An air conditioner energy-saving control method and system for a digital twin model

Abstract

The present invention relates to an air-conditioning energy-saving control method and system for a digital twin model, belonging to the field of air conditioning. The method includes: evenly dividing the temperature adjustment range of a target refrigeration air conditioner to obtain a plurality of reference set temperatures, and inputting each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted duration corresponding to the reference set temperature; taking the reference set temperature corresponding to the predicted power consumption with the smallest value as the priority energy-saving set temperature; Through the present invention, aiming at the technical problem that it is difficult for existing air conditioners to obtain the best initial refrigeration temperature, a digital twin model with a customized structure design is used to intelligently predict the multiple powers consumed and the multiple durations sustained by the refrigeration air conditioner at different initial set temperatures to reach the same desired refrigeration temperature by using a traversal processing mechanism, thus solving the above technical problem.

Description

Technical Field

[0001] The present invention relates to the field of air conditioning, and in particular, to an energy-saving control method and system for an air conditioner based on a digital twin model. Background Art

[0002] With the global warming, the energy problem has become increasingly serious, and energy conservation and environmental protection have become an indispensable part of our daily life. As one of the main devices consuming electricity in fixed places such as homes and offices, how to save electricity and reduce energy consumption during the use of air conditioners is an issue that people need to pay attention to. For example, when a refrigeration air conditioner is operating, after reaching the preset refrigeration temperature, the outdoor unit of the air conditioner (i.e., the compressor and the fan) will receive the temperature controller sensing signal obtained by the air conditioner computer main board and instruct the outdoor unit to stop operating. Of course, when the indoor temperature exceeds the preset temperature, the outdoor unit of the air conditioner will also receive the temperature controller sensing signal obtained by the air conditioner computer main board and instruct the outdoor compressor to operate for refrigeration.

[0003] Therefore, when starting a refrigeration air conditioner, people are accustomed to setting the temperature to the lowest, which actually prolongs the operating time of the air conditioner compressor. The longer the air conditioner compressor operates, the more power is consumed accordingly. Therefore, experienced regulators choose to preset a relatively high temperature, such as 28°C, when starting the machine. When the indoor temperature reaches 28°C and the air conditioner runs for one working cycle, people can then gradually decrease the temperature according to their needs. Thus, even when the same desired refrigeration temperature needs to be achieved, using the same refrigeration air conditioner in the same set closed space, with different initial set temperatures, the power consumed and the duration during the refrigeration process to reach the desired refrigeration temperature are different for the same refrigeration air conditioner.

[0004] Exemplarily, Chinese Patent Publication No. CN107990487A discloses an air conditioner and a method and device for predicting the power consumption of an air conditioner. The method for predicting the power consumption of an air conditioner includes the following steps: establishing a prediction model for the power consumption of the air conditioner using the least squares support vector machine (LSSVM) algorithm; obtaining the current outdoor wet-bulb temperature, the current indoor wet-bulb temperature, and the current indoor ambient temperature; predicting the power consumption of the air conditioner based on the prediction model for the power consumption of the air conditioner, the current outdoor wet-bulb temperature, the current indoor wet-bulb temperature, and the current indoor ambient temperature to obtain the power consumption value of the air conditioner. This prediction method establishes a prediction model for the power consumption of the air conditioner through the LSSVM algorithm, omits complex intermediate variables, has no hypothesis conditions, and the established prediction model is simple, with good applicability and expandability.

[0005] For example, the Chinese invention patent publication text CN109959122A proposes an air-conditioning energy consumption prediction method based on a long short-term memory recurrent neural network, which includes the following steps: Step 1, collect data, collect the water-cooled central air-conditioning data and the corresponding environmental data during the normal operation of the water-cooled central air-conditioning project; Step 2, perform data preprocessing on the obtained water-cooled central air-conditioning data and the corresponding environmental data; Step 3, train the data set to achieve air-conditioning energy consumption prediction. The LSTM-RNN long short-term memory recurrent neural network is used. The preprocessed data set and the corresponding power consumption are used as the input of the LSTM-RNN long short-term memory recurrent neural network. After network training, the final prediction model is obtained; Step 4, input the test data into the prediction model to obtain the energy consumption value of the air-conditioning under the current working conditions. The present invention simplifies the model training process and improves the prediction accuracy.

[0006] However, the above-mentioned prior art is only limited to the power consumption prediction in a single working scenario of the air-conditioning equipment. For example, the power consumption prediction at a single set temperature cannot exhaustively traverse all the working scenarios corresponding to each set temperature for power consumption prediction, let alone automatically set the preferred set temperature based on the prediction results. At the same time, the basic data used to execute the prediction mechanism based on the digital twin model is not comprehensive and sufficient enough, resulting in the predicted power consumption data not being stable and reliable. In addition, the above-mentioned prior art is only limited to the prediction of power consumption and cannot predict other functional parameters of the air-conditioning equipment. Therefore, a prediction mechanism for multi-functional parameters that traverses all the working scenarios corresponding to each set temperature based on the digital twin model is needed to provide more effective basic data for the energy-saving control and function control of the air-conditioning equipment. Summary of the Invention

[0007] In order to solve the technical problems in the prior art, the present invention provides an energy-saving control method and system for an air-conditioning based on a digital twin model. By using a traversal processing mechanism, a digital twin model with a customized structure is used based on a number of comprehensively sufficient basic data selected specifically. The intelligent prediction is carried out on the multiple power consumptions and multiple durations consumed by the refrigeration air-conditioning to control the same set enclosed space to reach the same desired refrigeration temperature at different initial set temperatures, and the optimal initial set temperature is selected based on the intelligently predicted multiple power consumptions and multiple durations to automatically set the refrigeration air-conditioning, so as to balance the requirements of air-conditioning energy consumption savings and refrigeration efficiency requirements, and improve the energy-saving level and automatic control level of the air-conditioning.

[0008] According to the first aspect of the present invention, an energy-saving control method for an air-conditioning based on a digital twin model is provided, and the method includes:

[0009] Obtain the device data of the target refrigeration air conditioner, where the device data of the target refrigeration air conditioner is the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner;

[0010] Obtain multiple internal environment parameters of the set enclosed space and multiple external environment parameters of the set enclosed space. The set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation. The multiple internal environment parameters of the set enclosed space are the temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environment parameters of the set enclosed space are the temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space;

[0011] Perform a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner;

[0012] Evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain multiple reference set temperatures. Input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtain multiple copies of predicted power consumption and multiple copies of predicted consumption duration corresponding to the multiple reference set temperatures respectively. The number of the multiple reference set temperatures exceeds the set number threshold;

[0013] Send the reference set temperature corresponding to the predicted power consumption with the smallest value among the multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature.

