Fan-coil gear control method and device and air conditioner

Abstract

This invention provides a fan coil unit speed control method, device, and air conditioner, comprising: acquiring multiple operating environment parameters of the fan coil unit; converting the operating environment parameters based on a preset conversion formula to obtain input parameters; inputting the input parameters into a pre-trained neural network to output a speed control signal for the fan coil unit; the neural network being trained based on multiple preset operating environment parameters; and controlling the airflow of the fan coil unit according to the speed control signal. This method controls the airflow of the fan coil unit by training a neural network based on multiple preset operating environment parameters to take into account factors other than temperature, thereby improving the comfort level of the air conditioning output.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air conditioner control, and in particular to a fan coil gear control method and device and an air conditioner. BACKGROUND

[0002] At present, common control methods of fan coil include on-off control, rule-based control (RBC) and proportional-integral-derivative control (PID), which are widely used in actual projects due to their simple deployment. The logic of on-off control is to start the fan coil to cool the indoor air when the indoor temperature is higher than or equal to the upper limit of the control target, and to turn off the fan coil when the indoor temperature is lower than or equal to the lower limit of the control target, and to keep the air supply unchanged in other cases to reduce the change of the fan coil gear. Here, the rule-based control reserves a safety range Δt for the control of indoor temperature, and the fan coil is started to cool the indoor air at the maximum air volume when the indoor temperature is Δt away from the upper limit of the control target, to avoid further increase of the room temperature, and vice versa, the fan coil is turned off when the indoor temperature is Δt away from the lower limit of the control target, to avoid further decrease of the room temperature, and the fan coil is started at a fixed gear to maintain the indoor temperature in other cases.

[0003] However, the on-off control, rule-based control and PID control commonly used in fan coils at present only take indoor temperature as a single control object, which ignores the influence of indoor humidity on human health. In addition, the heating ventilation and air conditioning system is a highly nonlinear time-varying system, and the comfort control effect of the traditional linear control method is poor. SUMMARY

[0004] The present application aims to provide a fan coil gear control method and device and an air conditioner to improve the comfort control effect of the fan coil.

[0005] In a first aspect, the present application provides a fan coil gear control method, which includes: obtaining a plurality of working environment parameters of a fan coil; the working environment parameters are used to indicate influencing factors affecting the fan coil gear control effect; converting the working environment parameters based on a preset conversion formula to obtain input parameters; inputting the input parameters into a pre-trained neural network to output a gear control signal of the fan coil; the neural network is trained based on a plurality of preset working environment parameters; and controlling the air supply of the fan coil according to the gear control signal.

[0006] In an optional embodiment, after the step of inputting the input parameters into the pre-trained neural network and outputting the gear control signal of the fan coil, the method further comprises: obtaining a current indoor temperature and a current indoor relative humidity at a first preset time point after the fan coil outputs the control signal; and evaluating the control effect of the control signal based on the current indoor temperature and the current indoor relative humidity.

[0007] In an optional embodiment, the working environment parameters comprise an indoor temperature and a relative humidity, and the pre-trained neural network is trained through the following steps: Step 1: obtaining the indoor temperature and the relative humidity in which the fan coil is located; Step 2: converting the indoor temperature and the relative humidity based on a preset conversion formula to obtain experimental input parameters; Step 3: inputting the experimental input parameters into a preset initial neural network to output an experimental gear control signal of the fan coil; Step 4: controlling an experimental air supply amount of the fan coil according to the experimental gear control signal; Step 5: obtaining an intermediate indoor temperature and an intermediate relative humidity at a second preset time point after the fan coil outputs the experimental gear control signal; Step 6: determining a reward value of a preset reward function based on the intermediate indoor temperature and the intermediate relative humidity; Step 7: determining a weight and a bias term coefficient of the initial neural network based on the reward value; Step 8: determining an updated neural network based on the weight and the bias term coefficient; Step 9: determining an updated reward value corresponding to the updated neural network based on the indoor temperature and the relative humidity; and Step 10: repeating the steps 1 to 9 until the updated reward value reaches a preset threshold, and determining the updated neural network corresponding to the updated reward value as the pre-trained neural network.

