Optimal target frequency calculation method based on load identification, air conditioner frequency control system and computer readable medium

Abstract

The invention relates to the technical field of smart home, and particularly discloses an optimal target frequency calculation method based on load identification, an air conditioner frequency control system and a computer readable medium. Historical time period operation data are collected through related sensors, and historical air conditioner equipment operation data are obtained through an interface of a calculation platform; calculating the optimal target frequency, storing the optimal target frequency in a big data platform, and training historical effective operation data of the air conditioning equipment to obtain a load identification model; if the air conditioning equipment is newly started, the load identification model is called, the optimal compressor target frequency of the starting is automatically searched for, frequency control is conducted, the problems of hysteresis quality, high energy consumption and the like existing in a traditional PID control method are solved, and effective temperature control is achieved; and meanwhile, the effective operation data after the new startup is updated, so that the load identification model can be optimized, the indoor temperature of the air conditioner refrigeration chamber can be quickly reduced and stabilized, the effects of reducing energy consumption and saving energy are achieved, and more comfortable use experience is provided for a user.

Description

Technical Field

[0001] The present invention relates to the field of smart home technology, and in particular to an optimal target frequency calculation method based on load identification, an air conditioning frequency control system and a computer-readable medium. Background Art

[0002] The compressor is an extremely important component of air-conditioning equipment. When the indoor temperature is higher than the set temperature, the control system sends a specified command to the compressor, and the compressor is started and operates according to the frequency sent by PID. The compressor frequency given by PID is the empirical value frequency obtained by designers through experiments in the laboratory under different working conditions. Its value is related to parameters such as indoor temperature, outdoor temperature, and set temperature. However, in the actual installation process of the air conditioner, various components of the air conditioner may be affected, making the collected data inaccurate. At the same time, according to the usage of different users, the air conditioner will have different degrees of loss. The PID control program lacks parameter adaptive capabilities and cannot meet the control requirements under different environmental conditions. The room temperature will fluctuate, which will affect the user experience and increase the energy consumption of the air conditioner. Therefore, it is particularly important to optimize the compressor frequency control program.

[0003] Generally speaking, most air-conditioning equipment compressor frequency control adopts PID control, and the frequency is adjusted only according to the experience value of the experimental equipment in the factory stage of the compressor, and the power consumption of the refrigeration system is rarely considered. At the same time, different users have different equipment usage habits and installation conditions, so the user's actual working conditions will not match the compressor frequency parameters controlled by PID, which will eventually lead to the user's perceived temperature being too high or too low, and the ideal room temperature value will not be reached for a long time, which greatly affects the user's experience and also causes the energy consumption of the air-conditioning system to increase.

[0004] Although there is a document entitled "Research on Energy-Saving Control Method of Air Conditioner Based on Q-learning and Adaptive Network Constraints", which discloses the "trial and error principle" of Q-learning reinforcement learning to find the optimal control in different environments, quickly accumulate training sample data, generalize the application scenarios of the control method, and construct adaptive network constraints by combining the radial basis deep neural network model to achieve the dynamic nature of the control strategy; however, the optimal control of different air-conditioning equipment in the same environment varies greatly. This type of air-conditioning energy-saving control method has a low matching degree for control strategies in different environments, has a low universality for the use of air-conditioning equipment, and fails to achieve the purpose of energy saving in some environments.

[0005] Another example is the patent document with patent number CN202410018124.6, which discloses a frequency control method for variable frequency air conditioners based on a time fuzzy algorithm. The frequency of the variable frequency air conditioner compressor is controlled according to the change in the cooling or heating load of the variable frequency air conditioner and the change in the indoor temperature. The frequency is increased or decreased according to the set first frequency adjustment amount with a natural number N fuzzy control time unit t as the time interval, thereby improving the compatibility of the compressor frequency with the room temperature, reducing room temperature fluctuations, and improving consumer comfort. However, this invention does not take into account the differences between different devices, and different devices may not be able to achieve the purpose of energy saving by using this frequency control method. Summary of the invention

[0006] The purpose of the present invention is to provide an optimal target frequency calculation method based on load identification, an air conditioning frequency control system and a computer-readable medium for the frequency control or energy-saving control system of air-conditioning equipment in the prior art. Due to different air-conditioning working conditions or different air-conditioning equipment, there is a problem that the matching degree between the control strategy and the environment is not high, resulting in poor energy-saving effect.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] A method for calculating an optimal target frequency based on load identification comprises the following steps:

