Control method of a refrigeration appliance and refrigeration appliance

Abstract

The application relates to a control method of a refrigeration equipment and the refrigeration equipment. The control method of the refrigeration equipment comprises the following steps: collecting the temperature and humidity in a storage compartment of the refrigeration equipment; when the temperature and / or humidity in the storage compartment does not meet the requirements, optimizing three parameters kp, ki and kd of a PID controller through an Elman neural network; and controlling the refrigeration equipment to run through the optimized PID controller to adjust the temperature and humidity in the storage compartment. The three parameters kp, ki and kd of the PID control are online set by the Elman neural network self-adaption, self-organization and self-learning, so that the PID parameters are adjusted in errors, the temperature and humidity in the storage compartment can be quickly adjusted even if the external environment is strongly time-varying, the real-time performance is good, the temperature and humidity in the storage compartment can be more accurately controlled, and the requirements of constant temperature and constant humidity can be better met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of refrigeration equipment, in particular to a control method of a refrigeration equipment and the refrigeration equipment. BACKGROUND

[0002] The refrigeration equipment, such as a side-by-side cabinet, a wine cabinet and the like, for the target group of high-end consumer groups needs to be controlled very accurately to achieve higher product experience and meet the needs of high-end consumer groups.

[0003] The traditional control method of the refrigeration equipment adopts a control method of switching points in sections. The control method generally adopts the following control flow:

[0004] (1) Collect the temperature and humidity of the current chamber and transmit them back to the MCU for temperature and humidity judgment;

[0005] (2) Determine the switching point according to the temperature at this time and control the switching of the compressor and the fan;

[0006] (3) If the temperature is greater than the switching-on point, the compressor is turned on, and if the temperature is less than the switching-off point, the compressor is turned off; as the temperature approaches the required temperature range, it slowly reaches a stable state.

[0007] This control method only considers the temperature difference between the set temperature and the current temperature, and the temperature regulation hysteresis is very large, which makes it difficult to accurately control the temperature, and the temperature of the chamber is inevitably lower or higher than the set temperature, and the temperature drop is uneven and the fluctuation is large. Moreover, this control method does not consider that the smaller the temperature difference is, the voltage should be slightly reduced to reduce the temperature drop rate, and does not consider the influence of the temperature difference change rate on the supply voltage.

[0008] The traditional PID control method is often used to control the refrigeration equipment in the prior art, and the process is as follows:

[0009] (1) Collect the temperature and humidity of the current chamber and transmit them back to the MCU for temperature and humidity judgment;

[0010] (2) Determine the switching point according to the temperature at this time and control the switching of the compressor and the fan;

[0011] (3) The traditional PID control adopts an incremental digital PID control algorithm, that is, the control value of each output is the increment of the control, which adjusts the error in turn.

[0012] The traditional PID control method needs to adjust the P, I and D parameters of the control system; since the effects among the three are both mutual assistance and mutual influence and restriction, a lot of experience is needed in actual operation, and the determination of the parameters is very troublesome. The refrigeration equipment needs to control the temperature and humidity of the storage room, and has strong time-varying property and high nonlinearity, so the traditional PID control is difficult to quickly find the best combination in accordance with the actual situation, and there is a certain lag, and the adjustment time is long. SUMMARY

[0013] One object of the first aspect of the present application aims to overcome at least one defect of the prior art, and provide a control method capable of quickly adjusting the temperature and humidity in the storage room of the refrigeration equipment.

[0014] Another object of the first aspect of the present application is to improve the accuracy of the temperature and humidity control in the storage room.

[0015] The object of the second aspect of the present application is to provide a refrigeration equipment capable of quickly and accurately controlling the temperature and humidity in the storage room thereof.

[0016] According to the first aspect of the present application, the present application provides a control method of a refrigeration equipment, comprising:

[0017] Collecting the temperature and humidity in the storage room of the refrigeration equipment;

[0018] When the temperature and / or humidity in the storage room does not meet the requirements, optimizing the three parameters kp, ki and kd of the PID controller through the Elman neural network; and

[0019] Controlling the refrigeration equipment to run through the optimized PID controller, so as to adjust the temperature and humidity in the storage room.

