

[0033] Figure 14 This is a schematic diagram of the LSTM algorithm principle;

[0034] Figure 15 A flowchart for prediction using the LSTM model based on historical approximate component signals;

[0035] Figure 16 A diagram showing an example training result of the LSTM model;

[0036] Figure 17 A diagram showing an example training result of the LSTM model;

[0037] Figure 18 is a signal flow chart of an air conditioning device provided by one or more embodiments of the present invention;

[0038] Figure 19 is a schematic block diagram of the structure of a processing device in an air conditioning device provided by one or more embodiments of the present invention;

[0039] Figure 20 is a schematic block diagram of the structure of a processing device in an air conditioning device provided by one or more embodiments of the present invention;

[0040] Figure 21 is an example of a predictable time period option;

[0041] Figure 22 is an example of the available feature options;

[0042] In the figure: 1. Air conditioning equipment; 10. Detection device; 101. Temperature sensor; 102. Humidity sensor; 103. Particulate matter sensor; 104. Gas pollutant sensor; 1041. Carbon dioxide sensor; 1042. Sulfur dioxide sensor; 1043. Nitrogen oxide sensor; 1044. Carbon monoxide sensor; 1045. Ozone concentration sensor; 105. Volatile organic compound sensor; 20. Processing device; 21. Analysis unit; 22. Prediction unit; 23. Generation Unit; 24. Prediction unit; 25. Storage unit; 26. Query unit; 27. Time period configuration unit; 28. Function configuration unit; 301. Processor; 302. Non-volatile memory; 303. Volatile memory; 304. Display device; 305. Operating device; 306. Communication interface; 307. Drive device; 308. Bus; 309. Storage medium; 310. Storage medium; 40. "Comfort control" option; 50. Future predictable time period option; 60. Available function option. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0044] In the description of this application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0045] The terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, features specified as "first" and "second" may explicitly or implicitly include one or more of such features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0046] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood broadly. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0047] In the present invention, unless otherwise expressly specified or limited, a first feature being "above" or "below" a second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature therebetween. Moreover, a first feature being "above," "above," and "above" a second feature includes the first feature being directly above and obliquely above the second feature, or simply means that the first feature is at a higher level than the second feature. A first feature being "below," "below," and "below" a second feature includes the first feature being directly below and obliquely below the second feature, or simply means that the first feature is at a lower level than the second feature.

[0048] The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numerals and reference letters in different examples, and such repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and settings discussed. In addition, the present invention provides examples of various specific processes and materials, but those of ordinary skill in the art will appreciate the application of other processes and the use of other materials.

[0049] The air conditioning equipment provided by the present invention is introduced below. The air conditioning equipment provided by the present invention can improve the air quality and enhance the cleanliness, healthiness and comfort of the air.

[0050] In one or more embodiments of the present application, the air conditioning device 1 may be an air conditioner.

[0051] An air conditioner performs its refrigeration cycle by using a compressor, condenser, throttling element, and evaporator. The refrigeration cycle involves a series of processes involving compression, condensation, expansion, and evaporation to cool or heat the indoor space.

[0052] Low-temperature, low-pressure refrigerant enters the compressor, which compresses it into high-temperature, high-pressure refrigerant gas and discharges the compressed refrigerant gas. The discharged refrigerant gas flows into the condenser. The condenser condenses the compressed refrigerant into a liquid phase, releasing heat into the surrounding environment through the condensation process.

[0053] The throttling element (e.g., an electronic expansion valve) expands the high-temperature, high-pressure liquid refrigerant condensed in the condenser to a low-pressure liquid. The evaporator evaporates the refrigerant expanded in the electronic expansion valve and returns the low-temperature, low-pressure refrigerant gas to the compressor. The evaporator achieves a cooling effect by utilizing the latent heat of evaporation to exchange heat with the material being cooled. Throughout this cycle, the air conditioner regulates the temperature of the indoor space.

[0054] The outdoor unit of the air conditioner refers to a portion of a refrigeration cycle including a compressor and an outdoor heat exchanger, the indoor unit of the air conditioner includes an indoor heat exchanger, and the electronic expansion valve may be provided in the indoor unit or the outdoor unit.

