TWI670451B - Power saving control method of air conditioner in the medium and large space and system thereof - Google Patents

Abstract

An energy-saving control method for an air-conditioning device in a medium and large-sized space, the air-conditioning device comprising a compressor and a controller, the control method for the controller to determine an indoor comfort index target according to current moment and season information, and according to the The indoor comfort index target and a current humidity provided by an indoor hygrometer determine a target temperature, and the controller obtains a first difference between a current temperature provided by an indoor thermometer and the target temperature, and obtains the current moment. a second difference between the temperature and the temperature of the previous moment provided by the indoor thermometer, and an error difference between the second difference and the first difference, and the difference between the first difference and the error difference A control parameter is obtained to control the operation of the compressor according to the control parameter, and repeat the above steps to achieve energy saving.

Description

Energy-saving control method and system for air-conditioning device in medium and large space

The invention relates to a control method for an air conditioner, in particular to an energy-saving control method for an air conditioner in a medium and large space.

Generally, the indoor air conditioner is controlled by the user setting a desired temperature to the air conditioner through the control panel or the remote controller, and then the air conditioner controls the interior of the air conditioner based on a difference between the measured indoor temperature and the user set temperature. The compressor is operated to achieve the temperature set by the user. However, in addition to relying on the user's settings, and regardless of the number of indoors becoming more or less, the air conditioner will not automatically adjust its operation mode to reduce the room temperature or increase the room temperature, which will easily lead to more indoor space. Insufficient air-conditioning or cold due to small people, and air-conditioning devices can not adjust their operation mode in a timely manner to save energy, consuming and wasting power in vain, and does not meet the needs of today's environmental protection.

Therefore, the object of the present invention is to provide an energy-saving control method for a medium- and large-sized air-conditioning device capable of dynamically providing an optimal room temperature and timely energy saving according to the environment, time, and number of users in a medium-to-large space. And system.

Therefore, the present invention relates to an energy-saving control method for an air conditioner in a medium and large space, the air conditioner comprising a compressor and a controller electrically coupled to the compressor, and the controller is further connected to an indoor thermometer and an indoor hygrometer Coupling, the control method comprises the following steps: (A) a target temperature decision unit in the controller determines an indoor comfort index target according to the current time and season information, and according to the indoor comfort index target and the indoor The humidity provided by the hygrometer obtains a target temperature; (B) an artificial intelligence control unit in the controller obtains a first difference between a current temperature provided by the indoor thermometer and the target temperature, and obtains a a second difference between the lower temperature and the temperature of the previous moment provided by the indoor thermometer, and an error difference between the second difference and the first difference, and based on the first difference and the error difference, Obtaining a control parameter and controlling the operation of the compressor according to the control parameter; and (C) repeating the above steps (A) and (B).

In some embodiments of the present invention, in step (A), the target temperature decision unit is a first fuzzy operation module, which determines the indoor comfort index target according to the current time and season information. And determining the target temperature according to an indoor comfort index formula related to temperature and humidity, the indoor comfort index target, and the current humidity.

In some embodiments of the present invention, in addition to the current time and season information, the first fuzzy computing module determines the indoor comfort index target according to the current flow information of the current flow.

In some embodiments of the present invention, in step (A), the target temperature decision unit includes an indoor comfort index neural network model and a target temperature calculation module, and the indoor comfort index neural network The model is a neural network model that has been trained in advance using one-time data and one-season data, so that the indoor comfort index neural network model can determine the indoor comfort index target according to the current moment and seasonal information input. And the target temperature calculation module obtains the target temperature according to an indoor comfort index formula related to temperature and humidity, the indoor comfort index target, and the current humidity.

In some embodiments of the present invention, the indoor comfort index neural network model is a neural network model trained in advance using the time data, the seasonal data, and the one-person data, so that the indoor comfort index-like nerve The network model can determine the indoor comfort index target based on the current moment, season and flow information entered.

In some embodiments of the present invention, in step (B), the artificial intelligence control unit is a second fuzzy operation module, and the fuzzy operation is performed according to the first difference and the error difference to obtain the The control parameter includes a control time length and a work cycle, and the previous moment temperature refers to an indoor temperature measured by the indoor thermometer when the control time length ends.

