Temperature Control Method, Device, Equipment and Medium of Liquid Nitrogen Quick Freezing Machine

Abstract

The present application relates to a temperature control method, device, equipment and medium for a liquid nitrogen quick-freezing machine. The method includes: predicting the temperature of each fixed point and the corresponding opening of the liquid nitrogen regulating valve at the next time step based on the pre-trained liquid nitrogen regulating valve control model in the MPC controller according to the real-time temperature of each fixed point and the opening of the liquid nitrogen regulating valve at the current time step; using the MPC controller to optimize the temperature at the next time step and the corresponding opening of the liquid nitrogen regulating valve according to the real-time temperature and the opening of the liquid nitrogen regulating valve at the current time step to determine the optimal valve opening; sending the target value corresponding to the optimal valve opening to the execution module in the liquid nitrogen quick-freezing machine, and controlling the liquid nitrogen regulating valve in the execution module to adjust to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold. The present application can significantly reduce the delay and hysteresis of temperature control in the freezing chamber and avoid waste of liquid nitrogen.

Description

Technical Field

[0001] The present application relates to the field of liquid nitrogen quick-freezing, and particularly to a temperature control method for a liquid nitrogen quick-freezing machine, a corresponding device, an electronic device, and a computer-readable storage medium. Background Art

[0002] A liquid nitrogen quick-freezing machine is a device that uses liquid nitrogen to quickly freeze food. Its main purpose is to quickly freeze food to an extremely low temperature in order to effectively maintain the quality and freshness of food in a relatively short time. In addition, by quickly reducing the temperature of food, large ice crystals formed during the freezing process can be reduced, which is beneficial to maintaining the taste and nutritional components of food.

[0003] Currently, most temperature control systems of quick-freezing machines adopt traditional PID control methods, which are characterized by being simple and easy to implement and having high stability. However, since they cannot adjust parameters according to real-time changes in the state during the operation of the system, this results in an inability to meet the real-time requirements of system changes, easily causing overshoot and lag problems, leading to a decline in the freezing performance of the system, and also causing waste of liquid nitrogen.

[0004] In summary, in view of the problems in the prior art that the temperature control system of the quick-freezing machine is prone to overshoot and lag, resulting in a decline in the freezing performance of the system and waste of liquid nitrogen, the applicant has made corresponding explorations to solve this problem. Summary of the Invention

[0005] The purpose of the present application is to solve the above problems and provide a temperature control method for a liquid nitrogen quick-freezing machine, a corresponding device, an electronic device, and a computer-readable storage medium.

[0006] To achieve the various purposes of the present application, the present application adopts the following technical solutions:

[0007] A temperature control method for a liquid nitrogen quick-freezing machine proposed for one of the purposes of the present application includes:

[0008] In response to a temperature control instruction of the liquid nitrogen quick-freezing machine, collect historical state data in the freezing space of the liquid nitrogen quick-freezing machine, where the historical state data includes temperature values at each fixed point in the freezing space, the difference between the temperature value at each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve;

[0009] Based on the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve, a control model of the liquid nitrogen regulating valve is constructed. Among them, the control model of the liquid nitrogen regulating valve is constructed by training an RBF neural network with a particle swarm optimization algorithm. The basic network architecture of the control model of the liquid nitrogen regulating valve is an RBF neural network. The RBF neural network includes an input layer, a hidden layer, and an output layer. The input layer is the state vector x(t)=(x1,x2,...,x m ) T at the current time step that affects the temperature distribution in the freezing space. The number of input vectors is m. The state vector at the current time step is the temperature value of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve;

[0010] The hidden layer uses a Gaussian function, which is expressed as:

[0011]

[0012] In the formula, c i is the center of the i-th node, β i is the width of the i-th node, ||x - c i || is the Euclidean distance between the two, and n is the number of hidden layer nodes;

[0013] The state variable at the next time step predicted by the output layer is expressed as:

[0014]

[0015] In the formula, w is the weight, and the number of grid outputs is 2;

[0016] Based on the pre-trained control model of the liquid nitrogen regulating valve in the MPC controller, according to the real-time temperature of each fixed point and the opening degree of the liquid nitrogen regulating valve at the current time step, predict the temperature of each fixed point and the corresponding opening degree of the liquid nitrogen regulating valve at the next time step;

[0017] Use the MPC controller to optimize the temperature at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve according to the real-time temperature and the opening degree of the liquid nitrogen regulating valve at the current time step to determine the optimal valve opening;

[0018] Send the target value corresponding to the optimal valve opening to the execution module in the liquid nitrogen freezer, and control the liquid nitrogen regulating valve in the execution module to adjust to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold to complete the temperature control of the liquid nitrogen freezer.

[0019] Optionally, the step of constructing a liquid nitrogen regulating valve control model based on the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and a preset temperature, and the corresponding liquid nitrogen regulating valve opening includes:

[0020] The RBF neural network is trained using a particle swarm optimization algorithm to obtain the liquid nitrogen regulating valve control model;

[0021] Initialize the particle swarm: determine the number of particles in the swarm and the dimension of each particle, randomly initialize the position and velocity of each particle, and record the corresponding historical optimal position and global optimal position for each particle. Each particle represents a set of RBF neural network parameters.

[0022] Evaluate the fitness of the particle swarm: Use the updated RBF neural network parameters to perform forward propagation calculations on the training set to obtain the output results;

[0023] Iteratively update the position and velocity of the particles: for each particle, update the new velocity of the particle according to the current position and velocity, and update the new position of the particle according to the new velocity of the particle;

[0024] Update RBF neural network parameters: For each particle, update the corresponding RBF neural network parameters, the position of the RBF center, the width of the RBF function, and the output layer weight according to its historical optimal position;

[0025] Determine the termination condition: If the preset number of iterations is reached, or the preset fitness threshold is reached, the optimization process is terminated and the training of the liquid nitrogen regulating valve control model is completed.