[0020] Optionally, the control method further comprises:

[0021] Collecting the temperature and humidity in the storage room again;

[0022] When the temperature and / or humidity in the storage room still does not meet the requirements, updating the weights of the output layer and the hidden layer in the Elman neural network in turn, so as to optimize the three parameters kp, ki and kd of the PID controller again, until the temperature and humidity in the storage room meet the requirements.

[0023] Optionally, the judging step of whether the temperature and / or humidity in the storage room meets the requirements comprises:

[0024] Setting a target function according to the temperature and humidity function relationship;

[0025] calculating preset values of the target function according to the set temperature and humidity in the storage compartment and calculating output values of the target function according to the collected temperature and humidity in the storage compartment;

[0026] calculating a difference between the preset values and the output values of the target function; and

[0027] if the difference is out of a preset error range, determining that the temperature and / or humidity in the storage compartment do not meet the requirements, otherwise, determining that the temperature and / or humidity in the storage compartment meet the requirements.

[0028] Optionally, the step of optimizing three parameters kp, ki and kd of the PID controller by the Elman neural network comprises:

[0029] establishing the Elman neural network with the collected temperature, humidity and difference in the storage compartment as inputs and the three parameters kp, ki and kd of the PID controller as outputs;

[0030] learning the three parameters kp, ki and kd respectively by using a learning index function of the Elman neural network to obtain optimized values of the three parameters kp, ki and kd.

[0031] Optionally, the Elman neural network is expressed by the following formula:

[0032] y(k) = g [W2h(k)];

[0033] h(k) = f{W3X c (k) + W1[u(k-1)]};

[0034] x c (k) = x c (k-1) + h(k-1); wherein

[0035] W1 is a weight value from an input layer to a hidden layer, W2 is a weight value from the hidden layer to an output layer, and W3 is a weight value from a carry layer to the hidden layer; u is an input of the Elman neural network, h(k) and X c (k) are outputs of the hidden layer and the carry layer respectively, and y(k) is an output of the Elman neural network; f(·) is a S-type activation function, and g(·) is a transfer function of an output neuron.

[0036] Optionally, the learning index function of the Elman neural network is as follows:

[0037] wherein

[0038] E(w) is the learning index function of the Elman neural network, yk (w) is an output vector of the Elman neural network, y k (w)' is a target output vector.

[0039] Optionally, the step of learning the three parameters kp, ki and kd respectively by using the learning index function of the Elman neural network comprises:

[0040] First, a first-order gradient of the learning index function is calculated as follows:

[0041]

[0042]

[0043] Then, the three parameters kp, ki and kd are optimized by the following formula:

[0044] wherein

[0045] η is a learning rate.

[0046] Optionally, η is any value between 0.04 and 0.06.

[0047] Optionally, the step of controlling the refrigeration equipment to operate by using the optimized PID controller comprises:

[0048] inputting a preset value of the target function as input data and an output value of the target function as feedback data into the PID controller;

[0049] controlling a temperature adjusting device and a humidity adjusting device of the refrigeration equipment according to a deviation between the input data and the feedback data.

[0050] According to a second aspect of the present application, the present application further provides a refrigeration equipment comprising:

[0051] a collecting device configured to collect temperature and humidity in a storage compartment of the refrigeration equipment; and

[0052] a control device comprising a processor and a memory, wherein the memory stores a machine executable program, and the machine executable program is executed by the processor to implement the control method according to any one of the above solutions.

[0053] The control method of the refrigeration equipment of the present application collects the temperature and humidity in the storage compartment, and when the temperature and humidity in the storage compartment do not meet the requirements, optimizes the three parameters kp, ki and kd of the PID controller through the Elman neural network. Compared with the traditional PID control which determines the values of the parameters kp, ki and kd through experience, the present application self-adapts, self-organizes and self-learns to online set the values of the three parameters kp, ki and kd of the PID control, thereby making error adjustment to the PID parameters, and even when the external environment has strong time-varying nature, the temperature and humidity in the storage compartment can be quickly adjusted, and the real-time performance is good.