[0014] According to the second aspect of the present invention, there is provided an air conditioner energy-saving control system for a digital twin model. The system includes a memory and one or more processors. The memory stores a computer program, and the computer program is configured to be executed by the one or more processors to complete the following steps:

[0015] Obtain the device data of the target refrigeration air conditioner, where the device data of the target refrigeration air conditioner is the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner;

[0016] Obtain multiple internal environment parameters of a set enclosed space and multiple external environment parameters of the set enclosed space. The set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation. The multiple internal environment parameters of the set enclosed space are the air temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environment parameters of the set enclosed space are the air temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space;

[0017] Perform a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner;

[0018] Evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain multiple reference set temperatures. Input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtain multiple copies of predicted power consumption and multiple copies of predicted consumption duration corresponding to the multiple reference set temperatures respectively. The number of the multiple reference set temperatures exceeds the set number threshold;

[0019] Send the reference set temperature corresponding to the predicted power consumption with the smallest value among the multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature.

[0020] According to the third aspect of the present invention, there is provided an air conditioner energy-saving control system for a digital twin model. The system includes:

[0021] The first acquisition device is used to obtain various device data of the target refrigeration air conditioner. The various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner;

[0022] The second acquisition device is used to obtain multiple internal environment parameters of the set enclosed space and multiple external environment parameters of the set enclosed space. The set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation. The multiple internal environment parameters of the set enclosed space are the air temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environment parameters of the set enclosed space are the air temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space;

[0023] A multi-layer construction device is used to perform a set number of multiple trainings on a deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner;

[0024] A traversal analysis device is respectively connected to the first acquisition device, the second acquisition device, and the multi-layer construction device. It is used to evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain multiple reference set temperatures, input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtain multiple copies of predicted power consumption and multiple copies of predicted consumption duration respectively corresponding to the multiple reference set temperatures. The number of the multiple reference set temperatures exceeds a set number threshold;

[0025] An energy-saving control device is respectively connected to the traversal analysis device and the main controller of the target refrigeration air conditioner. It is used to send the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature.

[0026] Compared with the prior art, the present invention has at least the following five key inventive points:

[0027] Inventive point A: For the set closed space cooled by the target refrigeration air conditioner, traverse each set cooling temperature to respectively predict the predicted power consumption and predicted consumption duration of the target refrigeration air conditioner when the target refrigeration air conditioner reaches the user's desired cooling temperature at each set cooling temperature, so as to provide a prediction data basis for the screening of the set cooling temperature with the best energy-saving effect;

[0028] Inventive point B: Send the set cooling temperature corresponding to the predicted power consumption with the smallest value to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature for automatic setting of the set cooling temperature with the best energy-saving effect. Particularly importantly, when there are two predicted power consumptions with equal minimum values, among the two set cooling temperatures corresponding to the two predicted power consumptions with equal minimum values, the set cooling temperature with the shortest corresponding predicted consumption duration is used as the priority energy-saving set temperature, so as to balance air conditioner energy consumption savings and air conditioner cooling efficiency;

[0029] Inventive Point C: Introduce a digital twin model to simulate, analyze, and optimize the operation process of the air conditioner entity based on the collected data and environmental data of the air conditioner entity, and complete the intelligent prediction of the predicted power consumption and predicted duration of the target refrigeration air conditioner when it reaches the user-expected refrigeration temperature at each set refrigeration temperature. Specifically, the digital twin model is a deep neural network after multiple trainings, and the number of trainings is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner, so as to design digital twin models with different structures for different refrigeration air conditioners;

[0030] Inventive Point D: Use the working voltage and working frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environmental parameters of the set enclosed space, and multiple external environments of the set enclosed space as the item-by-item basic information for the digital twin model to perform intelligent prediction. The comprehensive and targeted screening of the above basic information ensures the stability and effectiveness of the intelligent prediction results of the digital twin model. Specifically, the various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner horsepower, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner. The multiple internal environmental parameters of the set enclosed space are the temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environmental parameters of the set enclosed space are the temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space, and the set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation;

[0031] Inventive Point E: In each training of the deep neural network, use the predicted power consumption and predicted duration corresponding to a certain past set temperature as the two output contents of the deep neural network, and use the user-expected refrigeration temperature, the certain past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environmental parameters of the set enclosed space corresponding to the certain past set temperature, and multiple external environmental parameters of the set enclosed space corresponding to the certain past set temperature as the item-by-item input contents of the deep neural network to complete this training, so as to ensure the training effect of each training of the deep neural network, and further ensure the stability and effectiveness of the intelligent prediction results of the digital twin model. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The following will describe the embodiments of the present invention with reference to the drawings, where:

[0033] Figure 1 It is a technical flow chart of the air conditioner energy-saving control method and system based on the digital twin model according to the present invention.

[0034] Figure 2The flowchart of the air conditioner energy-saving control method for the digital twin model shown in Embodiment 1 of the present invention.

[0035] Figure 3 The flowchart of the air conditioner energy-saving control method for the digital twin model shown in Embodiment 2 of the present invention.

[0036] Figure 4 The flowchart of the air conditioner energy-saving control method for the digital twin model shown in Embodiment 3 of the present invention.

[0037] Figure 5 The flowchart of the air conditioner energy-saving control method for the digital twin model shown in Embodiment 4 of the present invention.

[0038] Figure 6 The structural schematic diagram of the air conditioner energy-saving control system for the digital twin model shown in Embodiment 5 of the present invention.

[0039] Figure 7 The structural schematic diagram of the air conditioner energy-saving control system for the digital twin model shown in Embodiment 6 of the present invention. Detailed implementation manners

[0040] As Figure 1 shown, the technical flowchart of the air conditioner energy-saving control method and system for the digital twin model shown in the present invention is given.

[0041] As Figure 1 shown, the specific technical process of the present invention is as follows:

[0042] Technical process one: Design a digital twin model with a customized structure for the target refrigeration air conditioner to provide a solution mechanism for the intelligent prediction of the power consumption and consumption duration required for the target refrigeration air conditioner to reach the user's desired refrigeration temperature at each current set refrigeration temperature.

[0043] Specifically, the structural customization of the digital twin model lies in the following aspects:

[0044] First aspect: The digital twin model is a deep neural network after multiple trainings. For example, the deep neural network includes multiple hidden layers, a single input layer, and a single output layer, and the multiple hidden layers are arranged between the single input layer and the single output layer;

[0045] Second aspect: The number of trainings of the deep neural network is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner, so as to design digital twin models with different structures for different refrigeration air conditioners;

[0046] Third aspect: In each training of the deep neural network, the predicted power consumption and predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network. The user's desired cooling temperature, the past set temperature, the working voltage and frequency of the target cooling air conditioner, various device data of the target cooling air conditioner, multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the input contents item by item of the deep neural network to complete this training, thus ensuring the training effect of each training of the deep neural network and ensuring the stability and effectiveness of the intelligent prediction results of the digital twin model;

[0047] Technical process two: Screen multiple pieces of basic information specifically for the digital twin model. The comprehensive and targeted screening of the above-mentioned multiple pieces of basic information further ensures the stability and effectiveness of the intelligent prediction results of the digital twin model;

[0048] Specifically, the above-mentioned multiple pieces of basic information include the working voltage and frequency of the target cooling air conditioner, various device data of the target cooling air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environments of the set enclosed space;