[0026] Optionally, the step of optimizing the temperature at the next time step and the corresponding liquid nitrogen regulating valve opening according to the real-time temperature at the current time step and the liquid nitrogen regulating valve opening using an MPC controller to determine the optimal valve opening includes:

[0027] The RBF neural network model is:

[0028] x k+1 =A(x(k)+Bu(k)+Bg(k)+Cd(k)),

[0029] y k =Ax(k),

[0030] Among them, x(k) represents the system state, u(k) represents the opening of the liquid nitrogen control valve, g(k) represents the dynamic parameters of the control valve, and d(k) represents the external disturbance; y k represents the difference between the current temperature value and the preset temperature, A, B and C represent the system state parameter matrix;

[0031] The predicted output value y of the controlled object in the next period of time n is:

[0032] [y N (k + 1)|k), y N (k + 2)|k,..., y N (k + i)|k),... y N (k + N)|k)] T ,

[0033] where N represents the prediction horizon; y N (k + i)|k represents predicting the output at the (k + i)-th moment based on the state at the k-th moment;

[0034] Based on this, the input of the control system within the prediction horizon n is:

[0035]

[0036] u N (k + 1)|k represents predicting the control input for the (k + 1)-th moment in the future at the k-th moment, representing the opening of the liquid nitrogen regulating valve, and its corresponding desired control output r is: [r(k + 1)), r(k + 2),..., r(k + N)] T ;

[0037] The objective function for optimizing the temperature and the corresponding opening of the liquid nitrogen regulating valve at the next time step includes:

[0038] J = min u (J1, J2),

[0039]

[0040] J is the total objective function; J1 and J2 are the difference optimization function between the temperature value at the moment and the preset temperature and the optimization function of the opening of the liquid nitrogen regulating valve respectively; r1(i) is the desired control output value of the temperature difference at the i-th moment; y p (i|k) is the predicted temperature difference at the i-th moment based on the state at the k-th moment; r2(i) is the desired control output value of the opening of the liquid nitrogen regulating valve at the i-th moment; u p (i|k) is the predicted opening of the liquid nitrogen regulating valve at the i-th moment based on the state at the k-th moment;

[0041] Set the constraints: Constraints on the control input and the state:

[0042] u(k) ∈ [u(k) min , u(k) max ,

[0043] y(k) ∈ [y(k) min , y(k)max ],

[0044] u(k) min is the minimum limit of the liquid nitrogen regulating valve opening; u(k) max is the maximum limit of the liquid nitrogen regulating valve opening; y(k) min The minimum limit of the difference between the current temperature value and the preset temperature; y(k) max The maximum limit of the difference between the current temperature value and the preset temperature;

[0045] Solve the objective function to optimize the temperature in the next time step and its corresponding liquid nitrogen control valve opening:

[0046]

[0047] Get the optimal valve opening sequence u N (k+i)|k, then the target value M corresponding to the optimal valve opening is:

[0048] M=u N (k+i)|k.

[0049] Optionally, after the step of sending the target value corresponding to the optimal valve opening to an execution module in the liquid nitrogen quick freezing machine, and controlling the liquid nitrogen regulating valve in the execution module to adjust to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches a preset temperature difference threshold, the method includes:

[0050] In response to the timing monitoring instruction, it is detected whether the difference between the real-time temperature in the freezing space and the preset temperature exceeds the preset temperature difference threshold. If it exceeds, the target value corresponding to the optimal valve opening is sent to the execution module in the liquid nitrogen quick freezing machine, and the liquid nitrogen regulating valve in the execution module is controlled to be adjusted to the target value.

[0051] Optionally, after the step of collecting historical status data in the freezing space of the liquid nitrogen quick-freezing machine, the following steps are included:

[0052] In response to the data preprocessing instruction, the temperature value of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding liquid nitrogen regulating valve opening;

[0053] The temperature values of the respective fixed points, the differences between the temperature values of the respective fixed points and the preset temperatures, and the corresponding liquid nitrogen regulating valve openings are normalized based on a preset normalization algorithm, wherein the normalization algorithm is:

[0054]

[0055] Among them, x normalized is the normalized data, xmin is the minimum value of the data, x max is the maximum value of the data.

[0056] A temperature control device for a liquid nitrogen quick-freezing machine provided to meet another object of the present application, comprising:

[0057] A data acquisition module, configured to respond to a temperature control instruction of the liquid nitrogen quick-freezing machine and collect historical state data in the freezing space of the liquid nitrogen quick-freezing machine. The historical state data includes temperature values at each fixed point in the freezing space, the difference between the temperature value at each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve;

[0058] A control model construction module, configured to construct a liquid nitrogen regulating valve control model based on the temperature values at each fixed point in the freezing space, the difference between the temperature value at each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve. Among them, the liquid nitrogen regulating valve control model is constructed by training an RBF neural network with a particle swarm optimization algorithm. The basic network architecture of the liquid nitrogen regulating valve control model is an RBF neural network. The RBF neural network includes an input layer, a hidden layer, and an output layer. The input layer is the state vector x(t) = (x1, x2,..., x m ) T , the number of input vectors is m, and the state vector at the current time step is the temperature value at each fixed point in the freezing space, the difference between the temperature value at each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve;

[0059] The hidden layer adopts a Gaussian function, expressed as:

[0060]

[0061] In the formula, c i is the center of the i-th node, β i is the width of the i-th node, ||x - c i || is the Euclidean distance between the two, and n is the number of hidden layer nodes;