In some embodiments of the present invention, in step (B), the artificial intelligence control unit includes a neural network control model, and the neural network control model is a preset that has previously utilized the first difference. a range, a predetermined range of the error difference, and a trained neural network model of the predetermined range of the control parameter, such that the neural network control model can be based on the input first difference and the error difference The control parameter is obtained by performing an operation, wherein the control parameter includes a control time length and a work cycle, and the previous moment temperature refers to an indoor temperature measured by the indoor thermometer when the control time length ends.

In some embodiments of the present invention, the air conditioner is further electrically coupled to a near-end server to obtain the time data and the season data from the near-end server, and the indoor thermometer and the indoor hygrometer further Providing the measured indoor temperature and indoor humidity to the controller through the near-end server, and the near-end server communicates with a remote server through the Internet of Things, so that the remote server issues the controller A control command can be transmitted to the controller through the near-end server, and the controller controls the operation of the compressor according to the control command.

Furthermore, the energy-saving control system for a medium- and large-sized air-conditioning device in the above-described energy-saving control method comprises an air-conditioning device, an indoor thermometer and an indoor hygrometer, the air-conditioning device comprising a compressor and a compressor The controller is coupled, and the controller is further electrically coupled to the indoor thermometer and the indoor hygrometer, and can perform the energy-saving control method of the air-conditioning device in the medium and large space as described above by the controller.

In some embodiments of the present invention, the energy-saving control system of the air conditioning device in the medium and large space further includes a near-end server and a remote server, and the air-conditioning device further includes an electrical coupling with the controller. a communication interface, and the controller is electrically coupled to the near-end server through the communication interface to obtain the time data and the season data from the near-end server, and the indoor thermometer and the indoor hygrometer also pass through the near The end server provides the measured indoor temperature and indoor humidity to the controller, and the near-end server communicates with the remote server through the Internet of Things, so that the remote server issues a control command to the controller. The controller can be transmitted to the controller through the near-end server and the communication interface, so that the controller controls the operation of the compressor according to the control command.

The utility model has the advantages that the controller of the air conditioner dynamically generates the optimal control parameter according to the current time, season, flow of people and indoor temperature and humidity, and dynamically controls the compression according to the control parameter. The machine operates, and the indoor environment in which the air conditioner is installed can produce the optimum temperature that is most comfortable and at the same time achieve energy saving.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same reference numerals.

1 is a main flow chart of an embodiment of an energy-saving control method for an air-conditioning device in a large space according to the present invention, and the embodiment is an energy-saving control system for an air-conditioning device in a medium and large space as shown in FIG. 2 ( The energy-saving control system is mainly referred to as an air-conditioning device 1, an indoor thermometer 3, and an indoor hygrometer 4. The air conditioner 100 mainly refers to, but is not limited to, any air conditioner or air conditioner apparatus used in a medium and large space, and mainly includes an (air conditioner) compressor 1, and a controller 2 electrically coupled to the compressor 1. The controller 2 is electrically coupled to the indoor thermometer 3 and the indoor hygrometer 4. The indoor thermometer 3 is disposed at a return air inlet (not shown) of the air conditioner 100 to detect a return air temperature, and the return air temperature generally represents an indoor temperature of an environment in which the air conditioner 100 is located. The indoor hygrometer 4 is also disposed at the air return opening of the air conditioner 100 to detect a return air humidity, and the return air humidity generally represents the indoor humidity of the environment in which the air conditioner 100 is located, but is not limited thereto. The indoor thermometer 3 and the indoor hygrometer 4 can also be set in other locations suitable for detecting indoor temperature and humidity according to actual needs, and transmit the measured indoor temperature through a wired or wireless communication network (such as a regional network). The indoor humidity is given to the controller 2.

As shown in FIG. 3, an embodiment of the controller 2 mainly includes a target temperature decision unit 21 and an artificial intelligence control unit 23. In order to make the air conditioning device 100 have the most comfortable temperature and timely energy saving, when the air conditioning device 100 starts to operate, as in step S1 of FIG. 1, the target temperature decision unit 21 of the controller 2 according to the current moment. , season and flow information, determine a target temperature.