[0049] More specifically, the various device data of the target cooling air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, cooling power, heating power, circulating air volume, and frequency conversion identifier of the target cooling air conditioner. The multiple internal environment parameters of the set enclosed space are the temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environment parameters of the set enclosed space are the temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space. Among them, the set enclosed space is the space where the target cooling air conditioner performs temperature regulation;

[0050] Technical process three: For the set enclosed space cooled by the target cooling air conditioner, the digital twin model with a customized structure design, based on multiple pieces of basic information screened specifically, traverses each set cooling temperature to respectively predict the predicted power consumption and predicted duration of the target cooling air conditioner when the target cooling air conditioner reaches the user's desired cooling temperature at each set cooling temperature;

[0051] The respective predicted power consumptions and respective predicted durations corresponding to each set cooling temperature obtained by traversing each set cooling temperature provide a prediction data basis for screening the set cooling temperature with the best energy-saving effect automatically;

[0052] Technical Process 4: Obtain the predicted power consumption and predicted duration corresponding to each set cooling temperature output by Technical Process 3. Use the set cooling temperature corresponding to the predicted power consumption with the smallest value as the priority energy-saving set temperature and send it to the main controller of the target refrigeration air conditioner to automatically set the set cooling temperature for the best energy-saving effect;

[0053] Specifically, when there are two equal minimum predicted power consumptions, among the two set cooling temperatures corresponding to the two equal minimum predicted power consumptions respectively, use the set cooling temperature with the shortest corresponding predicted duration as the priority energy-saving set temperature, thus taking into account both air conditioner energy consumption savings and air conditioner cooling efficiency, and improving the intelligence level of the refrigeration air conditioner;

[0054] Particularly crucial is to evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain multiple reference set temperatures. The number of the multiple reference set temperatures exceeds the set number threshold. Through the above settings, it is possible to exhaust as many as possible the selectable set temperature values within the temperature adjustment range of the target refrigeration air conditioner and perform traversal intelligent prediction, thereby providing data guarantee for the accuracy of the selection of the priority energy-saving set temperature;

[0055] Thus, through the design of the above technical processes, it is possible to intelligently predict the multiple powers consumed and multiple durations taken for the same refrigeration air conditioner to control the same set closed space to reach the same desired cooling temperature under different initial set temperatures, and based on the intelligently predicted multiple power consumptions and multiple durations, select the optimal initial set temperature to automatically set the refrigeration air conditioner, thereby taking into account the air conditioner energy consumption savings requirement and the refrigeration efficiency requirement, and improving the energy-saving level and automatic control level of the air conditioner.

[0056] The key point of the present invention lies in: as Figure 1 shown, introducing a digital twin model for simulating and analyzing the operation process of the air conditioner entity based on the collected data and environmental data of the air conditioner entity and providing a data basis for subsequent optimization, completing the intelligent prediction of the predicted power consumption and predicted duration of the target refrigeration air conditioner when reaching the user's desired cooling temperature at each set cooling temperature, and automatically selecting and setting the best set cooling temperature that takes into account both power consumption and duration requirements.

[0057] Next, the air conditioner energy-saving control method and system of the digital twin model of the present invention will be specifically described by way of embodiments.

[0058] Embodiment 1

[0059] Figure 2 It is a step flow chart of the air conditioner energy-saving control method of the digital twin model shown in Embodiment 1 according to the present invention.

[0060] As Figure 2 shown, the air-conditioning energy-saving control method of the digital twin model includes the following specific steps:

[0061] Step S11: Obtain various device data of the target refrigeration air conditioner. The various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion flag of the target refrigeration air conditioner;

[0062] Exemplarily, obtaining various device data of the target refrigeration air conditioner, where the various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion flag of the target refrigeration air conditioner includes: Multiple different Internet of Things acquisition components can be used to separately acquire the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion flag of the target refrigeration air conditioner;

[0063] Step S12: Obtain multiple internal environment parameters of the set enclosed space and multiple external environment parameters of the set enclosed space. The set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation. The multiple internal environment parameters of the set enclosed space are the air temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environment parameters of the set enclosed space are the air temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space;

[0064] Specifically, obtaining multiple internal environment parameters of the set enclosed space and multiple external environment parameters of the set enclosed space. The set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation. The multiple internal environment parameters of the set enclosed space are the air temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environment parameters of the set enclosed space are the air temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space includes: Storing the multiple internal environment parameters of the set enclosed space and the multiple external environment parameters of the set enclosed space into different physical storage spaces respectively;

[0065] Step S13: Perform a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner;

[0066] Exemplarily, a set number of multiple trainings are performed on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner, including: a numerical mapping function can be used to represent the numerical mapping relationship between the value of the set number and the compression ratio of the compressor of the target refrigeration air conditioner;

[0067] Specifically, in the numerical mapping function, the compression ratio of the compressor of the target refrigeration air conditioner is the input data of the numerical mapping function, and the value of the set number is the output data of the numerical mapping function;

[0068] And exemplarily, performing a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The positive correlation between the value of the set number and the compression ratio of the compressor of the target refrigeration air conditioner further includes: when the compression ratio of the compressor of the target refrigeration air conditioner is 3, the value of the set number is 100; when the compression ratio of the compressor of the target refrigeration air conditioner is 5, the value of the set number is 150; when the compression ratio of the compressor of the target refrigeration air conditioner is 7, the value of the set number is 180; and when the compression ratio of the compressor of the target refrigeration air conditioner is 9, the value of the set number is 200, and so on;

[0069] Step S14: Uniformly divide the temperature adjustment range of the target refrigeration air conditioner to obtain multiple reference set temperatures, input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtain multiple copies of predicted power consumption and multiple copies of predicted consumption duration corresponding to the multiple reference set temperatures respectively. The number of the multiple reference set temperatures exceeds a set number threshold;

[0070] Specifically, it can be selected to use the MATLAB toolbox to implement the simulation and test of the numerical processing process of inputting each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature;

[0071] Step S15: Send the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature;

[0072] Exemplarily, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature includes: the main controller of the target refrigeration air conditioner is an MCU controller, such as the 51 single-chip microcomputer series or the ARM series;

[0073] Among them, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple predicted power consumptions as the priority energy-saving set temperature to the main controller of the target refrigeration air conditioner includes: after receiving the priority energy-saving set temperature, the main controller of the target refrigeration air conditioner automatically sets the priority energy-saving set temperature as the refrigeration set temperature of the target refrigeration air conditioner;

[0074] Among them, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple predicted power consumptions as the priority energy-saving set temperature to the main controller of the target refrigeration air conditioner further includes: when there are two predicted power consumptions with equal minimum values among multiple predicted power consumptions, taking the two reference set temperatures corresponding to the two predicted power consumptions with equal minimum values as the first energy-saving set temperature and the second energy-saving set temperature, taking the predicted consumption duration corresponding to the first energy-saving set temperature as the first predicted consumption duration, taking the predicted consumption duration corresponding to the second energy-saving set temperature as the second predicted consumption duration, and taking the energy-saving set temperature corresponding to the predicted consumption duration with the smallest value among the first predicted consumption duration and the second predicted consumption duration as the priority energy-saving set temperature;