[0008] In an optional embodiment, before the step of obtaining the indoor temperature and the relative humidity in which the fan coil is located, the method comprises: obtaining a preset hyperparameter; and constructing the preset initial neural network based on the hyperparameter.

[0009] In an optional embodiment, after the step of determining the updated neural network corresponding to the updated reward value as the pre-trained neural network, the method further comprises: obtaining a preset random number, a preset initial random number, and a decay coefficient of a unit time step corresponding to the initial random number; determining a random gear control signal of the fan coil based on the random number, the initial random number, and the decay coefficient; determining an updated weight and an updated bias term coefficient corresponding to the updated neural network based on the random gear control signal; and determining a neural network adjusted based on a greedy exploration strategy based on the updated weight and the updated bias term coefficient.

[0010] In a preferred embodiment, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: determining whether the indoor temperature of the fan coil unit is greater than a preset upper limit of indoor temperature; if so, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: generating a maximum speed control signal for the fan coil unit based on the indoor temperature; and controlling the air volume of the fan coil unit based on the maximum speed control signal.

[0011] In a preferred embodiment, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: determining whether the indoor temperature of the fan coil unit is lower than a preset indoor temperature lower limit; if so, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: generating a shutdown control signal for the fan coil unit based on the indoor temperature; and controlling the fan coil unit to stop supplying air based on the shutdown control signal.

[0012] In a preferred embodiment, the above-mentioned working environment parameters include: indoor temperature, indoor relative humidity, indoor air humidity, outdoor temperature, outdoor relative humidity, outdoor air humidity, indoor wind speed, and occupancy rate.

[0013] Secondly, embodiments of the present invention provide a fan coil unit gear control device, comprising: a parameter acquisition module for acquiring multiple working environment parameters of the fan coil unit; the working environment parameters are used to indicate factors affecting the gear control effect of the fan coil unit; a parameter conversion module for converting the working environment parameters based on a preset conversion formula to obtain input parameters; a neural network control module for inputting the input parameters into a pre-trained neural network and outputting a gear control signal of the fan coil unit; the neural network is trained based on multiple preset working environment parameters; and a control module for controlling the air volume of the fan coil unit according to the gear control signal.

[0014] Thirdly, the present invention provides an air conditioner, comprising: a connected intelligent agent and a fan coil unit; the intelligent agent is used to implement any one of the fan coil unit speed control methods in the first aspect to the seventh possible implementation of the first aspect; the fan coil unit is used to supply air in response to the speed control signal.

[0015] The embodiments of the present invention bring the following beneficial effects:

[0016] This invention provides a fan coil unit speed control method, device, and air conditioner, comprising: acquiring multiple operating environment parameters of the fan coil unit; the operating environment parameters indicating factors affecting the speed control effect of the fan coil unit; converting the operating environment parameters based on a preset conversion formula to obtain input parameters; inputting the input parameters into a pre-trained neural network to output a speed control signal for the fan coil unit; the neural network being trained based on multiple preset operating environment parameters; and controlling the air volume of the fan coil unit according to the speed control signal. This method controls the air volume of the fan coil unit by training a neural network based on multiple preset operating environment parameters, thereby taking into account factors other than temperature, and improving the comfort level of the air conditioning output.

[0017] Other features and advantages disclosed in this embodiment will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above.

[0018] To make the above-mentioned objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0019] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a fan coil unit speed control method provided in an embodiment of the present invention;