Specifically, in this embodiment, a first implementation manner of the target temperature decision unit 21 may be a first fuzzy operation module, which can determine a fuzzy operation according to current moments, seasons, and flow information. The indoor comfort index target, and the target temperature is determined according to an indoor comfort index formula related to temperature and humidity, the indoor comfort index target, and the current humidity. That is, the first fuzzy computing module will be based on pre-acquired time data, one-season data, one-person data, and a pre-obtained indoor comfort index (also known as Predicted Mean Vote, hereinafter referred to as PMV). ), in which a plurality of (three) PMV fuzzy rule tables corresponding to different numbers of people are established, as shown in Table 1, Table 2 and Table 3 below. Among them, PMV is an indicator adopted by American Society of Heating, Refrigerating and Air-Conditioning Engineers ASHRAE, which is mainly used to express the human body's feelings of cold and heat in the environment. In addition, according to the linguistic variables (opt, hop, medium, hhp, pt) corresponding to the peaks and peaks of the stream, opt represents the peak, hop represents the semi-offer, medium represents the ordinary, hhp represents the semi-spike, pt Representing the peak; according to the language variables (cold, cool, medium, warm, hot) corresponding to the degree of cold and heat of the weather, cold represents cold, cool represents cool, medium represents ordinary, warm represents warm, hot represents heat; According to the corresponding language variables (few, medium, many), the few represent the small, the medium represents the ordinary, the many represents the many; the language variables corresponding to the indoor comfort (PMV) (vc, cold, cool, comfortable, warming, Hot, vh), vc is very cold, cold is cold, cool is cool, comfort is comfortable, warming is warm, hot is hot, vh is very hot.                    Time opt hop medium hhp pt cold vh vh hot warming comfortable cool vh hot warming comfortable cool medium hot warming comfortable cool cold warm warm comfortable cold cold cold hot comfortable cool cold cold cold Table 1 (person flow = few)        Opt hop medium hhp pt cold hot warm warming comfortable cool cool hot warming comfortable cool cool medium warming comfortable cool cool cold warm comfortable cool cool cold cold hot cool cool cold cold vc Table 2 (person flow = medium)        Time opt hop medium hhp pt cold warming warming comfortable cool cold cool warming comfortable cool cold cold medium comfortable cool cold cold vc warm cool cold cold vc vc hot cold cold vc vc vc table 3 (person flow = many)  

And the first fuzzy operation module selects one of the PMV fuzzy rule tables according to the current flow information, for example, the flow = many, for example, the above table 3, and according to the selected PMV rule table (ie, the table) 3) and the current moment, such as pt (spike) and season, such as cool (cool), from Table 3 determines a PMV target is cold (cold).

Then, the first fuzzy computing module (ie, the target temperature determining unit 21) obtains the target temperature according to the PMV target and a current humidity provided by the indoor hygrometer 3; specifically, the first fuzzy computing module is configured according to A PMV formula related to indoor temperature and indoor humidity ( ), the PMV target (ie, PMV _act in the formula) and the current indoor humidity (ie, RH (relative humidity) in the formula) determine the target temperature (ie, T _a in the formula) corresponding to the PMV target. That is, the first fuzzy computing module first converts the PMV target into a clear value by, for example, the center of gravity defuzzification method, and then converts the PMV target (value) converted to a clear value and the indoor humidity into the PMV formula. The target temperature is reversed and the target temperature is output to the artificial intelligence control unit 22. It is worth mentioning that this embodiment can also be used according to actual needs of the environment. For example, when the flow of people in the environment is not changed or the flow of people tends to be fixed, the flow of people can be omitted without considering the flow of people, and only one of the above PMVs is established. A fuzzy rule table, such as Table 1. Therefore, the first fuzzy computing module can determine the indoor comfort index target only by performing a fuzzy operation according to the current time and season information.