[0009] S1. Obtaining the operation data of the air conditioner in cooling mode each time it is turned on in a historical time period, wherein the operation data includes: switch flag, operation mode, control windshield, actual windshield, set temperature, outdoor temperature, indoor temperature, and compressor frequency value;

[0010] S2. Acquire first operating data of the air-conditioning device that meets the temperature reaching condition from the operating data of a single startup;

[0011] S3, selecting second operation data that meets the requirements of complete startup from the first operation data;

[0012] S4, selecting from the second operation data the third operation data of the air conditioner having a constant windshield, a constant temperature and a compressor without stopping;

[0013] S5. Obtaining a compressor frequency value list for a single startup period from the third operation data, and calculating a maximum value of the compressor frequency value for a single startup according to the compressor frequency value list, wherein the maximum value is an optimal target frequency;

[0014] S6. According to the frequency hopping point interval of the air conditioner, filter out the optimal target frequency in the frequency hopping point interval.

[0015] The optimal target frequency calculation method based on load identification described in the present invention can filter out the operating data with low reference value when the set temperature is not reached; then filter out the second operating data that meets the standard startup time to avoid the collection of incomplete startup operating data, and by screening the third operating data with unchanged windshield, unchanged temperature and non-stop compressor, ensure that the collected operating data is data of normal operation of the air conditioner, calculate the corresponding compressor frequency for each startup of each device and under different working conditions to characterize the optimal target frequency required for the current working condition, and obtain more real and reliable operating data under more actual working conditions, and then determine the optimal target frequency value; then filter out the optimal target frequency at the frequency hopping point, and exclude the operating data that may cause the compressor to fail to operate normally as a training parameter; through steps S1 to S6, under different air-conditioning working conditions or different air-conditioning equipment, the load identification platform can obtain a load identification model that is more in line with the actual working conditions through the training of the optimal target frequency to control the frequency of the air-conditioning equipment, improve the matching degree of the control strategy and the environment and the air-conditioning equipment, and achieve rapid and stable cooling and stabilization of the indoor temperature under different working conditions, so as to achieve the effect of reducing energy consumption and saving energy, and provide a comfortable experience for users.

[0016] Preferably, the optimal target frequency calculation method based on load identification described in the present invention further includes the following steps after S6: using the operating data corresponding to the operating conditions and the optimal target frequency as training data for subsequent load classification and load identification, and training to obtain a load identification model.

[0017] As the preferred solution of the present invention, and the corresponding operating data; finally, by saving the optimal target frequency parameters of the big data platform equipment under all working conditions, the load identification model is trained. Under the same load condition of the new startup, the air-conditioning compressor automatically finds the optimal target frequency parameters by calling the load identification model and performs precise and efficient control, which solves the problems of lag and high energy consumption in the traditional PID control method. At the same time, the new startup data is used to update the parameters of the entire load identification model to achieve model optimization.

[0018] Preferably, in the optimal target frequency calculation method based on load identification described in the present invention, the temperature reaching condition is: indoor temperature - set temperature ≤ 1°C.

[0019] As a preferred solution of the present invention, by setting the temperature reaching condition to indoor temperature - set temperature ≤ 1°C and filtering the first operating data into a more accurate range, the reliability of load identification model training can be further improved.

[0020] Preferably, in the optimal target frequency calculation method based on load identification described in the present invention, S2 also includes the following steps: determining whether the operating data of a single startup meets the temperature reaching condition; if the temperature reaching condition is met, executing S3; if the temperature reaching condition is not met, returning to execute step S1 to obtain the operating data for the next startup.

[0021] As a preferred solution of the present invention, by judging whether the operating data of a single startup meets the temperature reaching condition, if it does not meet the temperature reaching condition, returning to execute step S1, it can ensure that the operating data that does not meet the temperature reaching condition is screened out each time the machine is started, avoiding a large amount of redundant data, and improving the accuracy of the first operating data screening, thereby improving the matching degree of the optimal target frequency to the environment and air-conditioning equipment, and further improving the reliability of load identification model training.