[0021] Figure 2 This is a flowchart illustrating a neural network training method provided in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the structure of a fan coil unit gear control device provided in an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0024] Icons: 31-Parameter acquisition module; 32-Parameter conversion module; 33-Neural network control module; 34-Control module; 41-Memory; 42-Processor; 43-Bus; 44-Communication interface. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0026] Currently, common control methods for fan coil units include on / off control, rule-based control (RBC), and proportional-integral-derivative control (PID). These methods are widely used in practical projects due to their ease of deployment. On / off control works as follows: when the indoor temperature is higher than or equal to the upper limit of the control target, the fan coil unit is turned on to cool the room; when the indoor temperature is lower than or equal to the lower limit of the control target, the fan coil unit is turned off. Otherwise, the airflow remains constant to minimize changes in the fan coil unit's speed. Rule-based control, on the other hand, provides a safety margin Δt for controlling the indoor temperature. When the indoor temperature is Δt away from the upper limit of the control target, the fan coil unit operates at maximum airflow to cool the indoor air and prevent further temperature increases. Conversely, when the indoor temperature is Δt away from the lower limit of the control target, the fan coil unit is turned off to prevent further temperature decreases. Otherwise, the fan coil unit operates at a fixed speed to maintain the indoor temperature. However, the on / off control, rule-based control, and PID control commonly used in fan coil units currently only focus on indoor temperature as the single control object, neglecting the impact of indoor humidity on human health. Furthermore, HVAC systems are highly nonlinear and time-varying systems, and traditional linear control methods produce poor comfort control results.

[0027] Based on this, embodiments of the present invention provide a fan coil unit speed control method, device, and air conditioner. This method uses a neural network trained based on multiple preset working environment parameters to control the airflow of the fan coil unit across multiple working environment parameters, thereby taking into account factors other than temperature to improve the comfort level of the air conditioner's output. To facilitate understanding of the embodiments of the present invention, a detailed description of the fan coil unit speed control method disclosed in these embodiments will be provided first.

[0028] Example 1

[0029] In this embodiment, Figure 1 This is a flowchart illustrating a fan coil unit speed control method provided in an embodiment of the present invention.

[0030] Depend on Figure 1As can be seen, the above methods include:

[0031] Step S101: Obtain multiple working environment parameters of the fan coil unit; the above working environment parameters are used to indicate the influencing factors affecting the gear control effect of the above fan coil unit.

[0032] In this embodiment, the above-mentioned working environment parameters include: indoor temperature, indoor relative humidity, indoor air humidity, outdoor temperature, outdoor relative humidity, outdoor air humidity, indoor wind speed, and occupancy rate.

[0033] Step S102: Convert the above working environment parameters based on the preset conversion formula to obtain the input parameters.

[0034] In this embodiment, indoor temperature and relative humidity are used as examples for illustration. For instance, the indoor temperature and relative humidity are converted according to the following formula.

[0035]

[0036]

[0037] In the formula, tem and RH represent the indoor temperature and relative humidity before conversion, while tem′ and RH′ represent the temperature and relative humidity after conversion. The purpose of the above formula is to apply the formula when the indoor temperature tem is at a set upper limit value T. upper bound and setting a lower limit value T lowerbound When tem' is between -1 and 1, tem' is distributed between -1 and 1; when tem is greater than the set upper limit value T upper bound Or less than the set lower limit value T lowerbound When tem′ is linearly increased or decreased, the formula above applies the same principle when the indoor relative humidity RH is at a set upper limit RH. upperbound and setting the lower limit value RH lower bound When RH is between -1 and 1, RH′ will be distributed between -1 and 1; when RH is greater than the set upper limit value RH upperbound Or less than the set lower limit value RH lowerbound When RH increases or decreases by 10, RH′ increases or decreases by 1. This conversion makes tem′ and RH′ similar in magnitude.

[0038] Step S103: Input the above input parameters into the pre-trained neural network and output the gear control signal of the above fan coil unit; the above neural network is trained based on multiple preset working environment parameters.

[0039] In this embodiment, the gear control signal can be pre-divided into multiple gears, and in step S103, the gear control signal of the fan coil unit is output after being matched with the input parameters by a pre-trained neural network.

[0040] Step S104: Control the air volume of the fan coil unit according to the above gear control signal.

[0041] In one embodiment, after step S103, the method further includes: first, at a first preset time point after the fan coil unit outputs the control signal, acquiring the current indoor temperature and the current indoor relative humidity; then, evaluating the control effect of the control signal based on the current indoor temperature and the current indoor relative humidity.