[0055] The indoor heat exchanger and the outdoor heat exchanger function as a condenser or an evaporator. When the indoor heat exchanger functions as a condenser, the air conditioner functions as a heater in heating mode, and when the indoor heat exchanger functions as an evaporator, the air conditioner functions as a cooler in cooling mode.

[0056] In one or more embodiments of the present application, the air conditioning device 1 may be a fresh air device. The fresh air device includes a filter and a fan, which are used to remove particulate matter and harmful gases from the air while introducing fresh air to maintain fresh indoor air. The fresh air device may be a standalone device or integrated into an air conditioner as a functional module of the air conditioner.

[0057] In one or more embodiments of the present application, the air conditioning device 1 may be an air purifier.

[0058] An air purifier is a device used to improve indoor air quality by removing particulate matter, gaseous pollutants and harmful substances in the air, as well as eliminating odors and killing bacteria, making the air fresher and healthier.

[0059] Air purifiers can be equipped with a filtration system to remove particulate matter from the air. The filtration system can be a HEPA (High Efficiency Particulate Air Filter), an activated carbon filter, or the like.

[0060] Air purifiers can also be equipped with ultraviolet lamps to kill bacteria, viruses and other microorganisms in the air.

[0061] An ion generator can also be installed in the air purifier to charge the particles in the air and make them easier to be captured by the filter.

[0062] In one or more embodiments of the present application, the air conditioning device 1 may be a total heat exchanger.

[0063] A heat exchanger can transfer heat between incoming and outgoing air, recovering energy and pre-conditioning the temperature of fresh air entering a building. This reduces the energy required for heating or air conditioning, lowering energy consumption and saving energy costs. A heat exchanger can also adjust humidity where appropriate, helping to maintain indoor air comfort and quality.

[0064] In one or more embodiments of the present application, the air conditioning device 1 may be a dehumidifier.

[0065] In one or more embodiments of the present application, the air conditioning device 1 may be a humidifier.

[0066] In one or more embodiments of the present application, the air conditioning device 1 may be an air circulation fan.

[0067] Air circulation fans are used to promote air movement, helping to evenly distribute airborne pollutants and temperature.

[0068] In one or more embodiments of the present application, the air conditioning equipment 1 may also be a combination of multiple air conditioners, fresh air equipment, air purifiers, full heat exchangers, dehumidifiers, humidifiers, and air circulation fans.

[0069] like Figure 1 As shown, the air conditioning device 1 includes a detection device 10. The detection device 10 is configured to obtain air parameters of the air-conditioned room to generate a detection signal.

[0070] like Figure 2 As shown, in one or more embodiments of the present application, the detection device 10 is a temperature sensor 101, and the air parameter of the air-conditioned room is the air temperature of the air-conditioned room.

[0071] In one or more embodiments of the present application, the detection device 10 is a humidity sensor 102, and the air parameter of the air-conditioned room is the air humidity of the air-conditioned room.

[0072] In one or more embodiments of the present application, the detection device 10 is a particle sensor 103. The particle sensor 103 is used to measure suspended particulate matter in the air. Particles are classified into PM10, PM2.5, etc. based on their size. The air parameter of the air-conditioned room is the concentration of airborne particulate matter in the air-conditioned room.

[0073] like Figure 5 As shown, in one or more embodiments of the present application, the detection device 10 is a gas pollutant sensor 104. The gas pollutant sensor 104 includes a carbon dioxide sensor 1041, a sulfur dioxide sensor 1042, a nitrogen oxide sensor 1043, a carbon monoxide sensor 1044, and an ozone concentration sensor 1045. The air parameter of the air-conditioned room is the concentration of gas pollutants in the air-conditioned room.

[0074] In one or more embodiments of the present application, the detection device 10 is a volatile organic compound (VOC) sensor 105 that measures volatile organic compounds in the air. The air parameter of the air-conditioned room is the concentration of volatile organic compounds in the air-conditioned room.

[0075] In one or more embodiments of the present application, the detection device 10 includes one or more sensors as described above.

[0076] In one or more embodiments of the present application, the number of each type of sensor in the detection device 10 may be one or more.

[0077] like Figure 3 As shown, the air conditioning device 1 further comprises a processing device 20 .