[0062] The state variable at the next time step predicted by the output layer, its expression is:

[0063]

[0064] In the formula, w is the weight, and the number of grid outputs is 2;

[0065] The control valve opening prediction module is set to predict the temperature of each fixed point and its corresponding liquid nitrogen control valve opening at the next time step based on the real-time temperature of each fixed point and the liquid nitrogen control valve opening at the current time step based on the pre-trained liquid nitrogen control valve control model in the MPC controller;

[0066] an optimal valve opening determination module configured to optimize the temperature of the next time step and the corresponding liquid nitrogen regulating valve opening according to the real-time temperature of the current time step and the liquid nitrogen regulating valve opening using an MPC controller to determine the optimal valve opening;

[0067] The temperature control module is configured to send the target value corresponding to the optimal valve opening to the execution module in the liquid nitrogen quick freezing machine, and control the liquid nitrogen regulating valve in the execution module to adjust to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches a preset temperature difference threshold, thereby completing the temperature control of the liquid nitrogen quick freezing machine.

[0068] An electronic device provided to meet another purpose of the present application includes a central processing unit and a memory, wherein the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the liquid nitrogen quick freezing machine temperature control method described in the present application.

[0069] A computer-readable storage medium is provided to meet another purpose of the present application, which stores a computer program implemented according to the liquid nitrogen quick-freezing machine temperature control method in the form of computer-readable instructions. When the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

[0070] Compared with the prior art, the present application addresses the problems that the temperature control system of the quick freezer in the prior art is prone to overshoot and hysteresis, resulting in a decrease in the freezing performance of the system and a waste of liquid nitrogen. The present application includes but is not limited to the following beneficial effects:

[0071] First, the liquid nitrogen quick freezer temperature control method of the present application addresses the problem that the traditional PID algorithm cannot adjust parameters in real time according to changes in system status. It proposes a method combining an improved RBF neural network and MPC. Based on the temperature distribution characteristics in the freezing chamber collected by multiple temperature sensors, the liquid nitrogen flow rate is achieved by controlling the opening of the liquid nitrogen regulating valve, thereby controlling the temperature in the freezing chamber to make the actual temperature as close as possible to the preset temperature and stably maintain it within a certain range, thereby improving the temperature field distribution and the freezing effect.

[0072] Second, for the temperature control method of the liquid nitrogen quick-freezing machine in this application, taking advantage of the fact that the RBF neural network can better capture the non-linear relationship in the temperature control system, intelligent decision-making and prediction are carried out in real time for the currently actually measured temperature difference value and temperature distribution, and the opening degree of the liquid nitrogen regulating valve is quickly adjusted; the PSO algorithm used can be used to train the parameters of the RBF neural network. By optimizing the weights and biases of the network, the network can better fit the tasks of temperature prediction and control;

[0073] Third, for the temperature control method of the liquid nitrogen quick-freezing machine in this application, based on the RBF neural network model and the prediction results, through model predictive control technology, it is possible to predict and optimize the objective function for the temperature difference and temperature distribution within a certain period in the future on the basis of considering the dynamic characteristics and constraint conditions of the system, and according to the real-time feedback information of the system, dynamically adjust the control strategy to adapt to different working conditions and temperature changes, calculate in advance the optimal valve opening sequence, and quickly adjust the valve opening, significantly reducing the delay and lag in temperature control in the freezing chamber, reducing the temperature control error, and reducing the waste of liquid nitrogen. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The above and / or additional aspects and advantages of this application will become obvious and easy to understand from the following description of the embodiments in conjunction with the drawings, where:

[0075] Figure 1 is a schematic flow chart of the temperature control method of the liquid nitrogen quick-freezing machine in the embodiment of this application;

[0076] Figure 2 is a flow block diagram of the temperature control method of the liquid nitrogen quick-freezing machine in the embodiment of this application;

[0077] Figure 3 is an exemplary architecture of the temperature control system of the liquid nitrogen quick-freezing machine in the embodiment of this application;

[0078] Figure 4 is a schematic diagram of the overall structural principle of the temperature control system of the liquid nitrogen quick-freezing machine in the embodiment of this application;

[0079] Figure 5 is a schematic diagram of the RBF neural network structure in the embodiment of this application;

[0080] Figure 6 is a flow block diagram of the PSO optimization algorithm in the embodiment of this application;

[0081] Figure 7 is a principle block diagram of the temperature control device of the liquid nitrogen quick-freezing machine in the embodiment of this application;

[0082] Figure 8 is a schematic diagram of the structure of the computer device in the embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0083] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and are not to be construed as limiting the present application.

[0084] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or wireless couplings. The term "and / or" used herein includes all or any units and all combinations of one or more associated listed items.

[0085] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0086] It will be understood by those skilled in the art that the terms "client," "terminal," and "terminal device" as used herein include both devices that are wireless signal receivers, i.e., devices that only have wireless signal receivers without transmission capabilities, and devices that have receiving and transmitting hardware capable of two-way communication over a two-way communication link. Such devices may include: cellular or other communication devices such as personal computers and tablet computers, which have single-line displays, multi-line displays, or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service), which may combine voice, data processing, fax, and / or data communication capabilities; PDA (Personal Digital Assistant), which may include a radio frequency receiver, a pager, Internet / Intranet access, a web browser, a notepad, a calendar, and / or a GPS (Global Positioning System) receiver; and conventional laptop and / or palmtop computers or other devices, which have and / or include a radio frequency receiver. As used herein, the terms "client," "terminal," or "terminal device" may be portable, transportable, installed in a vehicle (air, sea, and / or land), or adapted and / or configured to operate locally, and / or in a distributed manner, at any other location on Earth and / or in space. As used herein, the terms "client," "terminal," or "terminal device" may also refer to a communication terminal, an Internet terminal, or a music / video playback terminal, such as a PDA, an MID (Mobile Internet Device), and / or a mobile phone with a music / video playback function, or may refer to a smart TV, a set-top box, or other device.