[0042] This invention provides a fan coil unit speed control method, comprising: acquiring multiple working environment parameters of the fan coil unit; the working environment parameters indicating factors affecting the speed control effect of the fan coil unit; converting the working environment parameters based on a preset conversion formula to obtain input parameters; inputting the input parameters into a pre-trained neural network to output a speed control signal for the fan coil unit; the neural network being trained based on multiple preset working environment parameters; and controlling the air volume of the fan coil unit according to the speed control signal. This method controls the air volume of the fan coil unit by training a neural network based on multiple preset working environment parameters, thereby taking into account factors other than temperature, and improving the comfort level of the air conditioning output.

[0043] Example 2

[0044] In this embodiment, Figure 2 This is a flowchart illustrating a neural network training method provided in an embodiment of the present invention. Figure 2 It is mainly used to describe the training process of the pre-trained neural network in Example 1.

[0045] The working environment parameters include: indoor temperature and relative humidity.

[0046] Depend on Figure 2 As can be seen, the above methods include:

[0047] Step 1: Obtain the indoor temperature and relative humidity of the area where the fan coil unit is located.

[0048] In this embodiment, before obtaining the indoor temperature and relative humidity of the fan coil unit, the method includes: first, obtaining preset hyperparameters; then, constructing the preset initial neural network based on the hyperparameters.

[0049] Step 2: Convert the above indoor temperature and relative humidity based on the preset conversion formula to obtain the experimental input parameters.

[0050] Step 3: Input the above experimental input parameters into the preset initial neural network, and output the experimental gear control signal of the above fan coil unit.

[0051] Step 4: Control the experimental air volume of the fan coil unit according to the experimental gear control signal.

[0052] Step 5: At the second preset time point after the fan coil unit outputs the above-mentioned experimental gear control signal, obtain the intermediate indoor temperature and intermediate relative humidity.

[0053] Step 6: Determine the reward value of the preset reward function based on the above intermediate indoor temperature and intermediate relative humidity.

[0054] Step 7: Determine the weights and bias coefficients of the initial neural network based on the reward values ​​mentioned above.

[0055] Step 8: Determine the updated neural network based on the weights and bias coefficients mentioned above.

[0056] Step 9: Based on the indoor temperature and relative humidity mentioned above, determine the update reward value corresponding to the updated neural network.

[0057] Step 10: Repeat steps 1 to 9 above until the updated reward value reaches the preset threshold; determine the updated neural network corresponding to the updated reward value as the pre-trained neural network.

[0058] In this embodiment, after determining the updated neural network corresponding to the updated reward value as the pre-trained neural network, the method further includes the following steps A1-A4:

[0059] Step A1: Obtain a preset random number, a preset initial random number, and the decay coefficient per unit time step corresponding to the initial random number.

[0060] Step A2: Determine the random gear control signal of the fan coil unit based on the above random number, the above initial random number, and the above attenuation coefficient.

[0061] In practical operation, to avoid meaningless exploration and enhance the practicality of the control method, this invention requires action intervention on the agent. This invention selects an ε-greedy exploration strategy to explore more state-action pairs. Specifically, during the training phase, a random number is generated at each preset time step. If this random number is less than the current ε... i Then, a random speed control signal is randomly selected for any fan coil unit.

[0062] Where, ε i =ε0-ε decay ·step i ;

[0063] In the formula, ε decay Let ε be the attenuation coefficient, step i Let i be the i-th time step.

[0064] Step A3: Based on the above random gear control signal, determine the update weights and update bias term coefficients corresponding to the updated neural network.

[0065] Step A4: Based on the updated weights and the updated bias term coefficients, determine the neural network adjusted based on the greedy exploration strategy.

[0066] In this embodiment, the neural network adjusted based on the greedy exploration strategy can also be determined as a pre-trained neural network.

[0067] Furthermore, to avoid meaningless exploration and enhance the practicality of the control method, the present invention also intervenes in the actions of the intelligent agent through the following method.

[0068] In one embodiment, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: determining whether the indoor temperature of the fan coil unit is greater than a preset upper limit of indoor temperature; if so, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: generating a maximum speed control signal for the fan coil unit based on the indoor temperature; and controlling the air volume of the fan coil unit based on the maximum speed control signal.