[0075] Among them, inputting each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtaining multiple predicted power consumptions and multiple predicted consumption durations corresponding to multiple reference set temperatures respectively includes: using the digital twin model to intelligently predict, based on the user's desired refrigeration temperature, the reference set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space, and the multiple external environment parameters of the set enclosed space, the power consumed by the target refrigeration air conditioner when the set enclosed space reaches the user's desired refrigeration temperature after the reference set temperature is set as the refrigeration set temperature of the target refrigeration air conditioner as the predicted power consumption corresponding to the reference set temperature, and the duration of use of the target refrigeration air conditioner as the predicted consumption duration corresponding to the reference set temperature;

[0076] Among them, a set number of multiple trainings are performed on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner, including: in each training performed on the deep neural network, the predicted power consumption and predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network, and the user's desired refrigeration temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network to complete this training;

[0077] Specifically, in each training performed on the deep neural network, the predicted power consumption and predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network, and the user's desired refrigeration temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training includes: it is possible to choose to use an FPGA chip to implement the simulation and testing of each training performed on the deep neural network;

[0078] And among them, in each training performed on the deep neural network, the predicted power consumption and predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network, and the user's desired refrigeration temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training includes: the predicted power consumption and predicted duration corresponding to the past set temperature are the actual power consumption and actual usage duration of the target refrigeration air conditioner during the process from when the target refrigeration air conditioner is set to the past set temperature until the set enclosed space reaches the user's desired refrigeration temperature;

[0079] And among them, in each training of the deep neural network, the predicted power consumption and the predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network, and the user's desired cooling temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training further includes: The multiple internal environment parameters of the set enclosed space corresponding to the past set temperature are the multiple internal environment parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are the multiple external environment parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature.

[0080] Embodiment 2

[0081] Figure 3 It is a step flowchart of the air conditioner energy-saving control method of the digital twin model shown in Embodiment 2 of the present invention.

[0082] As Figure 3 shown, different from the embodiments in Figure 2 in the air conditioner energy-saving control method of the digital twin model, after performing a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner, and the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner, then, after step S13, the method further includes:

[0083] Step S16: Use the cloud computing storage node to store various model parameters of the digital twin model to complete the model storage of the digital twin model;

[0084] Exemplarily, using the cloud computing storage node to store various model parameters of the digital twin model to complete the model storage of the digital twin model includes: The cloud computing storage node includes each cloud computing storage network element for respectively storing various model parameters of the digital twin model.

[0085] Embodiment 3

[0086] Figure 4 It is a step flowchart of the air conditioner energy-saving control method of the digital twin model shown in Embodiment 3 of the present invention.

[0087] As Figure 4 shown, different fromFigure 2 Different from the embodiments in the embodiment, in the air conditioning energy-saving control method of the digital twin model, after sending the reference set temperature corresponding to the predicted power consumption with the minimum value among the multiple predicted power consumptions as the priority energy-saving set temperature to the main controller of the target refrigeration air conditioner, that is, after step S15, the method further includes:

[0088] Step S17: using an LED display array or a liquid crystal display screen to receive and display the priority energy-saving set temperature;

[0089] Alternatively, an LCD display array may be selected to replace the LED display array or the liquid crystal display screen to receive and display the priority energy-saving set temperature.

[0090] Example 4

[0091] Figure 5 This is a flowchart of the steps of the air conditioning energy-saving control method of the digital twin model according to Example 4 of the present invention.

[0092] like Figure 5 As shown, Figure 2 Different from the embodiments in the embodiment, in the air conditioning energy-saving control method of the digital twin model, after sending the reference set temperature corresponding to the predicted power consumption with the minimum value among the multiple predicted power consumptions as the priority energy-saving set temperature to the main controller of the target refrigeration air conditioner, that is, after step S15, the method further includes:

[0093] Step S18: wirelessly sending the received priority energy-saving setting temperature to a remote energy-saving management server using a frequency division duplex communication link or a time division duplex communication link;

[0094] For example, a frequency division duplex communication link or a time division duplex communication link is used to wirelessly send the received priority energy-saving set temperature to a remote energy-saving management server, including: the remote energy-saving management server is a big data service network element or a blockchain service network element, which is used to perform energy-saving management of air-conditioning equipment at various locations.

[0095] Next, various method embodiments of the present invention are described in detail.

[0096] In the air conditioning energy-saving control method of the digital twin model according to each method embodiment of the present invention:

[0097] Obtain the device data of the target refrigeration air conditioner. The device data of the target refrigeration air conditioner includes the energy efficiency ratio, applicable area, air conditioner capacity, cooling power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner. Specifically, when the frequency conversion identifier of the target refrigeration air conditioner is 0B01, it indicates that the target refrigeration air conditioner is a variable-frequency air conditioner; when the frequency conversion identifier of the target refrigeration air conditioner is 0B00, it indicates that the target refrigeration air conditioner is a fixed-frequency air conditioner.

[0098] Specifically, to obtain the device data of the target refrigeration air conditioner, the device data of the target refrigeration air conditioner includes the energy efficiency ratio, applicable area, air conditioner capacity, cooling power, heating power, circulating air volume, and frequency conversion identifier. It also includes that multiple different local storage devices can be used to separately store the energy efficiency ratio, applicable area, air conditioner capacity, cooling power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner.

[0099] And in the air conditioner energy-saving control method of the digital twin model according to each method embodiment of the present invention:

[0100] Using the digital twin model, based on the user's desired cooling temperature, the reference set temperature, the working voltage and frequency of the target refrigeration air conditioner, the device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space, and the multiple external environment parameters of the set enclosed space, intelligently predict the power consumed by the target refrigeration air conditioner when the set enclosed space reaches the user's desired cooling temperature after the reference set temperature is used as the cooling set temperature of the target refrigeration air conditioner, as the predicted power consumption corresponding to the reference set temperature, and the usage duration of the target refrigeration air conditioner as the predicted consumption duration corresponding to the reference set temperature. This includes parallelly inputting the user's desired cooling temperature, the reference set temperature, the working voltage and frequency of the target refrigeration air conditioner, the device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space, and the multiple external environment parameters of the set enclosed space into the digital twin model.

[0101] Among them, using the digital twin model, based on the user's desired refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environmental parameters of the set enclosed space, and multiple external environmental parameters of the set enclosed space, intelligently predict the power consumed by the target refrigeration air conditioner when the set enclosed space reaches the user's desired refrigeration temperature after the reference set temperature is the refrigeration set temperature of the target refrigeration air conditioner as the predicted power consumption corresponding to the reference set temperature, and the duration of use of the target refrigeration air conditioner as the predicted consumption duration corresponding to the reference set temperature. It further includes: running the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature output by the digital twin model.

[0102] Embodiment 5

[0103] Figure 6 It is a schematic structural diagram of an air conditioner energy-saving control system of a digital twin model shown in Embodiment 5 according to the present invention.