[0087] The hardware referred to by names such as "server", "client", and "service node" in this application is essentially an electronic device with capabilities equivalent to those of a personal computer. It is a hardware device that has the necessary components revealed by the von Neumann principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device. Computer programs are stored in its memory, and the central processing unit loads the program stored in the external memory into the internal memory for execution, executes the instructions in the program, and interacts with the input and output devices to complete specific functions.

[0088] It should be noted that the concept of "server" referred to in this application can also be extended to server clusters. Based on the network deployment principles understood by those skilled in the art, the servers described should be logically divided. In physical space, these servers can be independent of each other but callable through interfaces, or integrated into a single physical computer or a computer cluster. Those skilled in the art should understand this flexibility and should not use it to constrain the implementation of the network deployment method of this application.

[0089] Unless expressly specified, one or more technical features of the present application can be deployed on a server for implementation and accessed by a client through a remote call to obtain an online service interface provided by the server, or can be directly deployed and run on a client for implementation.

[0090] Unless expressly specified otherwise, the neural network models referenced or may be referenced in this application may be deployed on a remote server and remotely called on the client, or may be deployed and directly called on a client with sufficient device capabilities. In some embodiments, when it runs on the client, its corresponding intelligence may be obtained through transfer learning to reduce the requirements for the client's hardware operating resources and avoid excessive occupation of the client's hardware operating resources.

[0091] Unless explicitly specified, the various data involved in this application can be stored remotely on a server or on a local terminal device, as long as they are suitable for being called by the technical solution of this application.

[0092] Those skilled in the art should be aware that although the various methods of this application are described based on the same concept and thus exhibit commonality, unless otherwise specified, these methods can be independently executed. Similarly, the various embodiments disclosed in this application are all based on the same inventive concept. Therefore, concepts with the same expression, as well as concepts that are appropriately transformed for convenience despite different expression, should be understood as equivalent.

[0093] Unless expressly stated to be mutually exclusive, the various embodiments disclosed in this application may be cross-combined with the relevant technical features of the various embodiments to flexibly construct new embodiments, as long as such combination does not deviate from the creative spirit of this application and can meet the needs of the prior art or resolve certain deficiencies in the prior art. Those skilled in the art should be aware of such flexibility.

[0094] See also Figure 1 、 Figure 2 、 Figure 3 as well as Figure 4 In one embodiment, the temperature control method of the liquid nitrogen quick freezing machine of the present application includes:

[0095] Step S10: In response to the temperature control instruction of the liquid nitrogen quick-freezing machine, collect the historical state data in the freezing space of the liquid nitrogen quick-freezing machine. The historical state data includes the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve.

[0096] The liquid nitrogen quick-freezing machine temperature control system in the computer terminal device can respond to the temperature control instruction of the liquid nitrogen quick-freezing machine and collect the historical state data in the freezing space of the liquid nitrogen quick-freezing machine. The historical state data includes the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve.

[0097] Specifically, after the liquid nitrogen quick-freezing machine temperature control system is put on line, it can initialize the resources of each module, set the temperature monitoring value T1 and the temperature difference reference value △T1. The preset temperature is the temperature monitoring value T1. The temperature values of each fixed point in the freezing space of the liquid nitrogen quick-freezing machine can be collected based on the sensors corresponding to each fixed point, so as to determine the temperature values of each fixed point and the difference between the temperature value of each fixed point and the temperature monitoring value T1.

[0098] In some embodiments, after the step of collecting the historical state data in the freezing space of the liquid nitrogen quick-freezing machine, it includes:

[0099] Step S101: In response to the data preprocessing instruction, determine the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve.

[0100] Step S103: Perform normalization processing on the temperature values of each fixed point, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve based on a preset normalization processing algorithm. The normalization processing algorithm is:

[0101]

[0102] where x normalized is the normalized data, x min is the minimum value of the data, and x max is the maximum value of the data.

[0103] In some embodiments, the liquid nitrogen quick-freezing machine temperature control method of the present application can be implemented based on the liquid nitrogen quick-freezing machine temperature control system. The liquid nitrogen quick-freezing machine temperature control system includes a collection module, a control unit, and an execution module.

[0104] The acquisition module is a temperature sensor; the control unit includes an RBF neural network and an MPC controller; and the execution module includes the liquid nitrogen spray equipment, which includes a spray nozzle and a liquid nitrogen control valve. The temperature sensor is used to collect temperatures at multiple fixed points within the freezing chamber. The RBF neural network and MPC controller calculate the optimal valve opening value based on the temperature data collected by the temperature sensor and the opening and closing data of the liquid nitrogen control valve, and then transmit the opening and closing execution command to the actuator. The liquid nitrogen spray equipment adjusts the opening and closing of the liquid nitrogen control valve according to the signal transmitted by the controller until the temperature at each fixed point reaches the predetermined value.

[0105] The liquid nitrogen quick-freezing machine temperature control system uses an RBF neural network as the front-end layer to model and predict the system's dynamic behavior. By learning from historical data, the neural network can estimate the system's state evolution and response characteristics, thereby providing predictions of future temperature changes and providing preliminary decision support for subsequent control. The model predictive control (MPC) algorithm serves as the system's advanced layer. Based on the state of the current time step, MPC uses the trained neural network model and preliminary predictions to predict the state variables for the next time step. Based on the predicted state variables and control objectives, a rolling optimization process is used to generate the optimal liquid nitrogen valve opening control strategy.