[0054] Moreover, the Elman neural network has strong approximation function, can learn the feedback result of the storage compartment according to the temperature and humidity of the storage compartment, adjusts the values of kp, ki and kd output according to the feedback result, makes error adjustment to the PID parameters, and makes the temperature and humidity control in the storage compartment more accurate and closer to the requirements of constant temperature and humidity.

[0055] The above and other objects, advantages and features of the present application will become more apparent from the following detailed description of some embodiments thereof, when considered in conjunction with the attached drawings. BRIEF DESCRIPTION OF DRAWINGS

[0056] Some specific embodiments of the present application will be described in detail below with reference to the attached drawings, which are presented by way of illustration and not of limitation. Like reference numerals designate like parts or portions throughout the drawings. It should be understood that the drawings are not necessarily to scale. In the drawings:

[0057] Figure 1 is a schematic flow chart of the control method of the refrigeration equipment according to one embodiment of the present application;

[0058] Figure 2 is a schematic flow chart of the control method of the refrigeration equipment according to another embodiment of the present application;

[0059] Figure 3 is a schematic flow chart of the judgment step of whether the temperature and humidity in the storage compartment meet the requirements according to another embodiment of the present application;

[0060] Figure 4 is a schematic network structure diagram of the Elman neural network according to one embodiment of the present application;

[0061] Figure 5 is a schematic principle diagram of the refrigeration equipment using PID control according to one embodiment of the present application;

[0062] Figure 6 is a schematic structure block diagram of the refrigeration equipment according to one embodiment of the present application. DETAILED DESCRIPTION

[0063] The present application provides a control method of a refrigeration device, Figure 1 is a schematic flow chart of the control method of the refrigeration device according to an embodiment of the present application. Referring to Figure 1 The control method of the refrigeration device of the present application comprises:

[0064] Step S10, collecting the temperature and humidity in the storage compartment of the refrigeration device;

[0065] Step S20, determining whether the temperature and humidity in the storage compartment meet the requirements; if yes, ending (i.e. no need to make any adjustment to the temperature and humidity in the storage compartment), if not, turning to step S30;

[0066] Step S30, optimizing the three parameters kp, ki and kd of the PID controller through the Elman neural network;

[0067] Step S40, controlling the refrigeration device to run through the optimized PID controller to adjust the temperature and humidity in the storage compartment.

[0068] The control method of the refrigeration device of the present application collects the temperature and humidity in the storage compartment, and when the temperature and / or humidity in the storage compartment do not meet the requirements, optimizes the three parameters kp, ki and kd of the PID controller through the Elman neural network. Compared with the traditional PID control which determines the values of the parameters kp, ki and kd through experience, the present application adjusts the values of the three parameters kp, ki and kd of the PID control through the Elman neural network self-adaption, self-organization and self-learning, thereby making error adjustment to the PID parameters, and even when the external environment is strongly time-varying, the temperature and humidity in the storage compartment can be quickly adjusted, and the real-time performance is good.

[0069] Moreover, the Elman neural network has a strong approximation function, which can learn the feedback results of the storage compartment according to the temperature and humidity of the storage compartment, adjust the values of kp, ki and kd outputted according to the feedback results, make error adjustment to the PID parameters, and make the temperature and humidity control in the storage compartment more accurate and closer to the requirements of constant temperature and humidity.

[0070] Specifically, in step S10, the temperature and humidity in the storage compartment can be collected through a collecting device. The collecting device can include a temperature sensor, a humidity sensor, etc.

[0071] In some embodiments, the control method of the present application further comprises:

[0072] collecting the temperature and humidity in the storage compartment again; and

[0073] When the temperature and / or humidity in the storage compartment still do not meet the requirements, the weights of the output layer and the hidden layer in the Elman neural network are updated in sequence to optimize the three parameters kp, ki and kd of the PID controller again until the temperature and humidity in the storage compartment meet the requirements.

[0074] Specifically, Figure 2 is a schematic flow chart of the control method of the refrigeration equipment according to another embodiment of the present application. In this embodiment, after step S40, the control method of the present application further comprises:

[0075] Step S50, the temperature and humidity in the storage compartment are collected again; and

[0076] Step S60, it is judged whether the temperature and humidity in the storage compartment meet the requirements; if not, go to step S70; if yes, end;

[0077] Step S70, the weights of the output layer and the hidden layer in the Elman neural network are updated in sequence to optimize the three parameters kp, ki and kd of the PID controller again, and return to step S40.