[0022] Preferably, in the optimal target frequency calculation method based on load identification described in the present invention, S3 also includes the following steps: according to the power-on status of the air conditioner and the compressor, judging whether the first operating data of a single startup belongs to a complete startup, the complete startup includes air conditioner startup and compressor startup; if the first operating data of a single startup belongs to a complete startup, executing step S4; if the first operating data of a single startup does not belong to a complete startup, returning to execute step S1 to obtain the operating data of the next startup.

[0023] After the air conditioner has been turned on for a period of time, the compressor, as a preferred solution of the present invention, determines whether the first operating data of a single startup belongs to a complete startup. If the first operating data of a single startup does not belong to a complete startup, it returns to execute step S1 to obtain the operating data of the next startup. This can ensure that each startup operating data that does not meet the requirements of a complete startup is screened out, avoid a large amount of redundant data, improve the accuracy of the second operating data screening, and then improve the matching degree of the optimal target frequency to the environment and air-conditioning equipment, and further improve the reliability of load identification model training.

[0024] Preferably, in the optimal target frequency calculation method based on load identification described in the present invention, S6 also includes the following steps: determining whether the optimal target frequency is in the frequency hopping point interval of the air-conditioning equipment; if the optimal target frequency is outside the frequency hopping point interval, retaining the optimal target frequency; if the optimal target frequency is in the frequency hopping point interval, returning to execute step S1 to obtain the operating data for the next startup.

[0025] As a preferred solution of the present invention, by judging whether the optimal target frequency is in the frequency hopping point interval of the air-conditioning equipment; if the optimal target frequency is in the frequency hopping point interval, returning to execute step S1, it can ensure that the obtained optimal target frequency can meet the operation of the compressor, avoid the situation where the compressor cannot operate normally at the optimal target frequency, ensure the effectiveness of the optimal target frequency, and further improve the reliability of load identification model training.

[0026] To achieve the purpose of the present invention, another technical solution adopted by the present invention is:

[0027] An air conditioning frequency control system, comprising: an Internet of Things air conditioning device, a big data storage platform, a computing platform, a load identification platform, a calling platform and an air conditioning control module;

[0028] The Internet of Things air-conditioning device is used to upload working status data to the big data storage platform; the big data storage platform is used to store the operating data of the Internet of Things air-conditioning device; the calculation platform is used to screen and calculate the operating data, so as to obtain the optimal target frequency; the load identification platform is used to train the selected operating data and the corresponding optimal target frequency, so as to obtain a load identification model, and optimize the load identification model according to the newly started operating data; the calling platform can call the load identification model through the air-conditioning control program, so as to determine the optimal target frequency for this startup; the air-conditioning control module is used to control the operation of the Internet of Things air-conditioning device according to the optimal target frequency.

[0029] The air conditioning frequency control system described in the present invention can be trained to obtain a load identification model based on the operating data of the selected historical time period and the optimal target frequency. For the new startup of the equipment, the air conditioning control program calls the load identification model, automatically finds the optimal compressor target frequency for this startup, and controls it to achieve effective temperature control, control the stability of the indoor temperature, reduce the energy consumption of the air conditioning, and achieve the purpose of energy saving.

[0030] To achieve the purpose of the present invention, another technical solution adopted by the present invention is:

[0031] A computer readable medium, wherein when the computer program is executed by a processor, the optimal target frequency calculation method of the present invention is implemented.

[0032] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0033] 1. The optimal target frequency calculation method, through steps S1 to S6, under different air-conditioning working conditions or different air-conditioning equipment, the load identification platform can obtain a load identification model that is more in line with the actual working conditions through the training of the optimal target frequency to control the frequency of the air-conditioning equipment, improve the matching degree between the control strategy and the environment and the air-conditioning equipment, and realize rapid and stable cooling of the indoor temperature under different working conditions, achieve the effect of reducing energy consumption and saving energy, and provide users with a comfortable experience;

[0034] 2. The air conditioning frequency control system can be trained to obtain a load identification model based on the operating data of the selected historical time period and the optimal target frequency. For the new startup of the equipment, the air conditioning control program calls the load identification model to automatically find the optimal compressor target frequency for this startup and control it to achieve effective temperature control, control the stability of indoor temperature, reduce air conditioning energy consumption, and achieve energy saving. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flow chart of the optimal target frequency calculation method based on load identification of the present invention;

[0036] Figure 2 It is a schematic diagram of the module structure of the air-conditioning frequency control system of the present invention;

[0037] Figure 3 It is a schematic diagram of the execution flow of the air-conditioning frequency control system of the present invention. DETAILED DESCRIPTION

[0038] The present invention will be described in detail below in conjunction with the accompanying drawings.