[0106] Step S20: constructing a liquid nitrogen regulating valve control model based on the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding liquid nitrogen regulating valve opening, wherein the liquid nitrogen regulating valve control model is constructed by training an RBF neural network using a particle swarm optimization algorithm;

[0107] After collecting historical status data from the freezing space of the liquid nitrogen quick-freezing machine, a liquid nitrogen regulating valve control model is constructed based on the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding liquid nitrogen regulating valve opening. The liquid nitrogen regulating valve control model is constructed by training an RBF neural network using a particle swarm optimization algorithm;

[0108] Specifically, see Figure 5 The basic network architecture of the liquid nitrogen regulating valve control model is RBF neural network, which is composed of a multi-input and single-output neural network, including input layer, hidden layer and output layer.

[0109] Furthermore, the input layer is the state vector x(t)=(x1, x2, ..., x m ) T, the number of input vectors is m, and the state vector of the current time step is the temperature value of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding liquid nitrogen regulating valve opening;

[0110] The hidden layer adopts a Gaussian function, which is expressed as:

[0111]

[0112] Where c i is the center of the i-th node, β i is the width of the i-th node, ||xc i || is the Euclidean distance between the two, and n is the number of hidden layer nodes;

[0113] The state variable of the next time step predicted by the output layer is expressed as:

[0114]

[0115] Where w is the weight and the number of grid outputs is 2.

[0116] For further information, see Figure 6 , using particle swarm optimization algorithm to train the RBF neural network to obtain the liquid nitrogen regulating valve control model;

[0117] Step S201, initializing the particle swarm: determining the number of particles in the swarm and the dimension of each particle, randomly initializing the position and velocity of each particle, and recording the corresponding historical optimal position and global optimal position for each particle, wherein each particle represents a set of RBF neural network parameters;

[0118] Step S203, evaluating the fitness of the particle swarm: using the updated RBF neural network parameters to perform forward propagation calculation on the training set to obtain the output result;

[0119] Specifically, the mean square error is selected as the fitness value of the particle swarm, and its calculation formula is expressed as:

[0120]

[0121] Where: N is the number of samples; y i is the predicted value of the i-th sample; t i is the true value of the i-th sample;

[0122] Step S205, iteratively update the position and velocity of the particle: for each particle, update the new velocity of the particle according to the current position and velocity, and update the new position of the particle according to the new velocity of the particle;

[0123] Specifically, the particle velocity and position update formulas include:

[0124] Velocity update formula:

[0125]

[0126] Position update formula:

[0127] p i|k+1 = p i|k + v i|k+i ,

[0128] Where: is the velocity of particle i at the k-th iteration; c1 and c2 are learning factors; rand1 and rand2 are random numbers in the range [0, 1]; is the historical best position of particle i at the k-th iteration; is the position of particle i at the k-th iteration; g(best) k is the historical best position of the population at the k-th iteration.

[0129] Step S207, update the RBF neural network parameters: For each particle, update the corresponding RBF neural network parameters, the position of the RBF center, the width of the RBF function, and the output layer weights according to its historical best position;

[0130] Step S209, judge the termination condition: If the preset number of iterations is reached, or the preset fitness threshold is reached, then terminate the optimization process and complete the training of the liquid nitrogen regulating valve control model.

[0131] Step S30, based on the pre-trained liquid nitrogen regulating valve control model in the MPC controller, predict the temperature at each fixed point and the corresponding liquid nitrogen regulating valve opening at the next time step according to the real-time temperature at each fixed point and the liquid nitrogen regulating valve opening at the current time step;

[0132] After building the liquid nitrogen regulating valve control model, based on the pre-trained liquid nitrogen regulating valve control model in the MPC controller, predict the temperature at each fixed point and the corresponding liquid nitrogen regulating valve opening at the next time step according to the real-time temperature at each fixed point and the liquid nitrogen regulating valve opening at the current time step;

[0133] Specifically, as can be seen from the above steps, the basic network architecture of the liquid nitrogen regulating valve control model is an RBF neural network. After training the various model parameters of the RBF neural network with the particle swarm optimization algorithm, it can be put into use. By inputting the temperature values of each fixed point in the freezing space at the current time step, or the difference between the temperature values of each fixed point and the preset temperature, and their corresponding liquid nitrogen regulating valve openings, the temperature of each fixed point and its corresponding liquid nitrogen regulating valve opening at the next time step can be predicted, so that the MPC controller can optimize the temperature and its corresponding liquid nitrogen regulating valve opening at the next time step according to the real-time temperature and liquid nitrogen regulating valve opening at the current time step to determine the optimal valve opening.

[0134] Step S40: Use the MPC controller to optimize the temperature and its corresponding liquid nitrogen regulating valve opening at the next time step according to the real-time temperature and liquid nitrogen regulating valve opening at the current time step to determine the optimal valve opening.

[0135] After predicting the temperature of each fixed point and its corresponding liquid nitrogen regulating valve opening at the next time step, use the MPC controller to optimize the temperature and its corresponding liquid nitrogen regulating valve opening at the next time step according to the real-time temperature and liquid nitrogen regulating valve opening at the current time step to determine the optimal valve opening.