[0078] That is, when the adjusted temperature and / or humidity in the storage compartment still do not meet the requirements, the weights of the output layer and the hidden layer in the Elman neural network are updated in sequence, and through the continuous reverse repair of the weights, until the temperature and humidity in the storage compartment meet the requirements after the three parameters kp, ki and kd output by the Elman neural network are brought into the PID controller and adjusted by the PID controller.

[0079] Figure 3 is a schematic flow chart of the judging step of whether the temperature and humidity in the storage compartment meet the requirements according to another embodiment of the present application. In some embodiments, the step S20 of judging whether the temperature and humidity in the storage compartment meet the requirements can specifically comprise:

[0080] Step S21, a target function is set according to the temperature and humidity function relationship;

[0081] Step S22, a preset value of the target function is calculated according to the set temperature and the set humidity in the storage compartment, and an output value of the target function is calculated according to the collected temperature and humidity in the storage compartment;

[0082] Step S23, a difference between the preset value and the output value of the target function is calculated; and

[0083] Step S24, it is judged whether the difference is outside the preset error range; if yes, go to step S25; if no, go to step S26;

[0084] Step S25, determining whether the temperature and / or humidity in the storage compartment meets the requirements;

[0085] Step S26, determining whether the temperature and humidity in the storage compartment meet the requirements.

[0086] That is, a target function can be set according to the temperature-humidity function relationship between the temperature and the humidity, and the target function is used to determine whether the temperature and the humidity in the storage compartment meet the requirements, so as to simplify the control logic and further improve the speed of adjusting the temperature and humidity in the storage compartment.

[0087] Further, the temperature-humidity function relationship between the temperature and the humidity can be expressed by the following formula:

[0088] K = Et' - AP(t - t'); wherein

[0089] K is the target function, Et' is the saturation water vapor pressure of the chamber, A is the psychrometric coefficient of the psychrometer, which can change with the wind speed around the storage compartment, P is the current air pressure, t is the temperature of the storage compartment, and t' is the humidity of the storage compartment.

[0090] In some embodiments, the step of optimizing the three parameters kp, ki and kd of the PID controller by the Elman neural network can specifically include:

[0091] The collected temperature and humidity in the storage compartment and the difference calculated in step S23 are taken as inputs, and the three parameters kp, ki and kd of the PID controller are taken as outputs to establish the Elman neural network.

[0092] The three parameters kp, ki and kd are learned by using the learning index function of the Elman neural network, so as to obtain the optimized values of the three parameters kp, ki and kd.

[0093] Figure 4 is a schematic network structure diagram of the Elman neural network according to an embodiment of the present application. Specifically, the established Elman neural network can include multiple input layers, multiple hidden layers, multiple receiving layers and multiple output layers.

[0094] Further, the Elman neural network can be expressed by the following formula:

[0095] y(k) = g[W2h(k)];

[0096] h(k) = f{W3X c (k) + W1[u(k - 1)]};

[0097] x c (k) = x c (k - 1) + h(k - 1); wherein

[0098] W1 is the weight value from the input layer to the hidden layer, W2 is the weight value from the hidden layer to the output layer, and W3 is the weight value from the context layer to the hidden layer; u is the input of the Elman neural network, h(k) and X c (k) are the outputs of the hidden layer and the context layer respectively, y(k) is the output of the Elman neural network; f(·) is a S-type activation function, and g(·) is a transfer function of the output neuron.

[0099] In some embodiments, the learning index function of the Elman neural network is as follows:

[0100] wherein

[0101] E(w) is the learning index function of the Elman neural network, y k (w) is the output vector of the Elman neural network, y k (w) is the target output vector.

[0102] In some embodiments, the steps of learning the three parameters kp, ki and kd respectively by using the learning index function of the Elman neural network include:

[0103] First, the first-order gradient of the learning index function is calculated as follows:

[0104]

[0105]

[0106] Then, the three parameters kp, ki and kd are optimized by the formula as shown below:

[0107] wherein

[0108] η is the learning rate.