[0069] In another embodiment, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: determining whether the indoor temperature of the fan coil unit is lower than a preset indoor temperature lower limit; if so, after obtaining the indoor temperature and relative humidity of the fan coil unit, the method further includes: generating a shutdown control signal for the fan coil unit based on the indoor temperature; and controlling the fan coil unit to stop supplying air based on the shutdown control signal.

[0070] This invention provides a neural network training method. The aforementioned working environment parameters include indoor temperature and relative humidity. The pre-trained neural network is trained through the following steps: Step 1: Obtain the indoor temperature and relative humidity of the fan coil unit; Step 2: Convert the indoor temperature and relative humidity based on a preset conversion formula to obtain experimental input parameters; Step 3: Input the experimental input parameters into a preset initial neural network and output the experimental speed control signal of the fan coil unit; Step 4: Control the experimental air volume of the fan coil unit according to the experimental speed control signal; Step 5: After the fan coil unit outputs the experimental speed control signal, the next step... Step 2: Obtain intermediate indoor temperature and relative humidity at preset time points; Step 6: Determine the reward value of the preset reward function based on the intermediate indoor temperature and relative humidity; Step 7: Determine the weights and bias coefficients of the initial neural network based on the reward value; Step 8: Determine the updated neural network based on the weights and bias coefficients; Step 9: Determine the updated reward value corresponding to the updated neural network based on the indoor temperature and relative humidity; Step 10: Repeat steps 1 to 9 until the updated reward value reaches a preset threshold; The updated neural network corresponding to the updated reward value is identified as the pre-trained neural network. This method, by setting a reward function, determines the length of the agent's training process and the quality of the training effect, thereby obtaining a more accurate neural network.

[0071] Example 3

[0072] This embodiment provides a fan coil unit speed control device. Figure 3 This is a schematic diagram of a fan coil unit gear control device provided in an embodiment of the present invention.

[0073] Depend on Figure 3 As seen above, the aforementioned device includes:

[0074] The parameter acquisition module 31 is used to acquire multiple working environment parameters of the fan coil unit; the above working environment parameters are used to indicate the influencing factors that affect the gear control effect of the above fan coil unit.

[0075] The parameter conversion module 32 is used to convert the above working environment parameters based on a preset conversion formula to obtain input parameters;

[0076] The neural network control module 33 is used to input the above-mentioned input parameters into a pre-trained neural network and output the gear control signal of the above-mentioned fan coil unit; the above-mentioned neural network is trained based on multiple preset working environment parameters;

[0077] The control module 34 is used to control the air volume of the fan coil unit according to the above-mentioned gear control signal.

[0078] The parameter acquisition module 31, parameter conversion module 32, neural network control module 33, and control module 34 are connected in sequence.

[0079] The fan coil unit gear control device provided in this embodiment of the invention has the same technical features as the fan coil unit gear control method provided in the above embodiments, and therefore can solve the same technical problems and achieve the same technical effects. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0080] Example 4

[0081] This embodiment provides an electronic device, including a processor and a memory. The memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the steps of the fan coil unit gear control method.

[0082] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a fan coil unit gear control method.

[0083] See Figure 4 The diagram shows the structure of an electronic device, which includes a memory 41 and a processor 42. The memory 41 stores a computer program that can run on the processor 42. When the processor executes the computer program, it implements the steps provided by the above-mentioned fan coil unit gear control method.

[0084] like Figure 4 As shown, the device also includes a bus 43 and a communication interface 44, with the processor 42, the communication interface 44 and the memory 41 connected via the bus 43; the processor 42 is used to execute executable modules, such as computer programs, stored in the memory 41.

[0085] The memory 41 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 44 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0086] Bus 43 can be an ISA bus, PCI bus, or EISA bus, etc. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0087] The memory 41 stores the program, and the processor 42 executes the program after receiving the execution instruction. The method executed by the fan coil unit speed control device disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 42, or implemented by the processor 42. The processor 42 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit in the hardware of the processor 42 or by instructions in the form of software. The processor 42 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this invention can be directly manifested as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory 41, and processor 42 reads information from memory 41 and, in conjunction with its hardware, completes the steps of the above method.