[0136] Specifically, the step of using the MPC controller to optimize the temperature and its corresponding liquid nitrogen regulating valve opening at the next time step according to the real-time temperature and liquid nitrogen regulating valve opening at the current time step to determine the optimal valve opening includes:

[0137] Use the RBF neural network model trained by the particle swarm algorithm, that is, the liquid nitrogen regulating valve control model. The RBF neural network model is:

[0138] x k+1 = A(x(k)+Bu(k)+Bg(k)+Cd(k)),

[0139] y k = Ax(k),

[0140] where x(k) represents the system state, u(k) represents the liquid nitrogen regulating valve opening, g(k) represents the dynamic parameters of the regulating valve, and d(k) represents the external disturbance; y k represents the difference between the temperature value at the current moment and the preset temperature, and A, B, and C represent the system state parameter matrices;

[0141] The predicted output value y n of the controlled object in the next period of time is:

[0142] [y N(k + 1)|k), y N (k + 2)|k,..., y N (k + i)|k),... y N (k + N)|k)] T ,

[0143] where N represents the prediction horizon; y N (k + i)|k represents the prediction of the output at time k + i based on the state at time k;

[0144] Based on this, the input of the control system within the prediction horizon n is:

[0145]

[0146] u N (k + 1)|k represents the prediction of the control input at the (k + 1)-th future time at time k, representing the opening of the liquid nitrogen regulating valve, and its corresponding desired control output r is: [r(k + 1)), r(k + 2),..., r(k + N)] T ;

[0147] The objective function for optimizing the temperature at the next time step and the corresponding opening of the liquid nitrogen regulating valve includes:

[0148] J = min u (J1, J2),

[0149]

[0150] J is the total objective function; J1 and J2 are the difference optimization functions between the temperature value at a certain time and the preset temperature and the optimization function of the opening of the liquid nitrogen regulating valve, respectively; r1(i) is the desired control output value of the temperature difference at the i-th time; y p (i|k) is the predicted temperature difference at the i-th time based on the state at time k; r2(i) is the desired control output value of the opening of the liquid nitrogen regulating valve at the i-th time; u p (i|k) is the predicted opening value of the liquid nitrogen regulating valve at the i-th time based on the state at time k;

[0151] Set the constraints: Constraints on the control input and the state:

[0152] u(k) ∈ [u(k) min , u(k) max ,

[0153] y(k) ∈ [y(k) min , y(k) max ,

[0154] u(k) min is the minimum limit value of the opening of the liquid nitrogen regulating valve; u(k)max is the maximum limit of the liquid nitrogen regulating valve opening; y(k) min The minimum limit of the difference between the current temperature value and the preset temperature; y(k) max The maximum limit of the difference between the current temperature value and the preset temperature;

[0155] Solve the objective function to optimize the temperature in the next time step and its corresponding liquid nitrogen control valve opening:

[0156]

[0157] Get the optimal valve opening sequence u N (k+i)|k, then the target value M corresponding to the optimal valve opening is:

[0158] M=u N (k+i)|k.

[0159] According to the above steps, the target value corresponding to the optimal valve opening of the liquid nitrogen regulating valve in the liquid nitrogen freezer can be determined.

[0160] Step S50: Send the target value corresponding to the optimal valve opening to the execution module in the liquid nitrogen quick freezing machine, control the liquid nitrogen regulating valve in the execution module to adjust to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold, thereby completing the temperature control of the liquid nitrogen quick freezing machine.

[0161] After determining the target value corresponding to the optimal valve opening of the liquid nitrogen regulating valve in the liquid nitrogen quick-freezing machine, the target value corresponding to the optimal valve opening is sent to the execution module in the liquid nitrogen quick-freezing machine, and the liquid nitrogen regulating valve in the execution module is controlled to be adjusted to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold, thereby completing the temperature control of the liquid nitrogen quick-freezing machine. The system continuously monitors the system status and compares it with the temperature difference reference value ΔT1; when the difference between the real-time temperature of each fixed point and the preset temperature reaches the temperature difference reference value ΔT1 and remains stable, the control system enters a waiting state and continues monitoring.

[0162] In some embodiments, after the step of sending the target value corresponding to the optimal valve opening to an execution module in the liquid nitrogen quick freezing machine, and controlling the liquid nitrogen regulating valve in the execution module to adjust to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches a preset temperature difference threshold, the method includes:

[0163] The temperature control system of the liquid nitrogen quick-freezing machine can respond to the timing monitoring instruction, detect whether the difference between the real-time temperature and the preset temperature in the freezing space exceeds the preset temperature difference threshold. If it exceeds, the target value corresponding to the optimal valve opening is sent to the execution module in the liquid nitrogen quick-freezing machine to control the liquid nitrogen regulating valve in the execution module to be adjusted to the target value.

[0164] Compared with the prior art, for the problems in the prior art that the temperature control system of the quick-freezing machine is prone to overshoot and hysteresis, resulting in the decline of the system's freezing performance and the waste of liquid nitrogen, the present application includes but is not limited to the following beneficial effects:

[0165] First, for the problem that the traditional PID algorithm cannot adjust parameters in real time according to the change of the system state, the temperature control method of the liquid nitrogen quick-freezing machine in the present application proposes a method combining improved RBF neural network and MPC. According to the temperature distribution characteristics collected by the multi-point temperature sensor in the freezing chamber, the liquid nitrogen flow rate is achieved by controlling the opening of the liquid nitrogen regulating valve, and then the temperature in the freezing chamber is controlled, so that the actual temperature is as close as possible to the preset temperature and is stably maintained within a certain range, improving the temperature field distribution and the freezing effect.

[0166] Second, the temperature control method of the liquid nitrogen quick-freezing machine in the present application makes use of the advantage that the RBF neural network can better capture the non-linear relationship in the temperature control system to make intelligent decisions and predictions on the currently actually measured temperature difference value and temperature distribution in real time, and quickly adjust the opening of the liquid nitrogen regulating valve. The PSO algorithm used can be used to train the parameters of the RBF neural network. By optimizing the weights and biases of the network, the network can better fit the tasks of temperature prediction and control.