[0078] Figure 4 This is a block diagram showing the hardware structure of the processing device 20. The processing device 20 includes components such as a processor 301, a volatile memory 303, a non-volatile memory 302, a display device 304, an operating device 305, a communication interface 306, and a drive device 307, which are interconnected via a bus 308. The processor 301 can be a dedicated processor 301, a central processing unit 20 (CPU), etc. The processor 301 can access the storage unit 25 to execute instructions or applications stored in the storage unit 25 to implement related functions. The display device is a display device for displaying various information, the operating device is an operating device for receiving various operations, and the drive device is a hardware terminal that interacts with the storage medium. In one or more embodiments of the present application, the storage medium (shown as 310 in the figure) includes a medium that records information optically, electrically, or magnetically, such as a CD-ROM, a floppy disk, or a magneto-optical disk. The storage medium (shown as 309 in the figure) can also be a semiconductor memory that records information electrically, such as a ROM or flash memory.

[0079] The processing device 20 may be a controller of the air conditioning device 1 itself, such as a system on board built based on an MCU.

[0080] The processing device 20 may also be a host computer or a cloud server, or other intelligent mobile terminal. The processing device 20 and the detection device 10 may be connected via a LAN (Local Area Network), a signal line (such as an Ethernet cable, a coaxial cable, an optical fiber, a power line, a serial line, etc.), a wireless signal, LTE, 5G, or other network connection to enable communication.

[0081] The following is the attached Figure 3 , the functional structure of the processing device 20 is explained.

[0082] like Figure 3 and Figure 6 As shown, in one or more embodiments of the present application, the processing device 20 includes a parsing unit 21, a prediction unit 22, and a generation unit 23. Each of these units can be implemented by a processor running a program.

[0083] In one or more embodiments of the present application, the analyzing unit 21 is configured to perform a set wavelet transform analysis based on the time series of the detection signal, and decompose the detection signal into an approximate component signal and a plurality of detail component signals.

[0084] In one or more embodiments of the present application, the prediction unit 22 is configured to predict the variation trend of the approximate component signal to obtain a predicted approximate component signal and predict the variation trend of the detail component signal to obtain a predicted detail component signal.

[0085] In one or more embodiments of the present application, the generating unit 23 is configured to generate a predicted detection signal based on the predicted approximation component signal and the predicted detail component signal.

[0086] The detection device 10, such as the temperature sensor 101, humidity sensor 102, particulate matter sensor 103, gas pollutant sensor 104 and volatile organic compound sensor 105 listed above, outputs a time series of detection results with frequency domain characteristics and time domain characteristics. The frequency domain characteristics refer to the distribution of the signal in the frequency domain, while the time domain characteristics refer to the change pattern of the signal on the time axis.

[0087] Frequency domain features include periodic changes that occur at different time scales. For example, on shorter time scales, such as daytime and nighttime, CO2 concentrations may fluctuate due to human activity, ventilation conditions, and changes in indoor and outdoor temperatures. During the day, there is more indoor and outdoor air exchange, resulting in lower CO2 concentrations. On a daily basis, the detection signal exhibits periodic changes. Time domain features, on the other hand, include trends that change over time.

[0088] The analysis unit 21 is configured to perform a wavelet transform analysis based on the time series of the detection signal. Wavelet transform analysis is a signal processing tool. The frequency characteristics of the detection signal within a specific time period, or the signal characteristics within a time range at a specific frequency can be determined through wavelet transform analysis. Wavelet transform analysis is more suitable for processing non-stationary signals than traditional Fourier transform or other frequency domain methods. In the present invention, the trend of the detection signal may be linear or nonlinear, or may show a repetitive pattern with periodic fluctuations. At the same time, the length of the cycle may be uncertain due to external factors, and may also be affected by external events to form mutations. It is a typical non-stationary signal, and wavelet transform analysis is more suitable for processing the detection signal.

[0089] The analyzing unit 21 decomposes the detection signal through a set wavelet transform analysis, thereby making it easier to distinguish the noise characteristics and the signal characteristics, so that the separated noise and signal can be used more flexibly for subsequent analysis.