[0039] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0040] Embodiment 1:

[0041] refer to Figure 1 As shown, this embodiment provides an optimal target frequency calculation method based on load identification, comprising the following steps:

[0042] S1. Obtaining the operation data of the air conditioner in cooling mode each time it is turned on in a historical time period, wherein the operation data includes: switch flag, operation mode, control windshield, actual windshield, set temperature, outdoor temperature, indoor temperature, and compressor frequency value;

[0043] S2. Acquire first operating data of the air-conditioning device that meets the temperature reaching condition from the operating data of a single startup;

[0044] S3, selecting second operation data that meets the requirements of complete startup from the first operation data;

[0045] S4, selecting from the second operation data the third operation data of the air conditioner having a constant windshield, a constant temperature and a compressor without stopping;

[0046] S5. Obtaining a compressor frequency value list for a single startup period from the third operation data, and calculating a maximum value of the compressor frequency value for a single startup according to the compressor frequency value list, wherein the maximum value is an optimal target frequency;

[0047] S6. According to the frequency hopping point interval of the air conditioner, filter out the optimal target frequency in the frequency hopping point interval.

[0048] In the present invention, Figure 1 As shown, the operating data of the historical time period is collected by sensors of relevant dimensions when the air conditioner is running, where the dimensions are understood to be factors that affect the load of the air conditioner operation, such as the difference between the indoor temperature and the set temperature.

[0049] It should be noted that the first, second and third in the present invention are only distinguished for the purpose of clear description and are not interpreted in numerical or sequential order. The first operating data is understood as follows: Figure 1 The data shown meets the temperature reaching condition, for example, the temperature reaching condition can be set to: indoor temperature - set temperature ≤ 1°C; specifically, S2 also includes the following steps: determining whether the operating data of a single startup meets the temperature reaching condition; if it meets the temperature reaching condition, executing S3; if it does not meet the temperature reaching condition, returning to execute step S1 to obtain the operating data for the next startup.

[0050] After obtaining the first operating data that meets the temperature reaching condition as described in step S2, the second operating data that meets the complete startup condition can be selected by judging the temperature reaching time period. Specifically, the S3 also includes the following steps: judging whether the first operating data of a single startup belongs to a complete startup according to the startup status of the air conditioner and the compressor, and the complete startup includes air conditioner startup and compressor startup; if the first operating data of a single startup belongs to a complete startup, executing step S4; if the first operating data of a single startup does not belong to a complete startup, returning to execute step S1 to obtain the operating data for the next startup.

[0051] Specifically, after S6, the following steps are also included: using the operating data corresponding to the working condition and the optimal target frequency as training data for subsequent load classification and load identification, and training to obtain a load identification model.

[0052] Specifically, S6 also includes the following steps: determining whether the optimal target frequency is in the frequency hopping point interval of the air-conditioning equipment; if the optimal target frequency is outside the frequency hopping point interval, retaining the optimal target frequency; if the optimal target frequency is in the frequency hopping point interval, returning to execute step S1 to obtain the operating data for the next startup.

[0053] Embodiment 2:

[0054] refer to Figure 2 , Figure 3 As shown, this embodiment discloses an air-conditioning frequency control system, including: an Internet of Things air-conditioning device, a big data storage platform, a computing platform, a load identification platform, a calling platform and an air-conditioning control module; the Internet of Things air-conditioning device is used to upload working status data to the big data storage platform; the big data storage platform is used to store the operating data of the Internet of Things air-conditioning device; the computing platform is used to screen and calculate the operating data, so as to obtain the optimal target frequency; the load identification platform is used to train the selected operating data and the corresponding optimal target frequency, so as to obtain a load identification model, and optimize the load identification model according to the newly started operating data; the calling platform can call the load identification model through the air-conditioning control program, so as to determine the optimal target frequency of this startup; the air-conditioning control module is used to control the operation of the Internet of Things air-conditioning device according to the optimal target frequency.