[0109] In some embodiments, the learning rate η cannot be set too large or too small. After trial and error, the value of η is set to any value in the range of 0.04-0.06 for the best effect.

[0110] For example, the value of the learning rate η can be 0.04, 0.05 or 0.06.

[0111] Figure 5 is a schematic diagram of a refrigeration device using PID control according to an embodiment of the present application. In some embodiments, the step of controlling the refrigeration device to operate by using the optimized PID controller can specifically include:

[0112] input the preset value of the target function as input data and the output value of the target function as feedback data into the PID controller; and

[0113] control the temperature adjusting device and the humidity adjusting device of the refrigeration equipment according to the deviation between the input data and the feedback data.

[0114] That is, the preset value of the target function calculated according to the temperature and humidity function relationship between the set temperature and the set humidity can be input as the input data of the PID, the output value of the target function calculated according to the temperature and humidity function relationship between the measured temperature and the measured humidity can be input as the feedback of the PID, the deviation between the two can be calculated, and the temperature adjusting device and the humidity adjusting device of the refrigeration equipment can be controlled by using the optimized PID parameters.

[0115] Specifically, the temperature adjusting device of the refrigeration equipment can be a compressor, a supply air damper or other devices capable of adjusting the temperature of the storage compartment, and the humidity adjusting device of the refrigeration equipment can be a humidity adjusting fan or other devices capable of adjusting the humidity of the storage compartment.

[0116] The application further provides a refrigeration equipment, Figure 6 is a schematic structural block diagram of the refrigeration equipment according to an embodiment of the application. Referring to Figure 6 The refrigeration equipment 1 of the application comprises a collection device 10 and a control device 20.

[0117] The collection device 10 is used to collect the temperature and humidity in the storage compartment of the refrigeration equipment. Specifically, the collection device can comprise a temperature sensor or a temperature sensor and the like for collecting the temperature in the storage compartment, and a humidity sensor and the like for directly collecting the humidity in the storage compartment. Of course, the humidity in the storage compartment can also be indirectly obtained through the temperature in the storage compartment and the conversion relationship between temperature and humidity.

[0118] The control device 20 comprises a processor 21 and a memory 22, the memory 22 stores a machine executable program 23, and the machine executable program 23 is used to realize the control method described in any of the above embodiments when executed by the processor 21.

[0119] The refrigeration equipment of the application can optimize the three parameters kp, ki and kd of the PID controller through the Elman neural network when the temperature and / or humidity in the storage compartment does not meet the requirements. Compared with the traditional PID control which determines the values of the parameters kp, ki and kd by experience, the application adjusts the values of the three parameters kp, ki and kd of the PID control online through self-adaptation, self-organization and self-learning of the Elman neural network, thereby adjusting the PID parameters, and the temperature and humidity in the storage compartment can be quickly adjusted even if the external environment is strongly time-varying, and the real-time performance is good.

[0120] And, the Elman neural network has a strong approximation function, which can learn the feedback result of the storage compartment according to the temperature and humidity of the storage compartment, adjust the values of kp, ki and kd of the output according to the feedback result, and make error adjustment on the PID parameters, so that the temperature and humidity control in the storage compartment is more accurate and closer to the requirement of constant temperature and humidity.

[0121] Further, the refrigeration equipment 1 further comprises a temperature adjusting device for adjusting the temperature in the storage compartment and a humidity adjusting device for adjusting the humidity in the storage compartment. The temperature adjusting device may, for example, be a compressor, a supply air damper, etc., and the humidity adjusting device may, for example, be a humidity adjusting fan, etc.

[0122] In some embodiments, the refrigeration equipment 1 of the present application can be a high-end refrigeration equipment such as a side cabinet, a wine cabinet, etc.

[0123] In other embodiments, the refrigeration equipment 1 of the present application can also be a common refrigeration and freezing device. For example, the refrigeration equipment 1 can also be a household or commercial refrigeration equipment such as a refrigerator, a freezer, a refrigerator cabinet, etc.

[0124] In yet other embodiments, the refrigeration equipment 1 of the present application can also be a medical refrigeration equipment such as a refrigerated container.

[0125] In still other embodiments, the refrigeration equipment 1 of the present application can also be other types of refrigeration equipment with high control precision requirements on temperature and humidity.