[0090] In one or more embodiments of the present application, the analyzing unit 21 decomposes the detection signal into an approximate component signal and multiple detail component signals. The approximate component signal is a valid signal, and the detail component signal is a noise signal.

[0091] The wavelet transform analysis set in the present invention is introduced below.

[0092] The general form of the wavelet basis function in wavelet transform analysis is:

[0093]

[0094] Among them, a represents the scale factor, b represents the translation factor, It is the basic mother wavelet function, which can be used to construct a low-pass filter to extract the approximate component signal, that is, the low-frequency component in the detection signal. By varying the scale factor a, a series of mother wavelet functions of varying scales are generated. These functions are used to construct high-pass filters and extract detail components, or high-frequency components, from the detection signal, thereby enabling multi-scale decomposition and analysis of the detection signal. The translation factor b can localize the wavelet basis functions to characteristic locations within the detection signal, such as signal breakpoints and peaks. This helps capture important signal features within a specific timeframe.

[0095] In one or more embodiments of the present application, in a given wavelet transform analysis, a wavelet basis function is selected from the Db wavelet. The Db wavelet is a family of wavelet basis functions that have good compactness, orthogonality, and multi-resolution properties. The Db wavelet family generates wavelet basis functions of different orders by selecting different parameters, such as Db2, Db4, Db6, etc., which are not limited in this application.

[0096] Since the set wavelet transform analysis can adaptively adjust the decomposition scale according to the characteristics of the detection signal, the analysis unit 21 can decompose the signal into components of different frequencies, and the signal characteristics in different frequency ranges can be better captured, so that the effective information and noise information are effectively separated, and the essence of the detection signal is better understood. After decomposition, the detection information has reduced dimensions to a certain extent. Combined with the decomposition results, on the one hand, it reduces the difficulty of establishing and calculating the subsequent prediction model, and on the other hand, it also helps to establish a more effective prediction model.

[0097] Furthermore, the prediction unit 22 processes the effective information (approximate component signal) and the noise information (detail component signal) separately. On the one hand, it can more accurately capture the changing trend of the signal, thereby obtaining a more accurate prediction result. On the other hand, by separating the noise information from the effective information, it can reduce the impact of noise on the prediction result and improve the accuracy of the prediction.

[0098] Finally, the generation unit 23 integrates the effective information (approximate component signal) and the noise information (detail component signal) to obtain more comprehensive and integrated prediction data, thereby improving the credibility and practicality of the prediction detection signal.

[0099] In one or more embodiments of the present invention, the analysis unit 21 is configured to perform a predetermined wavelet transform analysis based on the time series of the detection signal to perform multi-level decomposition. Specifically, at each level of decomposition, the approximate component signal of the previous level is further decomposed into an approximate component signal and a detail component signal, and the approximate component signal of the last level and the detail component signals of each level are output to the prediction unit 22. The multiple detail component signals represent the noise of the detection device 10.

[0100] Multi-level decomposition can decompose the detection signal into approximate component signals and detail component signals in different frequency bands, which can more accurately distinguish signals of different frequencies, capture the transient characteristics of the signal, and better separate noise information and effective information; more comprehensive information can be obtained through the detail component signals at each level, and by using the approximate component signals of the last level and the detail component signals of each level, better feature extraction capabilities can be ensured in the prediction stage, providing more comprehensive information when reconstructing the prediction signal in the generation stage.

[0101] The analysis unit 21 is configured to perform a predetermined wavelet transform analysis based on the time series of the detection signal, thereby performing at least two levels of decomposition. In each level of decomposition, the approximate component signal of the previous level is further decomposed into an approximate component signal and a detail component signal. The approximate component signal of the last level and the detail component signals of each level are output to the prediction unit 22. The detail component signal output from the first level represents the precision noise of the detection device 10, while the detail component signals output from the remaining levels represent interference noise acting on the detection device 10.