In addition, as shown in FIG. 4, the second embodiment of the target temperature decision unit 21 may also include an indoor comfort level index (hereinafter referred to as PMV) neural network model 215 and a target temperature calculation module 216, wherein The PMV-like neural network model 215 is a recurrent neural network (RNN) model that has been trained in advance using the above-mentioned time data, the season data, and the person flow data. The recurrent neural network is two kinds of artificial nerves. The general term for the network, one is the recurrent neural network, also known as the recurrent neural network, such as the Long Short-Term Memory (LSTM) time recurrent neural network, and the other is structural recursion. Recursive neural network. Since the training method of the recurrent neural network is already a well-known technique and is not the focus of the present case, it will not be described here. Therefore, the PMV-like neural network model 215 can output the PMV target (value) according to the current moment, season and flow information input, and the target temperature calculation module 216 is based on the above-mentioned PMV related to indoor temperature and indoor humidity. formula( ), the PMV target (value) and the indoor humidity reversely determine the target temperature. It is worth mentioning that this embodiment can also be used according to the actual use environment requirements. For example, when the flow of people in the environment is not changed or the flow of people tends to be fixed, the flow data is omitted. Therefore, the PMV-like neural network model 215 only needs to be The training is performed in advance using the information of the time and the seasonal data, and the trained PMV-based neural network model 215 can calculate the PMV target (value) according to the current moment and season information input. .

Then, as shown in step S2 of FIG. 1 and FIG. 3, the artificial intelligence control unit 23 obtains a first difference between a current temperature (indoor temperature) and the target temperature provided by the indoor thermometer 4, and obtains the current difference. a second difference between the temperature and the previous temperature, and an error difference between the second difference and the first difference, and obtaining a control parameter according to the first difference and the error difference, The compressor operation is controlled in accordance with the control parameter. In this embodiment, a first implementation manner of the artificial intelligence control unit 23 may be a second fuzzy operation module, which can perform the fuzzy operation according to the first difference and the error difference. Control parameters, in this embodiment, the control parameters may include, for example, a control time length and a work cycle, but are not limited thereto. And the temperature of the previous moment may refer to the indoor temperature measured by the indoor thermometer 4 at the end of the control time period of the previous control period or before the air conditioner 100 is turned off, or refers to, for example, 5 minutes, 10 before the current time. The indoor temperature measured by the indoor thermometer 4 at a fixed or unfixed time point such as minutes or half an hour, but not limited thereto.

Specifically, the second fuzzy computing module pre-determines a predetermined range (eg, +4 degrees to -6 degrees) of the first difference between the current temperature (indoor temperature) and the target temperature, and the error difference. a predetermined range of values (eg, +4 degrees to -4 degrees), a predetermined range of the length of the control time (eg, 0 to 14 minutes), and a predetermined range of the duty cycle (eg, 0% to 100%) A control time fuzzy rule table and a work cycle fuzzy rule table as shown below are established. Among them, according to the language variables corresponding to the first difference (mcold, cold, cool, medium, warm, hot, mhot), mcold represents the coldest, cold represents cold, cool represents cool, medium represents ordinary, warm represents warmth. Hot represents hot, and mhot stands for hottest; according to the language variables (idrop, drop, down, medium, rise, soar, isoar) corresponding to the error difference, idrop represents immediate cooling, drop represents rapid cooling, and down represents cooling. , medium stands for ordinary, rise stands for warming, soar stands for rapid warming, and isoar stands for immediate warming. According to the control rules (moment, short, medium, long, faraway) corresponding to the length of control time, moment represents a moment, short represents short time, medium represents middle, long represents long time, faraway represents a long time; According to the corresponding control rules (close, low, medium, high, very high), close represents a work cycle of 0% or close to 0%, low represents a short work cycle, and medium represents a work cycle of half, such as 50%, high Represents a long work cycle, very high represents a long work cycle.             Cold differential cool medium warm hot m hot i drop faraway faraway far medium long medium medium medium drop faraway far medium long medium short short short down faraway long medium Moment short short short medium long medium moment Moment moment short long rise Medium medium short Moment short long faraway soar short short short short long faraway faraway i soar short short short long faraway faraway faraway control time fuzzy rule table           The difference in the medium medium medium high medium low medium medium medium medium high high low low low close close low close low low high low close Low low medium medium medium high very high soar low low high medium high very high very high i soar low high high high very high very high  