[0104] As Figure 6 shown, the air conditioner energy-saving control system of the digital twin model includes a memory and one or more processors. The memory stores a computer program, and the computer program is configured to be executed by the one or more processors to complete the following steps:

[0105] Step S11: Obtain various device data of the target refrigeration air conditioner. The various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner;

[0106] Exemplarily, obtaining various device data of the target refrigeration air conditioner, where the various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner includes: multiple different Internet of Things acquisition components can be used to respectively acquire the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner;

[0107] Step S12: Obtain multiple internal environmental parameters of the set enclosed space and multiple external environmental parameters of the set enclosed space. The set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation. The multiple internal environmental parameters of the set enclosed space are the temperature, humidity, air pressure, and space volume inside the set enclosed space, and the multiple external environmental parameters of the set enclosed space are the temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space;

[0108] Specifically, obtain multiple internal environment parameters of a set enclosed space and multiple external environment parameters of the set enclosed space. The set enclosed space is the space where the target refrigeration air conditioner performs temperature regulation. The multiple internal environment parameters of the set enclosed space are the air temperature, humidity, air pressure, and space volume inside the set enclosed space. The multiple external environment parameters of the set enclosed space are the air temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space, including: storing the multiple internal environment parameters of the set enclosed space and the multiple external environment parameters of the set enclosed space into different physical storage spaces respectively;

[0109] Step S13: Perform a set number of multiple trainings on the deep neural network to obtain the deep neural network after performing multiple trainings and output it as the digital twin model of the target refrigeration air conditioner. The value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner;

[0110] Exemplarily, performing a set number of multiple trainings on the deep neural network to obtain the deep neural network after performing multiple trainings and output it as the digital twin model of the target refrigeration air conditioner, where the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner includes: a numerical mapping function can be used to represent the numerical mapping relationship between the value of the set number and the compression ratio of the compressor of the target refrigeration air conditioner;

[0111] Specifically, in the numerical mapping function, the compression ratio of the compressor of the target refrigeration air conditioner is the input data of the numerical mapping function, and the value of the set number is the output data of the numerical mapping function;

[0112] And exemplarily, performing a set number of multiple trainings on the deep neural network to obtain the deep neural network after performing multiple trainings and output it as the digital twin model of the target refrigeration air conditioner, where the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner further includes: when the compression ratio of the compressor of the target refrigeration air conditioner is 3, the value of the set number is 100; when the compression ratio of the compressor of the target refrigeration air conditioner is 5, the value of the set number is 150; when the compression ratio of the compressor of the target refrigeration air conditioner is 7, the value of the set number is 180; and when the compression ratio of the compressor of the target refrigeration air conditioner is 9, the value of the set number is 200, and so on;

[0113] Step S14: Uniformly divide the temperature adjustment range of the target refrigeration air conditioner to obtain multiple reference set temperatures, input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtain multiple copies of predicted power consumption and multiple copies of predicted consumption duration corresponding to the multiple reference set temperatures respectively. The number of the multiple reference set temperatures exceeds a set number threshold;

[0114] Specifically, it can be selected to use the MATLAB toolbox to implement the simulation and test of the numerical processing process of inputting each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature;

[0115] Step S15: Send the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature;

[0116] Exemplarily, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner includes: The main controller of the target refrigeration air conditioner is an MCU controller, such as the 51 single-chip microcomputer series or the ARM series;

[0117] Among them, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner includes: After receiving the priority energy-saving set temperature, the main controller of the target refrigeration air conditioner automatically sets the priority energy-saving set temperature as the refrigeration set temperature of the target refrigeration air conditioner;

[0118] Among them, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner further includes: When there are two predicted power consumptions with equal minimum values among multiple copies of predicted power consumption, take the two reference set temperatures corresponding to the two predicted power consumptions with equal minimum values as the first energy-saving set temperature and the second energy-saving set temperature, take the predicted consumption duration corresponding to the first energy-saving set temperature as the first predicted consumption duration, take the predicted consumption duration corresponding to the second energy-saving set temperature as the second predicted consumption duration, and take the energy-saving set temperature corresponding to the predicted consumption duration with the smallest value among the first predicted consumption duration and the second predicted consumption duration as the priority energy-saving set temperature;

[0119] Among them, each reference set temperature is input into the digital twin model to obtain the predicted power consumption and predicted duration corresponding to the reference set temperature, and obtaining multiple copies of predicted power consumption and multiple copies of predicted duration corresponding to multiple reference set temperatures respectively includes: using the digital twin model to intelligently predict, based on the user's desired refrigeration temperature, the reference set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space, and the multiple external environment parameters of the set enclosed space, the power exhausted by the target refrigeration air conditioner when the set enclosed space reaches the user's desired refrigeration temperature after the reference set temperature is the refrigeration set temperature of the target refrigeration air conditioner as the predicted power consumption corresponding to the reference set temperature, and the duration of use of the target refrigeration air conditioner as the predicted duration corresponding to the reference set temperature;

[0120] Among them, performing a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner, and the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner includes: in each training performed on the deep neural network, taking the predicted power consumption and predicted duration corresponding to a known past set temperature as two output contents of the deep neural network, and taking the user's desired refrigeration temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature as the item-by-item input contents of the deep neural network to complete this training;

[0121] Specifically, in each training performed on the deep neural network, taking the predicted power consumption and predicted duration corresponding to a known past set temperature as two output contents of the deep neural network, and taking the user's desired refrigeration temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature as the item-by-item input contents of the deep neural network to complete this training includes: it is possible to choose to use an FPGA chip to implement the simulation and testing of each training performed on the deep neural network;

[0122] And among them, in each training performed on the deep neural network, the predicted power consumption and predicted power consumption duration corresponding to a known past set temperature are used as two output contents of the deep neural network. The user's desired cooling temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training includes: the predicted power consumption and predicted power consumption duration corresponding to the past set temperature are the actual power consumption and actual usage duration of the target refrigeration air conditioner from when the target refrigeration air conditioner is set to the past set temperature until the set enclosed space reaches the user's desired cooling temperature;

[0123] And among them, in each training performed on the deep neural network, the predicted power consumption and predicted power consumption duration corresponding to a known past set temperature are used as two output contents of the deep neural network. The user's desired cooling temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training further includes: the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature are the multiple internal environment parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are the multiple external environment parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature;

[0124] As Figure 6 shown, exemplarily, P processors are given, where P is a natural number greater than or equal to 1.

[0125] Embodiment 6

[0126] Figure 7 It is a schematic structural diagram of the air conditioner energy-saving control system of the digital twin model shown in Embodiment 6 of the present invention.