[0167] Third, based on the RBF neural network model and the prediction results, the temperature control method of the liquid nitrogen quick-freezing machine in the present application can predict and optimize the objective function of the temperature difference and temperature distribution in the next period of time on the basis of considering the dynamic characteristics and constraint conditions of the system through model predictive control technology, and dynamically adjust the control strategy according to the real-time feedback information of the system, adapt to different working conditions and temperature changes, calculate the optimal valve opening sequence in advance, and quickly adjust the valve opening, significantly reducing the delay and hysteresis of the temperature control in the freezing chamber, reducing the temperature control error, and reducing the waste of liquid nitrogen.

[0168] Please refer to Figure 7, A temperature control device for a liquid nitrogen quick-freezing machine provided to meet one of the purposes of this application, including a data acquisition module 1100, a control model construction module 1200, a regulating valve opening prediction module 1300, an optimal valve opening determination module 1400, and a temperature control module 1500. Among them, the data acquisition module 1100 is set to respond to the temperature control instruction of the liquid nitrogen quick-freezing machine and collect historical state data in the freezing space of the liquid nitrogen quick-freezing machine. The historical state data includes the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening of the liquid nitrogen regulating valve; the control model construction module 1200 is set to construct a liquid nitrogen regulating valve control model based on the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening of the liquid nitrogen regulating valve. Among them, the liquid nitrogen regulating valve control model is constructed by training an RBF neural network with a particle swarm optimization algorithm. The basic network architecture of the liquid nitrogen regulating valve control model is an RBF neural network. The RBF neural network includes an input layer, a hidden layer, and an output layer. The input layer is the state vector x(t)=(x1,x2,...,x m ) T , The number of input vectors is m, and the state vector of the current time step is the temperature value of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening of the liquid nitrogen regulating valve;

[0169] The hidden layer adopts a Gaussian function, expressed as:

[0170]

[0171] In the formula, c i is the center of the i-th node, β i is the width of the i-th node, ||x - c i || is the Euclidean distance between the two, and n is the number of hidden layer nodes;

[0172] The state variable of the next time step predicted by the output layer, its expression is:

[0173]

[0174] In the formula, w is the weight, and the number of grid outputs is 2;

[0175] The regulating valve opening prediction module 1300 is configured to predict the temperatures at each fixed point and the corresponding opening degrees of the liquid nitrogen regulating valve at the next time step based on the pre-trained liquid nitrogen regulating valve control model in the MPC controller according to the real-time temperatures at each fixed point and the opening degree of the liquid nitrogen regulating valve at the current time step; the optimal valve opening determination module 1400 is configured to optimize the temperatures at the next time step and the corresponding opening degrees of the liquid nitrogen regulating valve by using the MPC controller according to the real-time temperature and the opening degree of the liquid nitrogen regulating valve at the current time step to determine the optimal valve opening; the temperature control module 1500 is configured to send the target value corresponding to the optimal valve opening to the execution module in the liquid nitrogen freezer, and control the liquid nitrogen regulating valve in the execution module to be adjusted to the target value until the difference between the real-time temperatures at each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold, so as to complete the temperature control of the liquid nitrogen freezer.

[0176] Based on any embodiment of the present application, please refer to Figure 8 , another embodiment of the present application further provides an electronic device, which can be implemented by a computer device, as Figure 8 shown, the internal structure schematic diagram of the computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected through a system bus. Among them, the computer-readable storage medium of the computer device stores an operating system, a database, and computer-readable instructions. The control information sequence can be stored in the database. When the computer-readable instructions are executed by the processor, the processor can implement a temperature control method for a liquid nitrogen freezer. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device can store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute the temperature control method for the liquid nitrogen freezer of the present application. The network interface of the computer device is used to connect and communicate with the terminal. Those skilled in the art can understand that Figure 8 the structure shown in

[0177] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0177] In this embodiment, the processor is used to execute Figure 7 the specific functions of each module and its sub-modules in The memory stores the program codes and various types of data required to execute the above modules or sub-modules. The network interface is used for data transmission between the user terminal and the server. The memory in this embodiment stores the program codes and data required to execute all modules / sub-modules in the temperature control device of the liquid nitrogen freezer of the present application. The server can call the program codes and data of the server to execute the functions of all sub-modules.

[0178] The present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors are caused to execute the steps of the temperature control method of the liquid nitrogen quick-freezing machine according to any embodiment of the present application.

[0179] The present application also provides a computer program product, including a computer program / instructions. When the computer program / instructions are executed by one or more processors, the steps of the temperature control method of the liquid nitrogen quick-freezing machine according to any embodiment of the present application are implemented.

[0180] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments of the present application can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the foregoing storage medium can be a computer-readable storage medium such as a magnetic disk, an optical disc, a read-only memory (ROM), or a random access memory (RAM), etc.

[0181] The above are only some embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

[0182] In summary, based on the RBF neural network model and the prediction results, the temperature control method of the liquid nitrogen quick-freezing machine of the present application can, through model predictive control technology, predict and optimize the objective function for the temperature difference and temperature distribution within a certain period in the future on the basis of considering the dynamic characteristics and constraint conditions of the system, and dynamically adjust the control strategy according to the real-time feedback information of the system, adapt to different working conditions and temperature changes, calculate the optimal valve opening sequence in advance, and quickly adjust the valve opening, significantly reducing the delay and hysteresis of temperature control in the freezing chamber, reducing the temperature control error, and reducing the waste of liquid nitrogen.