Then, the second fuzzy operation module is configured according to the language variable corresponding to the first difference obtained in the moment, for example, “cold”, and the language variable corresponding to the error difference, for example, “down”, from the control time listed above The fuzzy rule table and the work cycle fuzzy rule table obtain a control time rule, such as "long" and a work cycle rule, such as "low". The first difference is obtained by, for example, the second fuzzy operation module subtracting the current temperature from the current temperature by using a subtractor (not shown), or by a subtractor in the fuzzy control unit 23 (Fig. The first difference is obtained by subtracting the current temperature from the current temperature, and the first difference is further provided to the second fuzzy operation module. Similarly, the second difference is also the second fuzzy operation. The module obtains the current temperature by subtracting the current temperature from the current subtractor, or the subtractor in the fuzzy control unit 23 subtracts the current temperature from the current temperature to obtain the second difference, and then provides the second difference. The second difference is given to the second fuzzy operation module. Therefore, the second fuzzy operation module can use the subtractor to subtract the first difference from the second difference to obtain the error difference.

Then, the second fuzzy operation module determines the length of the control time according to the control time rule "long" and the duty cycle rule "low", for example, 12 minutes and the work cycle, for example, 30%, that is, the second The fuzzy computing module converts the control time rule "long" and the duty cycle rule "low" into a control time length and the duty cycle expressed by an explicit numerical value using, for example, a center of gravity defuzzification method.

In addition, a second implementation manner of the artificial intelligence control unit 23 may also be a type of neural network control model, and the neural network control model is that the preset range of the first difference value has been previously utilized (for example, + 4 degrees to -6 degrees), the preset range of the error difference (for example, +4 degrees to -4 degrees), the preset range of the control time length (for example, 0 to 14 minutes), and the duty cycle Pre-defined range (eg 0%~100%) and other well-trained neural networks, such as recurrent neural networks (RNN), but not limited to this, and because the training method of recurrent neural networks is already known Technology is not the focus of this case, so it will not be repeated here. Therefore, the trained neural network control model can generate and output the control parameter according to the first difference and the error difference input at the moment, that is, the length of time and the duty cycle.

Thereby, the artificial intelligence control unit 23 can control the operation of the compressor 1 according to the control time length and the duty cycle until the control time length ends (at this time, the artificial intelligence control unit 23 obtains the indoor thermometer 4 to measure The indoor temperature, that is, the previous moment temperature), and repeating the above steps S1 and S2 until the air conditioner 100 is turned off (at this time, the artificial intelligence control unit 23 obtains the indoor temperature measured by the indoor thermometer 4, It is the initial temperature of the previous moment when the air conditioner 100 is turned on next time).

Thereby, the controller 2 of the air conditioner 100 can dynamically generate the optimal control parameter according to the current time, season, flow of people, indoor temperature, humidity, etc., and dynamically control the compressor 1 according to the control parameter. The operation allows the medium and large indoor environment space using the air conditioner 100 to have an optimum temperature that is most comfortable and at the same time achieves energy saving effects.

Furthermore, as shown in FIG. 5, the energy-saving control system of the embodiment further includes a near-end server 6 and a remote server 7, and the controller 2 of the air-conditioning device 100 can also utilize the air-conditioning device 100. a communication interface 5 is electrically coupled to the near-end server 6 through an Internet of Things, such as MQTT (Message Queuing Telemetry Transport, a communication protocol jointly developed by IBM and Eurotech) or other wireless or wired area network. The near-end server 6 obtains the current time data, the season data and the person flow data, wherein the person flow data may be preset fixed data or is detected by a person flow coupled to the near-end server 6 Device 7 is provided. The person flow detector 7 is disposed at an access passage or entrance of the indoor environment in which the air conditioner 100 is used to detect and count the number of people entering and leaving the indoor environment. Further, the indoor thermometer 3 and the indoor hygrometer 4 described above can also transmit the measured indoor temperature and indoor humidity to the controller 2 via the near-end server 6. The near-end server 6 can also communicate with the remote server 8 through the Internet of Things, such as the MQTT or the Internet, so that the remote server 8 issues the controller 2 of the air conditioner 100. The control command can be transmitted to the controller 2 through the near-end server 6, so that the controller 2 controls the operation of the compressor 1 according to the control command.