[0127] As Figure 7 shown, the air conditioner energy-saving control system of the digital twin model includes the following components:

[0128] A first acquisition device, configured to acquire various device data of a target refrigeration air conditioner, where the various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner;

[0129] Exemplarily, acquiring various device data of a target refrigeration air conditioner, where the various device data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier includes: multiple different Internet of Things acquisition components can be used to respectively acquire the energy efficiency ratio, applicable area, air conditioner capacity, refrigeration power, heating power, circulating air volume, and frequency conversion identifier of the target refrigeration air conditioner;

[0130] A second acquisition device, configured to acquire multiple internal environment parameters and multiple external environment parameters of a set enclosed space, where the set enclosed space is the space in which the target refrigeration air conditioner performs temperature regulation, the multiple internal environment parameters of the set enclosed space are the temperature, humidity, air pressure, and space volume inside the set enclosed space, and the multiple external environment parameters of the set enclosed space are the temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space;

[0131] Specifically, acquiring multiple internal environment parameters and multiple external environment parameters of a set enclosed space, where the set enclosed space is the space in which the target refrigeration air conditioner performs temperature regulation, the multiple internal environment parameters of the set enclosed space are the temperature, humidity, air pressure, and space volume inside the set enclosed space, and the multiple external environment parameters of the set enclosed space are the temperature, humidity, wind direction, wind speed, and sunlight illumination angle outside the set enclosed space includes: storing the multiple internal environment parameters and the multiple external environment parameters of the set enclosed space into different physical storage spaces respectively;

[0132] A multi-layer construction device, configured to perform a set number of multiple trainings on a deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner, where the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner;

[0133] Exemplarily, performing a set number of multiple trainings on a deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner, where the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner includes: a numerical mapping function can be used to represent the numerical mapping relationship between the value of the set number and the compression ratio of the compressor of the target refrigeration air conditioner;

[0134] Specifically, in the numerical mapping function, the compression ratio of the compressor of the target refrigeration air conditioner is the input data of the numerical mapping function, and the value of the set quantity is the output data of the numerical mapping function;

[0135] And by way of example, performing a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and outputting it as the digital twin model of the target refrigeration air conditioner. The value of the set quantity being positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner further includes: when the compression ratio of the compressor of the target refrigeration air conditioner is 3, the value of the set quantity is 100; when the compression ratio of the compressor of the target refrigeration air conditioner is 5, the value of the set quantity is 150; when the compression ratio of the compressor of the target refrigeration air conditioner is 7, the value of the set quantity is 180; and when the compression ratio of the compressor of the target refrigeration air conditioner is 9, the value of the set quantity is 200, and so on;

[0136] A traversal analysis device, configured to evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain a plurality of reference set temperatures, input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted duration corresponding to the reference set temperature, and obtain multiple copies of predicted power consumption and multiple copies of predicted duration corresponding to the respective reference set temperatures. The number of the plurality of reference set temperatures exceeds a set quantity threshold;

[0137] Specifically, it can be selected to use the MATLAB toolbox to implement the simulation and test of the numerical processing process of inputting each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted duration corresponding to the reference set temperature;

[0138] An energy-saving control device, configured to send the reference set temperature corresponding to the predicted power consumption with the smallest value among the multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner as the priority energy-saving set temperature;

[0139] By way of example, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among the multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner includes: the main controller of the target refrigeration air conditioner is an MCU controller, such as the 51 single-chip microcomputer series or the ARM series;

[0140] Among them, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among the multiple copies of predicted power consumption to the main controller of the target refrigeration air conditioner includes: after receiving the priority energy-saving set temperature, the main controller of the target refrigeration air conditioner automatically sets the priority energy-saving set temperature as the refrigeration set temperature of the target refrigeration air conditioner;

[0141] Among them, sending the reference set temperature corresponding to the predicted power consumption with the smallest value among multiple predicted power consumptions to the main controller of the target refrigeration air conditioner further includes: when there are two predicted power consumptions with equal minimum values among multiple predicted power consumptions, taking the two reference set temperatures corresponding to the two predicted power consumptions with equal minimum values as the first energy-saving set temperature and the second energy-saving set temperature, taking the predicted consumption duration corresponding to the first energy-saving set temperature as the first predicted consumption duration, taking the predicted consumption duration corresponding to the second energy-saving set temperature as the second predicted consumption duration, and taking the energy-saving set temperature corresponding to the predicted consumption duration with the smallest value among the first predicted consumption duration and the second predicted consumption duration as the priority energy-saving set temperature;

[0142] Among them, inputting each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtaining multiple predicted power consumptions and multiple predicted consumption durations corresponding to multiple reference set temperatures respectively includes: using the digital twin model to intelligently predict, based on the user's desired refrigeration temperature, the reference set temperature, the working voltage and frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space, and the multiple external environment parameters of the set enclosed space, the power consumed by the target refrigeration air conditioner when the set enclosed space reaches the user's desired refrigeration temperature after the reference set temperature is set as the refrigeration set temperature of the target refrigeration air conditioner as the predicted power consumption corresponding to the reference set temperature, and the duration of use of the target refrigeration air conditioner as the predicted consumption duration corresponding to the reference set temperature;

[0143] Among them, performing a set number of multiple trainings on the deep neural network to obtain the deep neural network after multiple trainings and output it as the digital twin model of the target refrigeration air conditioner, and the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner includes: in each training performed on the deep neural network, taking the predicted power consumption and predicted consumption duration corresponding to a known past set temperature as the two output contents of the deep neural network, and taking the user's desired refrigeration temperature, the past set temperature, the working voltage and frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature as the item-by-item input contents of the deep neural network to complete this training;

[0144] Specifically, in each training of the deep neural network, the predicted power consumption and predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network. The user's desired cooling temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training includes: optionally using an FPGA chip to implement the simulation and testing of each training of the deep neural network;

[0145] And among them, in each training of the deep neural network, the predicted power consumption and predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network. The user's desired cooling temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training includes: the predicted power consumption and predicted duration corresponding to the past set temperature are the actual power consumption and actual usage duration of the target refrigeration air conditioner during the process from when the target refrigeration air conditioner is set to the past set temperature until the set enclosed space reaches the user's desired cooling temperature;

[0146] And among them, in each training of the deep neural network, the predicted power consumption and predicted duration corresponding to a known past set temperature are used as two output contents of the deep neural network. The user's desired cooling temperature, the past set temperature, the working voltage and working frequency of the target refrigeration air conditioner, the various device data of the target refrigeration air conditioner, the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are used as the item-by-item input contents of the deep neural network. Completing this training further includes: the multiple internal environment parameters of the set enclosed space corresponding to the past set temperature are the multiple internal environment parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature, and the multiple external environment parameters of the set enclosed space corresponding to the past set temperature are the multiple external environment parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature.

[0147] In addition, the present invention can also incorporate the following technical contents in various places to highlight the significant technical progress of the present invention:

[0148] Parallelly inputting the user-expected refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environment parameters of the set enclosed space into the digital twin model includes: before parallelly inputting the user-expected refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environment parameters of the set enclosed space into the digital twin model, respectively performing numerical normalization processing on the user-expected refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environment parameters of the set enclosed space;

[0149] Exemplarily, before parallelly inputting the user-expected refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environment parameters of the set enclosed space into the digital twin model, respectively performing numerical normalization processing on the user-expected refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environment parameters of the set enclosed space includes: before parallelly inputting the user-expected refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environment parameters of the set enclosed space into the digital twin model, respectively performing octal numerical conversion processing on the user-expected refrigeration temperature, the reference set temperature, the operating voltage and frequency of the target refrigeration air conditioner, various device data of the target refrigeration air conditioner, multiple internal environment parameters of the set enclosed space, and multiple external environment parameters of the set enclosed space;

[0150] And wherein, running the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature output by the digital twin model includes: the predicted power consumption and predicted consumption duration corresponding to the reference set temperature output by the digital twin model are both in the numerical representation form after numerical normalization processing;

[0151] Exemplarily, the predicted power consumption and the predicted consumption duration corresponding to the reference set temperature output by the digital twin model are both in the form of numerical representation after numerical normalization processing, including: the predicted power consumption and the predicted consumption duration corresponding to the reference set temperature output by the digital twin model are both in the form of numerical representation after octal numerical conversion processing.