[0088] Furthermore, this embodiment of the invention also provides a machine-readable storage medium storing machine-executable instructions. When the machine-executable instructions are invoked and executed by the processor 42, the machine-executable instructions cause the processor 42 to implement the above-described fan coil unit gear control method.

[0089] The electronic devices and computer-readable storage media provided in the embodiments of the present invention have the same technical features, so they can also solve the same technical problems and achieve the same technical effects.

[0090] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0091] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

Claims

1. A method of step control of a fan-coil, characterized in that, The method comprises the following steps: acquiring a plurality of working environment parameters of the fan-coil unit; the working environment parameters are used to indicate influencing factors affecting the control effect of the fan-coil unit; transforming the working environment parameters based on a preset transformation formula to obtain input parameters; inputting the input parameters into a pre-trained neural network to output a gear control signal of the fan-coil unit; the neural network is trained based on a plurality of preset working environment parameters; controlling the air supply amount of the fan-coil unit according to the gear control signal; wherein the working environment parameters comprise indoor temperature and indoor relative humidity; the pre-trained neural network is trained by the following steps: Step 1: acquiring the indoor temperature and the indoor relative humidity of the fan-coil unit; Step 2: transforming the indoor temperature and the indoor relative humidity based on a preset transformation formula to obtain experimental input parameters; Step 3: inputting the experimental input parameters into a preset initial neural network to output an experimental gear control signal of the fan-coil unit; Step 4: controlling the experimental air supply amount of the fan-coil unit according to the experimental gear control signal; Step 5: acquiring intermediate indoor temperature and intermediate indoor relative humidity at a second preset time point after the fan-coil unit outputs the experimental gear control signal; Step 6: determining a reward value of a preset reward function according to the intermediate indoor temperature and the intermediate indoor relative humidity; Step 7: determining the weight and the bias term coefficient of the initial neural network according to the reward value; Step 8: determining an updated neural network according to the weight and the bias term coefficient; Step 9: determining an updated reward value corresponding to the updated neural network according to the indoor temperature and the indoor relative humidity; Step 10: repeating steps 1 to 9 until the updated reward value reaches a preset threshold; determining the updated neural network corresponding to the updated reward value as the pre-trained neural network; wherein after the step of determining the updated neural network corresponding to the updated reward value as the pre-trained neural network, the method further comprises: acquiring a preset random number, a preset initial random number, and a decay coefficient of a unit time step corresponding to the initial random number; determining a random gear control signal of the fan-coil unit according to the random number, the initial random number, and the decay coefficient; determining an updated weight and an updated bias term coefficient corresponding to the updated neural network according to the random gear control signal; determining a neural network adjusted based on a greedy exploration strategy according to the updated weight and the updated bias term coefficient; wherein the indoor temperature and the indoor relative humidity are transformed according to the following formula: in, and The indoor temperature and relative humidity before the conversion. and The converted temperature and indoor relative humidity, when the indoor temperature At the set upper limit value and setting a lower limit value In between, Distributed between -1 and 1; when Greater than the set upper limit value Or less than the set lower limit value At that time, Linear increase or decrease; when indoor relative humidity At the set upper limit value and setting a lower limit value In between, Distributed between -1 and 1; when Greater than the set upper limit value Or less than the set lower limit value hour, For every increase or decrease of 10, Then increase or decrease by 1 to make and Their magnitudes are similar.

2. The method of claim 1, wherein, after the step of inputting the input parameters into the pre-trained neural network to output the gear control signal of the fan-coil unit, the method further comprises: acquiring current indoor temperature and current indoor relative humidity at a first preset time point after the fan-coil unit outputs the control signal; According to the current indoor temperature and the current indoor relative humidity, an effect of the control signal is evaluated.

3. The method of claim 1, wherein, After the step of obtaining the indoor temperature and the indoor relative humidity where the fan coil is located, the method further comprises: determining whether the indoor temperature where the fan coil is located is greater than a preset upper limit of the indoor temperature; if yes, after the step of obtaining the indoor temperature and the indoor relative humidity where the fan coil is located, the method further comprises: generating a highest gear control signal of the fan coil according to the indoor temperature; controlling the air supply amount of the fan coil according to the highest gear control signal.