[0102] In one or more embodiments of the present application, the parsing unit 21 is configured to perform a set wavelet transform analysis based on the time series of the detection signal to perform a three-level decomposition, and in each level of decomposition, the approximate component signal of the previous level is decomposed into an approximate component signal and a detail component signal. The approximate component signal of the third level and the detail component signal of each level are output to the prediction unit 22. Figure 7 As shown, in the first level of decomposition, the detection signal is decomposed into an approximation component signal CA1 and a detail component signal CD1. In the second level of decomposition, the approximation component signal CA1 is further decomposed into an approximation component signal CA2 and a detail component signal CD2. In the third level of decomposition, the approximation component signal CA2 is further decomposed into an approximation component signal CA3 and a detail component signal CD3. After the three levels of decomposition, the approximation component signal CA3 reflects the main characteristic parameters of the detection signal, such as periodicity and trend. The detail component signal CD1 represents the precision noise of the detection device 10, and the detail component signals CD1, CD2, and CD3 represent the interference noise acting on the detection device 10.

[0103] Still taking the concentration of carbon dioxide as an example, the interference noise acting on the detection device 10 includes opening / closing windows, people entering and exiting, using the kitchen, etc.

[0104] Taking the classroom as an air-conditioned room as an example, the detection signal is the carbon dioxide concentration. Figure 8 is the time-frequency distribution diagram of the detection signal, Figure 9 is the time-frequency distribution diagram of the approximate component signal CA1, Figure 10 is the time-frequency distribution diagram of the detail component signal CD1, Figure 11is the time-frequency distribution diagram of the detail component signal CD2, Figure 12 is the video distribution diagram of the detail component signal CD3, Figures 8 to 12 The horizontal axis is the time axis, and the vertical axis is the scale axis. The energy of the signal is distributed in two dimensions: time and frequency.

[0105] The detail component signal CD1 output by the first stage represents the precision noise of the detection device 10. The hardware precision noise of the sensor itself is usually random and unpredictable, and will show different patterns in different measurements. Therefore, the precision noise can be regarded as irrelevant noise. In order to reduce or even eliminate irrelevant noise data and improve signal quality, in one or more embodiments of the present application, Figure 13 As shown, the processing device 20 also includes a filtering unit 24, which is configured to receive the approximate component signal output by the last stage and the detail component signal output by each stage from the self-analysis unit 21, and filter the detail component signal corresponding to the precision noise, and then output the other detail component signals and the approximate component signal output by the last stage to the prediction unit 22.

[0106] Continuing with the above example, the filtering unit 24 filters the detail component signal CD1 and then outputs the detail component signals CD2 and CD3 and the approximation component signal CA3 to the prediction unit 22 .

[0107] In one or more embodiments of the present application, the prediction unit 22 is configured with a first prediction model and a second prediction model. The first prediction model is configured to perform feature recognition of the approximate component signal and predict the change trend of the approximate component signal to obtain a predicted approximate component signal. The second prediction model is configured to perform feature recognition of the detail component signal and predict the telephone trend of the detail component signal to obtain a predicted detail component signal. That is, in the prediction unit 22, the approximate component signal and the detail component signal are predicted using independent prediction models respectively, and then the two prediction results are combined. On the one hand, the two prediction models can better capture the characteristics of the two groups of signals. For example, the approximate component signal may have a relatively stable trend, while the detail component signal may have a more complex change pattern. Using independent models can better capture these characteristics. On the other hand, the two prediction models can improve the overall robustness and reduce the impact of outliers or missing values.

[0108] In one or more embodiments of the present application, the first prediction model adopts a long short-term memory model.

[0109] Long short-term memory (LSTM) is a special variant of recurrent neural networks. Figure 14The LSTM algorithm principle includes three gating units: forget gate, input gate and output gate, which control the transmission and update of information through the gating units.

[0110] C(t-1) represents the memory cell at time step t-1, and C(t) represents the memory cell at time step t. These cells are the core components of LSTM. Memory cells store information that needs to be retained during processing. Information updates and transfers are controlled through the forget gate, input gate, and output gate, as well as the tanh function.

[0111] h(t-1) represents the hidden state at time step t-1, and h(t) represents the hidden state at time step t. It is the output of the LSTM model and the information passed during processing. It is obtained by transforming the memory unit (for example, through the output gate and activation function). The hidden state can be used as the output of the model or passed to the next time step.

[0112] x(t) represents the input at time step t, and the input of each time step can be a feature vector.

[0113] σ is the sigmoid activation function, and tanh is the hyperbolic tangent activation function.