[0152] The above are only the preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A digital twin model air conditioning energy-saving control method, characterized in that: The method comprises: Acquire various equipment data of the target refrigeration air conditioner, wherein the various equipment data of the target refrigeration air conditioner are energy efficiency ratio, applicable area, air conditioner number, cooling power, heating power, circulating air volume and frequency conversion mark of the target refrigeration air conditioner; Acquire multiple internal environmental parameters of a set closed space and multiple external environmental parameters of the set closed space, wherein the set closed space is a space in which the target refrigeration air conditioner performs temperature control, the multiple internal environmental parameters of the set closed space are the temperature, humidity, air pressure and space volume inside the set closed space, and the multiple external environmental parameters of the set closed space are the temperature, humidity, wind direction, wind speed and sunlight angle outside the set closed space; Performing a set number of multiple trainings on the deep neural network to obtain a deep neural network after the multiple trainings and outputting the deep neural network as the digital twin model of the target refrigeration air conditioner, wherein the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner; Evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain a plurality of reference set temperatures, input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtain a plurality of predicted power consumption and a plurality of predicted consumption durations corresponding to the plurality of reference set temperatures, wherein the number of the plurality of reference set temperatures exceeds a set number threshold; Sending a reference set temperature corresponding to the predicted power consumption with the minimum value among the multiple predicted power consumptions as a priority energy-saving set temperature to a main controller of the target refrigeration air conditioner; Wherein, after receiving the priority energy-saving setting temperature, the main controller of the target refrigeration air conditioner automatically sets the priority energy-saving setting temperature as the refrigeration setting temperature of the target refrigeration air conditioner; Among them, when two sets of predicted power consumption with the same minimum value appear in multiple sets of predicted power consumption, the two reference set temperatures corresponding to the two sets of predicted power consumption with the same minimum value are used as the first energy-saving set temperature and the second energy-saving set temperature, the predicted consumption time corresponding to the first energy-saving set temperature and the second energy-saving set temperature are used as the first predicted consumption time and the second predicted consumption time, and the energy-saving set temperature corresponding to the predicted consumption time with the minimum value of the two predicted consumption time is used as the priority energy-saving set temperature; Among them, the digital twin model is used to intelligently predict the power consumed by the target refrigeration air conditioner when the closed space reaches the user's expected cooling temperature after the reference set temperature is set as the refrigeration set temperature of the target refrigeration air conditioner based on the user's expected cooling temperature, the reference set temperature, the operating voltage and operating frequency of the target refrigeration air conditioner, various equipment data of the target refrigeration air conditioner, multiple internal environmental parameters of the set closed space, and multiple external environmental parameters of the set closed space as the predicted power consumption corresponding to the reference set temperature, and the usage time of the target refrigeration air conditioner as the predicted consumption time corresponding to the reference set temperature.

2. The air conditioning energy-saving control method of the digital twin model according to claim 1, characterized in that: Performing a set number of multiple trainings on the deep neural network to obtain a deep neural network after multiple trainings and outputting it as the digital twin model of the target refrigeration air conditioner, wherein the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner, including: in each training performed on the deep neural network, using the predicted power consumption and the predicted consumption time corresponding to a known past set temperature as two output contents of the deep neural network, and using the user's expected refrigeration temperature, the past set temperature, the operating voltage and operating frequency of the target refrigeration air conditioner, various equipment data of the target refrigeration air conditioner, multiple internal environmental parameters of the set enclosed space corresponding to the past set temperature, and multiple external environmental parameters of the set enclosed space corresponding to the past set temperature as input contents of the deep neural network item by item to complete this training; Among them, in each training performed on the deep neural network, the predicted power consumption and predicted consumption time corresponding to a known past set temperature are used as two output contents of the deep neural network, and the user's expected cooling temperature, the past set temperature, the operating voltage and operating frequency of the target cooling air conditioner, various equipment data of the target cooling air conditioner, multiple internal environmental parameters of the set closed space corresponding to the past set temperature, and multiple external environmental parameters of the set closed space corresponding to the past set temperature are used as the input content of the deep neural network item by item, and completing this training includes: the predicted power consumption and predicted consumption time corresponding to the past set temperature are the actual power consumption and actual usage time of the target cooling air conditioner from the time when the target cooling air conditioner is set to the past set temperature to the time when the set closed space reaches the user's expected cooling temperature; Among them, in each training performed on the deep neural network, the predicted power consumption and predicted consumption time corresponding to a known past set temperature are used as two output contents of the deep neural network, and the user's expected refrigeration temperature, the past set temperature, the operating voltage and operating frequency of the target refrigeration air conditioner, various equipment data of the target refrigeration air conditioner, multiple internal environmental parameters of the set enclosed space corresponding to the past set temperature, and multiple external environmental parameters of the set enclosed space corresponding to the past set temperature are used as the input content of the deep neural network item by item. Completing this training also includes: the multiple internal environmental parameters of the set enclosed space corresponding to the past set temperature are the multiple internal environmental parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature, and the multiple external environmental parameters of the set enclosed space corresponding to the past set temperature are the multiple external environmental parameters of the set enclosed space when the target refrigeration air conditioner is set to the past set temperature.

3. The air conditioning energy-saving control method of the digital twin model according to claim 2, characterized in that: After performing a set number of multiple trainings on the deep neural network to obtain a deep neural network after the multiple trainings and outputting it as the digital twin model of the target refrigeration air conditioner, and the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner, the method further includes: A cloud computing storage node is used to store various model parameters of the digital twin model to complete the model storage of the digital twin model.

4. The air conditioning energy-saving control method of the digital twin model according to claim 2, characterized in that: After sending the reference set temperature corresponding to the predicted power consumption with the minimum value among the multiple predicted power consumptions as the priority energy-saving set temperature to the main controller of the target refrigeration air conditioner, the method further includes: An LED display array or a liquid crystal display screen is used to receive and display the priority energy-saving set temperature.

5. The air conditioning energy-saving control method of the digital twin model according to claim 2, characterized in that: After sending the reference set temperature corresponding to the predicted power consumption with the minimum value among the multiple predicted power consumptions as the priority energy-saving set temperature to the main controller of the target refrigeration air conditioner, the method further includes: The received priority energy-saving set temperature is wirelessly sent to a remote energy-saving management server using a frequency division duplex communication link or a time division duplex communication link.

6. The air conditioning energy-saving control method of the digital twin model according to claim 5, characterized in that: Obtain various equipment data of the target refrigeration air conditioner, wherein the various equipment data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioning number, cooling power, heating power, circulating air volume and frequency conversion mark of the target refrigeration air conditioner, including: when the frequency conversion mark of the target refrigeration air conditioner is 0B01, it indicates that the target refrigeration air conditioner is a variable frequency air conditioner; when the frequency conversion mark of the target refrigeration air conditioner is 0B00, it indicates that the target refrigeration air conditioner is a fixed frequency air conditioner.