It can be seen that when a plurality of air conditioners 100 are installed in the medium and large indoor environment space, the air conditioners 100 can obtain the required materials for the near-end server 5, unless the required materials are stored in the far The end server 8 then obtains the required data from the remote server 8 through the near-end server 5. Therefore, the air conditioning device 100 disposed in different medium and large indoor field spaces does not need to directly communicate with the remote server 8, but respectively passes through each of the near end servers 5 and the remote end disposed in the field. The server 8 communicates, thereby reducing the bandwidth requirements of the remote server 8 to communicate with the air conditioners 100 in a plurality of indoor field domains. And the remote server 8 can send a control command to each of the designated air conditioners 100 through the respective near-end servers 5 provided in different indoor field domains.

In summary, the above embodiment dynamically generates the optimal control parameter by the controller 2 of the air conditioner 100 according to the current time, season, flow information, indoor temperature, humidity, and the like, and according to the control parameter. The operation of the compressor 1 is dynamically controlled, and the indoor environment in which the air conditioner 100 is installed can produce an optimum temperature that is most comfortable, and at the same time achieve energy-saving effects, and indeed achieve the effects and purposes of the present invention.

However, the above is only the embodiment of the present invention, and the scope of the invention is not limited thereto, and all the equivalent equivalent changes and modifications according to the scope of the patent application and the patent specification of the present invention are still The scope of the invention is covered.

100‧‧‧Air conditioning unit

1‧‧‧Compressor

2‧‧‧ Controller

21‧‧‧Target temperature decision unit

215‧‧‧Interior comfort index (PMV) neural network model

216‧‧‧Target temperature calculation module

23‧‧‧Artificial Intelligence Control Unit

3‧‧‧ indoor thermometer

4‧‧‧ Indoor Hygrometer

5‧‧‧Communication interface

6‧‧‧ Near-end server

7‧‧‧A person flow detector

8‧‧‧Remote Server

S1~S2‧‧‧ steps

Other features and effects of the present invention will be clearly shown in the embodiments of the present invention. FIG. 1 is a main flow chart of an embodiment of a dynamic control method of an air conditioning apparatus according to the present invention; FIG. 3 is a block diagram showing a detailed configuration of a controller included in the controller of the air conditioner of the present embodiment; FIG. 4 is a schematic diagram of a detailed configuration of the controller of the air conditioner of the present embodiment; The second embodiment of the target temperature decision unit includes a detailed block diagram; and FIG. 5 shows that the air conditioner of the embodiment further communicates with a near-end server through a communication interface.

Claims (10)