[0114] In the LSTM algorithm, the output of the forget gate is first calculated to control which historical information needs to be forgotten. Then, the output of the input gate and the candidate memory unit are calculated to control which new information needs to be remembered. Then, the memory unit is updated according to the output of the forget gate and the input gate. Finally, the output and hidden state of the output gate are calculated to control which information needs to be output. In this way, the flow of information is controlled through the forget gate, input gate and output gate, so that historical information can be remembered or forgotten to achieve prediction.

[0115] The first prediction model performs feature recognition of the approximate component signal and predicts the change trend of the approximate component signal to obtain the predicted approximate component signal, that is, the LSTM model is used to perform prediction based on the historical approximate component signal.

[0116] It includes multiple steps as shown in the figure:

[0117] Step S11: Perform data preprocessing on the historical approximate component signal to define a training set. Data preprocessing on the historical approximate component signal includes denoising, normalization, and standardization. The processed data is divided into a training set and a test set, or into a training set, a test set, and a validation set.

[0118] Step S12: Define the LSTM model, including setting LSTM network parameters.

[0119] Step S13: Training the LSTM model, including: setting the optimizer and loss function, training the LSTM model using the training set, and evaluating the performance of the trained LSTM model using the test set.

[0120] Step S14: Initialize the LSTM network.

[0121] Step S15: Based on the historical approximate component signal, use the trained LSTM model to perform single-step prediction to obtain the predicted approximate component signal.

[0122] Figure 16 and Figure 17 This is an example of the LSTM model training results. After training, the LSTM model has a RMSE of < 0.03 (e.g. Figure 16 As shown), the loss function approaches 0 (as shown Figure 17 shown).

[0123] In one or more embodiments of the present application, the second prediction model is an ARMA model. The ARMA model can predict future detail component signals based on historical detail component signals. The ARMA model can be constructed using methods disclosed in the prior art, including stationarity testing and model order determination, etc., which will not be repeated here.

[0124] The ARMA model is a classic linear time series model with the advantages of simplicity and computational efficiency. For detail component signals (high-frequency signals), the ARMA model provides ideal fitting and forecasting results due to their relatively simple fluctuations. In particular, the ARMA model excels at capturing short-term correlations, which aligns with the data characteristics of detail component signals and makes it an excellent modeling and forecasting tool.

[0125] Decomposing the original signal into detail component signals and approximate component signals and using different models for prediction can reduce the complexity of prediction. The detail component signals and approximate component signals have different characteristics. The LSTM model and the ARMA model can better adapt to their respective characteristics. The predicted approximate component signal and the predicted detail component signal are finally synthesized to give full play to the advantages of each sub-model and improve the overall prediction accuracy. Taking the selection of humidity sensor, carbon dioxide concentration sensor and PM2.5 sensor as an example, the entire architecture is as follows Figure 18 shown.

[0126] In one or more embodiments of the present application, the ARMA model can predict CD2 and CD3 separately.

[0127] Assuming that the prediction result of the LSTM model is PCA3, and the prediction results of the ARMA model are PCD3 and PCD2, the prediction detection signal finally generated by the generation unit 23 can be expressed as PC=PCA3+PCD3+PCD2.

[0128] In one or more embodiments of the present application, the processing device 20 is implemented by a cloud server.

[0129] It is not difficult to understand that, as more historical data is available, the data for a longer period of time in the future can be more accurately predicted, and the air conditioning device 1 can achieve more diverse and flexible functions. In one or more embodiments of the present application, the air conditioning device 1 can be configured to automatically enable corresponding functions based on historical data without manual intervention.

[0130] Specifically, if Figure 19 As shown, the processing device 20 further includes a storage unit 25 , a query unit 26 and a time period configuration unit 27 .

[0131] In one or more embodiments of the present application, the storage unit 25 is configured to operatively store a detection signal time series, such as a carbon dioxide concentration detection signal output by the carbon dioxide sensor 1041. The storage unit 25 can continuously store the detection signal time series, or can store the detection signal time series for a specific time period according to a setting.

[0132] In one or more embodiments of the present application, the query unit 26 is configured to query the length of the time period covered by the detection signal time series stored in the storage unit 25 according to operation information specified by the user.