7. The air conditioning energy-saving control method based on the digital twin model according to claim 5, characterized in that: Inputting the user's expected cooling temperature, the reference set temperature, the operating voltage and operating frequency of the target cooling air conditioner, various equipment data of the target cooling air conditioner, multiple internal environmental parameters of the set enclosed space, and multiple external environmental parameters of the set enclosed space into the digital twin model in parallel; Wherein, the digital twin model is run to obtain the predicted power consumption and predicted consumption time corresponding to the reference set temperature output by the digital twin model.

8. An air conditioning energy-saving control system based on a digital twin model, characterized in that: The system comprises a memory and one or more processors, the memory storing a computer program, the computer program being configured to be executed by the one or more processors to perform the following steps: Acquire various equipment data of the target refrigeration air conditioner, wherein the various equipment data of the target refrigeration air conditioner are energy efficiency ratio, applicable area, air conditioner number, cooling power, heating power, circulating air volume and frequency conversion mark of the target refrigeration air conditioner; Acquire multiple internal environmental parameters of a set closed space and multiple external environmental parameters of the set closed space, wherein the set closed space is a space in which the target refrigeration air conditioner performs temperature control, the multiple internal environmental parameters of the set closed space are the temperature, humidity, air pressure and space volume inside the set closed space, and the multiple external environmental parameters of the set closed space are the temperature, humidity, wind direction, wind speed and sunlight angle outside the set closed space; Performing a set number of multiple trainings on the deep neural network to obtain a deep neural network after the multiple trainings and outputting the deep neural network as the digital twin model of the target refrigeration air conditioner, wherein the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner; Evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain a plurality of reference set temperatures, input each reference set temperature into the digital twin model to obtain the predicted power consumption and predicted consumption duration corresponding to the reference set temperature, and obtain a plurality of predicted power consumption and a plurality of predicted consumption durations corresponding to the plurality of reference set temperatures, wherein the number of the plurality of reference set temperatures exceeds a set number threshold; Sending a reference set temperature corresponding to the predicted power consumption with the minimum value among the multiple predicted power consumptions as a priority energy-saving set temperature to a main controller of the target refrigeration air conditioner; Wherein, after receiving the priority energy-saving setting temperature, the main controller of the target refrigeration air conditioner automatically sets the priority energy-saving setting temperature as the refrigeration setting temperature of the target refrigeration air conditioner; Among them, when two sets of predicted power consumption with the same minimum value appear in multiple sets of predicted power consumption, the two reference set temperatures corresponding to the two sets of predicted power consumption with the same minimum value are used as the first energy-saving set temperature and the second energy-saving set temperature, the predicted consumption time corresponding to the first energy-saving set temperature and the second energy-saving set temperature are used as the first predicted consumption time and the second predicted consumption time, and the energy-saving set temperature corresponding to the predicted consumption time with the minimum value of the two predicted consumption time is used as the priority energy-saving set temperature; Among them, the digital twin model is used to intelligently predict the power consumed by the target refrigeration air conditioner when the closed space reaches the user's expected cooling temperature after the reference set temperature is set as the refrigeration set temperature of the target refrigeration air conditioner based on the user's expected cooling temperature, the reference set temperature, the operating voltage and operating frequency of the target refrigeration air conditioner, various equipment data of the target refrigeration air conditioner, multiple internal environmental parameters of the set closed space, and multiple external environmental parameters of the set closed space as the predicted power consumption corresponding to the reference set temperature, and the usage time of the target refrigeration air conditioner as the predicted consumption time corresponding to the reference set temperature.

9. An air conditioning energy-saving control system based on a digital twin model, characterized in that: The system comprises: The first acquisition device is used to obtain various equipment data of the target refrigeration air conditioner, wherein the various equipment data of the target refrigeration air conditioner are the energy efficiency ratio, applicable area, air conditioner number, cooling power, heating power, circulating air volume and frequency conversion mark of the target refrigeration air conditioner; The second acquisition device is used to obtain a plurality of internal environmental parameters of a set closed space and a plurality of external environmental parameters of a set closed space, wherein the set closed space is a space in which the target refrigeration air conditioner performs temperature control, the plurality of internal environmental parameters of the set closed space are the temperature, humidity, air pressure and space volume inside the set closed space, and the plurality of external environmental parameters of the set closed space are the temperature, humidity, wind direction, wind speed and sunlight angle outside the set closed space; A multi-layer construction device, used to perform a set number of multiple trainings on the deep neural network to obtain a deep neural network after the multiple trainings and output it as a digital twin model of the target refrigeration air conditioner, wherein the value of the set number is positively correlated with the compression ratio of the compressor of the target refrigeration air conditioner; A traversal analysis device is respectively connected to the first acquisition device, the second acquisition device and the multi-layer construction device, and is used to evenly divide the temperature adjustment range of the target refrigeration air conditioner to obtain a plurality of reference set temperatures, and input each reference set temperature into the digital twin model to obtain a predicted power consumption and a predicted consumption duration corresponding to the reference set temperature, and obtain a plurality of predicted power consumption and a plurality of predicted consumption durations corresponding to the plurality of reference set temperatures, wherein the number of the plurality of reference set temperatures exceeds a set number threshold; an energy-saving control device, connected to the ergodic analysis device and the main controller of the target refrigeration air conditioner, respectively, for sending a reference set temperature corresponding to the predicted power consumption with the minimum value among the multiple predicted power consumptions as a priority energy-saving set temperature to the main controller of the target refrigeration air conditioner; Wherein, after receiving the priority energy-saving setting temperature, the main controller of the target refrigeration air conditioner automatically sets the priority energy-saving setting temperature as the refrigeration setting temperature of the target refrigeration air conditioner; Among them, when two sets of predicted power consumption with the same minimum value appear in multiple sets of predicted power consumption, the two reference set temperatures corresponding to the two sets of predicted power consumption with the same minimum value are used as the first energy-saving set temperature and the second energy-saving set temperature, the predicted consumption time corresponding to the first energy-saving set temperature and the second energy-saving set temperature are used as the first predicted consumption time and the second predicted consumption time, and the energy-saving set temperature corresponding to the predicted consumption time with the minimum value of the two predicted consumption time is used as the priority energy-saving set temperature; Among them, the digital twin model is used to intelligently predict the power consumed by the target refrigeration air conditioner when the closed space reaches the user's expected cooling temperature after the reference set temperature is set as the refrigeration set temperature of the target refrigeration air conditioner based on the user's expected cooling temperature, the reference set temperature, the operating voltage and operating frequency of the target refrigeration air conditioner, various equipment data of the target refrigeration air conditioner, multiple internal environmental parameters of the set closed space, and multiple external environmental parameters of the set closed space as the predicted power consumption corresponding to the reference set temperature, and the usage time of the target refrigeration air conditioner as the predicted consumption time corresponding to the reference set temperature.

Images