[0055] The air conditioning frequency control system described in the present invention can be trained to obtain a load identification model based on the operating data of the selected historical time period and the optimal target frequency. For the new startup of the equipment, the air conditioning control program calls the load identification model, automatically finds the optimal compressor target frequency for this startup, and controls it to achieve effective temperature control, control the stability of the indoor temperature, reduce the energy consumption of the air conditioning, and achieve the purpose of energy saving.

[0056] It should be noted that the single startup described in the present invention is understood as a startup of the air-conditioning equipment in a historical time period, and each startup is understood as each single startup; the current startup is understood as the startup of the air-conditioning equipment for operation.

[0057] Embodiment 3:

[0058] refer to Figure 1 As shown, this embodiment discloses a computer-readable medium on which a computer program is stored. When the computer program is executed by a processor, the optimal target frequency calculation method described in Example 1 is implemented.

[0059] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for calculating the optimal target frequency based on load identification, characterized in that: The following steps are involved: S1. Obtaining the operation data of the air conditioner in cooling mode each time it is turned on in a historical time period, wherein the operation data includes: switch flag, operation mode, control windshield, actual windshield, set temperature, outdoor temperature, indoor temperature, and compressor frequency value; S2. Acquire first operating data of the air-conditioning device that meets the temperature reaching condition from the operating data of a single startup; S3, selecting second operation data that meets the requirements of complete startup from the first operation data; S4, selecting from the second operation data the third operation data of the air conditioner having a constant windshield, a constant temperature and a compressor without stopping; S5. Obtaining a compressor frequency value list for a single startup period from the third operation data, and calculating a maximum value of the compressor frequency value for a single startup according to the compressor frequency value list, wherein the maximum value is an optimal target frequency; S6. According to the frequency hopping point interval of the air conditioner, filter out the optimal target frequency in the frequency hopping point interval.

2. The optimal target frequency calculation method based on load identification according to claim 1 is characterized in that: The step S6 also includes the following steps: using the operating data corresponding to the working condition and the optimal target frequency as training data for subsequent load classification and load identification, and training to obtain a load identification model.

3. The optimal target frequency calculation method based on load identification according to claim 1 is characterized in that: The temperature reaching condition is: indoor temperature - set temperature ≤ 1°C.

4. The optimal target frequency calculation method based on load identification according to claim 1 is characterized in that: The S2 also includes the following steps: determining whether the operating data of a single startup meets the temperature reaching condition; if the temperature reaching condition is met, executing the S3; if the temperature reaching condition is not met, returning to execute step S1 to obtain the operating data for the next startup.

5. The optimal target frequency calculation method based on load identification according to claim 1 is characterized in that: The S3 also includes the following steps: judging whether the first operating data of a single startup belongs to a complete startup according to the startup status of the air conditioner and the compressor, wherein the complete startup includes the startup of the air conditioner and the startup of the compressor; if the first operating data of a single startup belongs to a complete startup, executing step S4; if the first operating data of a single startup does not belong to a complete startup, returning to executing step S1 to obtain the operating data of the next startup.

6. The optimal target frequency calculation method based on load identification according to claim 1 is characterized in that: The S6 also includes the following steps: determining whether the optimal target frequency is within the frequency hopping point interval of the air-conditioning equipment; if the optimal target frequency is outside the frequency hopping point interval, retaining the optimal target frequency; if the optimal target frequency is within the frequency hopping point interval, returning to execute step S1 to obtain the operating data for the next startup.

7. An air conditioning frequency control system, characterized in that: A method for calculating an optimal target frequency according to any one of claims 1 to 6, comprising: an Internet of Things air conditioning device, a big data storage platform, a computing platform, a load identification platform, a calling platform and an air conditioning control module; The Internet of Things air-conditioning device is used to upload working status data to the big data storage platform; the big data storage platform is used to store the operating data of the Internet of Things air-conditioning device; the calculation platform is used to screen and calculate the operating data, so as to obtain the optimal target frequency; the load identification platform is used to train the selected operating data and the corresponding optimal target frequency, so as to obtain a load identification model, and optimize the load identification model according to the newly started operating data; the calling platform can call the load identification model, so as to determine the optimal target frequency for this startup; the air-conditioning control module is used to control the operation of the Internet of Things air-conditioning device according to the optimal target frequency.

8. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the optimal target frequency calculation method according to any one of claims 1 to 7 is implemented.

Images