An energy-saving control method for an air conditioner in a medium and large space, the air conditioner comprising a compressor and a controller electrically coupled to the compressor, and the controller is further electrically coupled to an indoor thermometer and an indoor hygrometer, The control method comprises the following steps: (A) a target temperature decision unit in the controller determines an indoor comfort index target according to the current time and season information, and provides according to the indoor comfort index target and the indoor hygrometer a current target temperature is obtained by the humidity; (B) an artificial intelligence control unit in the controller obtains a first difference between a current temperature provided by the indoor thermometer and the target temperature, and obtains the current temperature and the a second difference between the temperature of the previous moment provided by the indoor thermometer and an error difference between the second difference and the first difference, and obtaining a control according to the first difference and the error difference Parameters, and controlling the operation of the compressor according to the control parameters; and (C) repeating the above steps (A) and (B). The energy-saving control method for the air-conditioning device in the medium-large space according to claim 1, wherein in the step (A), the target temperature decision unit is a first fuzzy operation module, which performs a fuzzy operation according to the current time and season information. The indoor comfort index target is determined, and the target temperature is determined according to an indoor comfort index formula related to temperature and humidity, the indoor comfort index target, and the current humidity. The energy-saving control method for the air-conditioning device in the medium-large space according to the claim 2, wherein the first fuzzy computing module except the current time and season information The indoor comfort level target is also determined based on the current flow information of the person flow information. The energy-saving control method for the air-conditioning device in the medium-large space according to claim 1, wherein in the step (A), the target temperature decision unit comprises an indoor comfort index neural network model and a target temperature calculation module, the indoor The comfort index-like neural network model is a neural network model that has been trained in advance using one-time data and one-season data, so that the indoor comfort index-like neural network model can be based on the current moment and season information input. Determining the indoor comfort index target, and the target temperature calculation module determines the target temperature according to an indoor comfort index formula related to temperature and humidity, the indoor comfort index target, and the current humidity; and the air conditioner further And electrically coupled to a near-end server to obtain the time data and the seasonal data from the near-end server. The energy-saving control method for an air-conditioning device in a medium-large space according to claim 4, wherein the indoor comfort index-like neural network model is a neural network model trained in advance using the time data, the seasonal data, and the one-person flow data. The indoor comfort index neural network model can determine the indoor comfort index target according to the current moment, season and flow information input; and the indoor thermometer and the indoor hygrometer are also provided through the near-end server The measured indoor temperature and indoor humidity are given to the controller, and the near-end server communicates with a remote server through the Internet of Things, so that a remote controller sends a control command to the controller through the near The end server transmits to the controller, and the controller controls the operation of the compressor according to the control command. The energy-saving control method for the air-conditioning device in the medium-large space according to claim 1, wherein in step (B), the artificial intelligence control unit is a second fuzzy operation module, and the error difference is based on the first difference The control parameter is obtained by performing a fuzzy operation, and the control parameter includes a control time length and a work cycle, and the previous moment temperature refers to an indoor temperature measured by the indoor thermometer when the control time length ends. In the energy saving control method of the air conditioning device in the medium and large space according to claim 1, in the step (B), the artificial intelligence control unit includes a neural network control model, and the neural network control model has been pre-utilized. a predetermined range of the difference, a predetermined range of the error difference, and a trained neural network model of the predetermined range of the control parameter, such that the neural network control model can be based on the first input The difference parameter and the error difference are calculated to obtain the control parameter, wherein the control parameter includes a control time length and a duty cycle, and the previous moment temperature refers to the indoor thermometer measured at the end of the control time length Room temperature. The energy-saving control method for an air conditioner in a medium-large space according to any one of claims 1 to 3, wherein the air conditioner is further electrically coupled to a near-end server, and the indoor thermometer and the indoor humidity The controller also provides the measured indoor temperature and indoor humidity to the controller through the near-end server, and the near-end server also communicates with a remote server through the Internet of Things, so that the remote server communicates with the controller A control command issued can be transmitted to the controller through the near-end server, and the controller controls the operation of the compressor according to the control command. An energy-saving control system for an air conditioner in a medium and large-sized space, comprising an air conditioner, an indoor thermometer and an indoor hygrometer, the air conditioner comprising a compressor and a controller electrically coupled to the compressor, and the controller further And the indoor thermometer and the indoor hygrometer are electrically coupled to each other, and the energy-saving control method of the air-conditioning device in the medium-large space according to any one of claims 1 to 3, 6 to 7 can be executed by the controller; and the method The energy-saving control system further includes a near-end server and a remote server, the air-conditioning device further includes a communication interface electrically coupled to the controller, and the controller is electrically coupled to the near-end server through the communication interface The indoor thermometer and the indoor hygrometer further provide the measured indoor temperature and indoor humidity to the controller through the near-end server, and the near-end server communicates with the remote server through the Internet of Things, so that The control command issued by the remote server to the controller can be transmitted to the controller through the near-end server and the communication interface, so that the controller controls the compression according to the control command. Machine operation. An energy-saving control system for an air conditioner in a medium and large-sized space, comprising an air conditioner, an indoor thermometer and an indoor hygrometer, the air conditioner comprising a compressor and a controller electrically coupled to the compressor, and the controller further Electrically coupled with the indoor thermometer and the indoor hygrometer, and capable of performing an energy-saving control method for a medium-sized and large-sized air-conditioning device according to claim 4; wherein the energy-saving control system includes the near-end server and a remote server, and the air conditioner further includes a communication interface electrically coupled to the controller, and the controller is electrically coupled to the near-end server through the communication interface to obtain from the near-end server The time data and the seasonal data, and the indoor thermometer and the indoor hygrometer are also measured through the near-end server. The indoor temperature and the indoor humidity are given to the controller, and the near-end server communicates with the remote server through the Internet of Things, so that a remote controller sends a control command to the controller through the near-end servo. And the communication interface is transmitted to the controller, so that the controller controls the operation of the compressor according to the control command.