[0126] At this point, those skilled in the art should recognize that, although the present application has been shown and described in detail in the above embodiments, many other variations or modifications can be directly determined or deduced from the disclosure of the present application according to the principles of the present application without departing from the spirit and scope of the present application. Therefore, the scope of the present application should be understood and recognized as covering all these other variations or modifications.

Claims

1. A control method of a refrigeration device, comprising: collecting temperature and humidity in a storage compartment of the refrigeration device; optimizing three parameters kp, ki and kd of a PID controller by an Elman neural network when the temperature and / or humidity in the storage compartment does not meet a requirement; and controlling the refrigeration device to run by the PID controller optimized to adjust the temperature and humidity in the storage compartment; wherein the step of judging whether the temperature and / or humidity in the storage compartment meets the requirement comprises: setting a target function according to a temperature-humidity function relationship; calculating a preset value of the target function according to a set temperature and a set humidity in the storage compartment and calculating an output value of the target function according to the collected temperature and humidity in the storage compartment; calculating a difference between the preset value and the output value of the target function; and determining that the temperature and / or humidity in the storage compartment does not meet the requirement if the difference is outside a preset error range; and the step of optimizing the three parameters kp, ki and kd of the PID controller by the Elman neural network comprises: establishing the Elman neural network with the collected temperature, humidity and difference in the storage compartment as inputs and the three parameters kp, ki and kd of the PID controller as outputs; and learning the three parameters kp, ki and kd respectively by a learning index function of the Elman neural network to obtain values of the three parameters kp, ki and kd optimized. 2.The control method of claim 1, further comprising: collecting the temperature and humidity in the storage compartment again; updating weights of output layers and hidden layers in the Elman neural network in sequence to optimize the three parameters kp, ki and kd of the PID controller again until the temperature and humidity in the storage compartment meet the requirement when the temperature and / or humidity in the storage compartment still does not meet the requirement. The step of judging whether the temperature and / or humidity in the storage compartment meets the requirement further comprises: determining that the temperature and / or humidity in the storage compartment meets the requirement if the difference is within the preset error range. 4.The control method of claim 1, wherein the Elman neural network is expressed by the following formula: y (k) =g [W2h (k) ]. 5.The control method of claim 4, wherein the learning index function of the Elman neural network is as follows: and the step of learning the three parameters kp, ki and kd respectively by the learning index function of the Elman neural network comprises: first calculating a first-order gradient of the learning index function as follows: and then optimizing the three parameters kp, ki and kd by the following formula: wherein η is a learning rate. 7.The control method of claim 6, wherein η is any value between 0.04 and 0.

06. The step of controlling the refrigeration device to run by the PID controller optimized comprises: ​ ​ ​ ​ ​ ​ ​ ​ 3. The control method according to claim 2, wherein ​ ​ ​ ​ ​ h(k) = f{W3X c (k) + W1[u(k - 1)]; x c (k) = x c (k - 1) + h(k - 1); where W1 is the weight from input layer to hidden layer, W2 is the weight from hidden layer to output layer, and W3 is the weight from context layer to hidden layer; u is the input of the Elman neural network, h(k) and X c (k) are the outputs of hidden layer and context layer respectively, and y(k) is the output of the Elman neural network; f(·) is the S-type activation function, and g(·) is the transfer function of output neuron. ​ ​ wherein E(w) is the learning index function of the Elman neural network, y k (w) is the output vector of the Elman neural network, y k (w) is the target output vector.

6. The control method according to claim 5, wherein ​ ​ ​ wherein ​ ​ ​ 8. The control method according to claim 1, wherein ​ a preset value of the target function is input as input data, and an output value of the target function is input as feedback data to the PID controller; a temperature adjusting device and a humidity adjusting device of the refrigeration equipment are controlled according to a deviation between the input data and the feedback data.

9. A refrigeration equipment, comprising: a collecting device configured to collect temperature and humidity in a storage compartment of the refrigeration equipment; and a control device including a processor and a memory, the memory storing a machine executable program, and the machine executable program, when executed by the processor, is configured to implement the control method according to any one of claims 1-8. ​

Images