[0133] In one or more embodiments of the present application, the time period configuration unit 27 is configured to generate a future predictable time period option 50 based on the length of the time period, and the future predictable time period option 50 is available for user operation and selection; the number of predictable time periods corresponding to the future predictable time period option 50 is positively correlated with the length of the time period.

[0134] In one or more embodiments of the present application, the prediction unit 22 predicts the change trend of the approximate component signal to obtain a predicted approximate component signal and the change trend of the detail component signal to obtain a predicted detail component signal according to the prediction period selected by the user in the future predictable period option 50.

[0135] In one or more embodiments of the present application, the generating unit 23 is configured to generate a predicted detection signal based on the predicted approximation component signal and the predicted detail component signal.

[0136] like Figure 20As shown, a selection interface similar to the "Comfort Control" option can be configured on the operation terminal. When the user selects the "Comfort Control" option 40, the user's operation selection is received. The query unit 26, based on the user's specified operation information, queries the storage unit 25 for the length of the time period covered by the detection signal time series. For example, the length of the time period covered is greater than one month but less than one quarter, greater than one quarter but less than two quarters, greater than two quarters but less than one year, or greater than one year. The time period length can also be set in other ways, which are not limited here.

[0137] The time period configuration unit 27 generates a future predictable time period option 50 based on the time period length. The number of predictable time periods corresponding to the future predictable time period option 50 is positively correlated with the time period length.

[0138] Exemplarily, when the coverage time period is longer than one month but not longer than one quarter, the future predictable time period option is one, including, for example, 24-hour intelligent control; when the coverage time period is longer than one quarter but not longer than two quarters, the future predictable time period options are two, including 24-hour intelligent control and 7-day intelligent control; when the coverage time period is longer than two quarters but not longer than one year, the future predictable time period options are three, including 24-hour intelligent control, 7-day intelligent control and 30-day intelligent control; when the coverage time period is longer than one year, the future predictable time period options are four, including 24-hour intelligent control, 7-day intelligent control, 30-day intelligent control and 90-day intelligent control.

[0139] In one or more embodiments of the present application, based on the length of the time period covered by the detection signal time series queried by the query unit 26 and stored in the storage unit 25, the corresponding future predictable time period option is automatically displayed on the operation terminal in the form of a user interface (UI) and selected by the user. The prediction unit 22 performs the prediction of the corresponding time period based on the result of the user's selection. On the one hand, it can save the computing power of the cloud server, and on the other hand, it can prevent excessive prediction of air quality parameters and affect the user experience.

[0140] like Figure 21 As shown, in one or more embodiments of the present application, the processing device 20 further includes a function configuration unit 28. The function configuration unit 28 is configured to generate available function options 60 based on the length of the time period, and the available function options 60 are available for user selection. The number of available functions corresponding to the available function options 60 is positively correlated with the length of the time period.

[0141] For example, when the user selects the "Comfort Control" option 40, the user's operation selection is received, and the query unit 26 queries the storage unit 25 for the length of the time period covered by the detection signal time series based on the operation information specified by the user. For example, the length of the time period covered is greater than one month but less than one quarter, greater than one quarter but less than two quarters, greater than two quarters but less than one year, or greater than one year. The time period length may also be set in other forms, which are not limited here.

[0142] like Figure 22 As shown, the function configuration unit 28 is configured to generate available function options 60 based on the length of the time period, and the number of available functions corresponding to the available function options 60 is positively correlated with the length of the time period.

[0143] For example, when the coverage period is longer than one month but shorter than one quarter, the available function option 60 is one, including automatic on / off control; when the coverage period is longer than one quarter but shorter than two quarters, the available functions are two, including automatic on / off control and output capacity pre-adjustment; when the coverage period is longer than two quarters but shorter than one year, the available functions are three, including automatic on / off control, output capacity pre-adjustment, and holiday control; and when the coverage period is longer than one year, the available functions are four, including automatic on / off control, output capacity pre-adjustment, holiday control, and seasonal control. Available function option 60 may also include other intelligent control functions or quantities, which are not listed here.