4. The method of claim 3, wherein, After the step of obtaining the indoor temperature and the indoor relative humidity where the fan coil is located, the method further comprises: determining whether the indoor temperature where the fan coil is located is less than a preset lower limit of the indoor temperature; if yes, after the step of obtaining the indoor temperature and the indoor relative humidity where the fan coil is located, the method further comprises: generating a closing control signal of the fan coil according to the indoor temperature; controlling the fan coil to stop air supply according to the closing control signal.

5. The method of claim 1, wherein, The working environment parameters include: indoor temperature, indoor relative humidity, indoor air moisture content, outdoor temperature, outdoor relative humidity, outdoor air moisture content, indoor wind speed, and the rate of people in the room.

6. A fan coil unit gear control apparatus, characterized by, comprises: a parameter acquisition module configured to acquire a plurality of working environment parameters of a fan coil; the working environment parameters are used to indicate influencing factors affecting the control effect of the fan coil gear; a parameter conversion module configured to convert the working environment parameters based on a preset conversion formula to obtain input parameters; a neural network control module configured to input the input parameters into a pre-trained neural network to output a gear control signal of the fan coil; the neural network is trained based on a plurality of preset working environment parameters; a control module configured to control the air supply amount of the fan coil according to the gear control signal; wherein the working environment parameters include: indoor temperature and indoor relative humidity; the pre-trained neural network is trained by the following steps: Step 1: Obtain the indoor temperature and the indoor relative humidity where the fan coil is located; Step 2: Convert the indoor temperature and the indoor relative humidity based on a preset conversion formula to obtain experimental input parameters; Step 3: Input the experimental input parameters into a preset initial neural network to output an experimental gear control signal of the fan coil; Step 4: Control the experimental air supply amount of the fan coil according to the experimental gear control signal; Step 5: At a second preset time point after the fan coil outputs the experimental gear control signal, obtain the intermediate indoor temperature and the intermediate indoor relative humidity; Step 6: Determine the reward value of a preset reward function according to the intermediate indoor temperature and the intermediate indoor relative humidity; Step 7: Determine the weight and the bias term coefficient of the initial neural network according to the reward value; Step 8: Determine an updated neural network according to the weight and the bias term coefficient. Step 9: determining an updated reward value corresponding to the updated neural network according to the indoor temperature and the indoor relative humidity; Step 10: repeating the steps 1 to 9 until the updated reward value reaches a preset threshold; determining the updated neural network corresponding to the updated reward value as the pre-trained neural network; After the step of determining the updated neural network corresponding to the updated reward value as the pre-trained neural network, the method further comprises: obtaining a preset random number, a preset initial random number and a decay coefficient of a unit time step corresponding to the initial random number; determining a random gear control signal of the fan coil according to the random number, the initial random number and the decay coefficient; determining an updated weight and an updated bias term coefficient corresponding to the updated neural network according to the random gear control signal; and determining a neural network adjusted based on a greedy exploration strategy according to the updated weight and the updated bias term coefficient. The indoor temperature and the indoor relative humidity are converted according to the following formula: in, and The indoor temperature and relative humidity before the conversion. and The converted temperature and indoor relative humidity, when the indoor temperature At the set upper limit value and setting a lower limit value In between, Distributed between -1 and 1; when Greater than the set upper limit value Or less than the set lower limit value At that time, Linear increase or decrease; when indoor relative humidity At the set upper limit value and setting a lower limit value In between, Distributed between -1 and 1; when Greater than the set upper limit value Or less than the set lower limit value hour, For every increase or decrease of 10, Then increase or decrease by 1 to make and Their magnitudes are similar.

7. An air conditioner characterized by comprising: The method comprises: An intelligent agent and a fan coil connected to each other; the intelligent agent is configured to implement the fan coil gear control method according to any one of claims 1 to 5; and the fan coil is configured to perform air supply in response to the gear control signal.

Images