[0144] In one or more embodiments of the present application, based on the length of the time period covered by the detection signal time series queried by the query unit 26 and stored in the storage unit 25, the corresponding available function options 60 are automatically displayed on the operation terminal in the form of a user interface (UI) for selection by the user. Alternatively, the available function options 60 may be nested within the predictable time period options in the form of a hierarchical UI.

[0145] For example, when the predictable time period option includes 24-hour intelligent control, the available function options 60 include automatic switch control; when the predictable time period option includes 7-day intelligent control, the available function options 60 include output capacity pre-adjustment and holiday control, etc.

[0146] In the description of the above embodiments, specific features, structures, materials or characteristics may be combined in an appropriate manner in any one or more embodiments or examples.

[0147] The above are only specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. Air conditioning equipment, including: a detection device configured to obtain air parameters of the air-conditioned room to generate a detection signal; It is characterized by further comprising: A processing device comprising: an analyzing unit configured to perform a set wavelet transform analysis based on a time series of the detection signal to decompose the detection signal into an approximate component signal and a detail component signal; a prediction unit configured to respectively predict a change trend of the approximate component signal to obtain a predicted approximate component signal and predict a change trend of the detail component signal to obtain a predicted detail component signal; and A generating unit is configured to generate a predicted detection signal based on the predicted approximation component signal and the predicted detail component signal.

2. The air conditioning equipment according to claim 1, characterized in that The analysis unit is configured to perform a wavelet transform analysis set based on the time series of the detection signal to perform multi-level decomposition: in each level of decomposition, the approximate component signal of the previous level is further decomposed into an approximate component signal and a detail component signal; the approximate component signal of the last level and the detail component signal of each level are output to the prediction unit; multiple detail component signals represent the noise of the detection device.

3. The air conditioning equipment according to claim 1, characterized in that The analysis unit is configured to perform a wavelet transform analysis based on the time series of the detection signal to perform at least two levels of decomposition: in each level of decomposition, the approximate component signal of the previous level is further decomposed into an approximate component signal and a detail component signal; the approximate component signal of the last level and the detail component signal of each level are output to the prediction unit; wherein the detail component signal output by the first level represents the precision noise of the detection device, and the detail component signals output by other levels represent the interference noise acting on the detection device.

4. The air conditioning equipment according to claim 3, characterized in that The processing device further includes: A filtering unit is configured to receive the approximate component signal output by the last stage and the detail component signal output by each stage from the analysis unit, filter the detail component signal corresponding to the precision noise, and then output the other detail component signals and the approximate component signal output by the last stage to the prediction unit.

5. The air conditioning equipment according to claim 4, characterized in that The prediction unit is configured with: a first prediction model configured to perform feature recognition on the approximate component signal and predict a change trend of the approximate component signal to obtain a predicted approximate component signal; and A second prediction model is configured to perform feature recognition on the detail component signal and predict a change trend of the detail component signal to obtain a predicted detail component signal.

6. The air conditioning equipment according to claim 5, characterized in that The first prediction model is a long short-term memory model.

7. The air conditioning equipment according to claim 5, characterized in that The second prediction model is an ARMA model.

8. The air conditioning equipment according to any one of claims 1 to 7, characterized in that: The processing device further includes: a storage unit configured to operably store the detection signal time series; a query unit configured to query the length of a time period covered by the detection signal time series stored in the storage unit according to operation information specified by a user; and a time period configuration unit configured to generate a future predictable time period option based on the time period length, wherein the future predictable time period option is selectable by a user; the number of predictable time periods corresponding to the future predictable time period option is positively correlated with the time period length; The prediction unit predicts the change trend of the approximate component signal to obtain a predicted approximate component signal and the change trend of the detail component signal to obtain a predicted detail component signal according to the prediction period selected by the user in the future predictable period option; The generating unit is configured to generate a predicted detection signal based on the predicted approximation component signal and the predicted detail component signal.

9. The air conditioning equipment according to claim 8, characterized in that The processing device further includes: The function configuration unit is configured to generate available function options based on the length of the time period, wherein the available function options are available for user selection; the number of available functions corresponding to the available function options is positively correlated with the length of the time period.

10. The air conditioning equipment according to claim 1, characterized in that The detection device includes a temperature sensor, a humidity sensor, a particulate matter sensor, a gas pollutant sensor and / or a volatile organic compound sensor.

Images