Claims

1. A temperature control method for a liquid nitrogen quick-freezing machine, characterized in that, Including: Responding to the temperature control instruction of the liquid nitrogen quick-freezing machine, collecting the historical state data in the freezing space of the liquid nitrogen quick-freezing machine, where the historical state data includes the temperature values of each fixed point in the freezing space, the difference between the temperature values of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve; Based on the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve, a control model of the liquid nitrogen regulating valve is constructed. Among them, the control model of the liquid nitrogen regulating valve is constructed by training an RBF neural network with a particle swarm optimization algorithm. The basic network architecture of the control model of the liquid nitrogen regulating valve is an RBF neural network. The RBF neural network includes an input layer, a hidden layer, and an output layer. The input layer is the state vector x(t)=(x1,x2,...,x m ) T at the current time step that affects the temperature distribution in the freezing space. The number of input vectors is m. The state vector at the current time step is the temperature values of each fixed point in the freezing space, the difference between the temperature value of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve; The hidden layer adopts a Gaussian function, expressed as: where c i is the center of the i-th node, β i is the width of the i-th node, ||x - c i || is the Euclidean distance between them, and n is the number of hidden layer nodes; The state variable at the next time step predicted by the output layer, its expression is: In the formula, w is the weight, and the number of grid outputs is 2; Based on the pre-trained liquid nitrogen regulating valve control model in the MPC controller, according to the real-time temperature of each fixed point and the opening degree of the liquid nitrogen regulating valve at the current time step, predict the temperature of each fixed point at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve; Use the MPC controller to optimize the temperature at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve according to the real-time temperature and the opening degree of the liquid nitrogen regulating valve at the current time step to determine the optimal valve opening degree; Send the target value corresponding to the optimal valve opening degree to the execution module in the liquid nitrogen quick-freezing machine, control the liquid nitrogen regulating valve in the execution module to be adjusted to the target value until the difference between the real-time temperature of each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold to complete the temperature control of the liquid nitrogen quick-freezing machine.

2. The temperature control method of the liquid nitrogen quick-freezing machine according to claim 1, characterized in that, Based on the temperature values of each fixed point in the freezing space, the difference between the temperature values of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve, the steps of constructing the liquid nitrogen regulating valve control model include: Use the particle swarm optimization algorithm to train the RBF neural network to obtain the liquid nitrogen regulating valve control model; Initialize the particle swarm: Determine the number of the particle swarm and the dimension of each particle, randomly initialize the position and velocity of each particle, and record the corresponding historical optimal position and global optimal position for each particle, where each particle represents a set of parameters of the RBF neural network; Evaluate the fitness of the particle swarm: Use the updated RBF neural network parameters to perform forward propagation calculation on the training set to obtain the output result; Iteratively update the position and velocity of the particle: For each particle, update the new velocity of the particle according to the current position and velocity, and update the new position of the particle according to the new velocity of the particle; Update the RBF neural network parameters: For each particle, update the corresponding RBF neural network parameters, the position of the RBF center, the width of the RBF function, and the output layer weight according to its historical optimal position; Judge the termination condition: If the preset number of iterations is reached, or the preset fitness threshold is reached, terminate the optimization process and complete the training of the liquid nitrogen regulating valve control model.

3. The temperature control method of the liquid nitrogen quick-freezing machine according to claim 1, characterized in that The steps of using the MPC controller to optimize the temperature at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve according to the real-time temperature and the opening degree of the liquid nitrogen regulating valve at the current time step to determine the optimal valve opening degree include: The RBF neural network model is: x k+1 = A(x(k) + Bu(k) + Bg(k) + Cd(k)), y k = Ax(k), Among them, x(k) represents the system state, u(k) represents the opening of the liquid nitrogen regulating valve, g(k) represents the dynamic parameters of the regulating valve, and d(k) represents the external disturbance; y k represents the difference between the temperature value at the current moment and the preset temperature, and A, B, and C represent the system state parameter matrices; The predicted output value y of the controlled object in a future period of time n is as follows: [y N (k + 1)|k), y N (k + 2)|k,..., y N (k + i)|k),... y N (k + N)|k)] T , where N represents the prediction horizon; y N (k + i)|k represents the output predicted at the (k + i)-th time step using the state at the k-th time step; On this basis, the input of the control system within the prediction horizon n is obtained as: u N (k + 1)|k represents the prediction of the control input at the (k + 1)-th future moment at time k, representing the opening of the liquid nitrogen regulating valve. Its corresponding desired control output r is: [r(k + 1)), r(k + 2),..., r(k + N)] T ; The objective function for optimizing the temperature at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve includes: J = min u (J1, J2), J is the total objective function; J1 and J2 are the difference optimization function between the temperature value at a moment and the preset temperature and the optimization function of the opening of the liquid nitrogen regulating valve respectively; r1(i) is the expected control output value of the temperature difference at the i-th moment; y p (i|k) is the predicted temperature difference at the i-th moment based on the state at the k-th moment; r2(i) is the expected control output value of the opening of the liquid nitrogen regulating valve at the i-th moment; u p (i|k) is the predicted opening value of the liquid nitrogen regulating valve at the i-th moment based on the state at the k-th moment; Set constraints: Constraints on control input and state: u(k) ∈ [u(k) min , u(k) max , y(k)∈[y(k) min ,y(k) max , u(k) min is the minimum limit value of the opening degree of the liquid nitrogen regulating valve; u(k) max is the maximum limit value of the opening degree of the liquid nitrogen regulating valve; y(k) min is the minimum limit value of the difference between the temperature value at the current moment and the preset temperature; y(k) max is the maximum limit value of the difference between the temperature value at the current moment and the preset temperature; Solve the objective function for optimizing the temperature at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve: Obtain the optimal valve opening sequence u N (k + i)|k, then the target value M corresponding to the optimal valve opening is: M = u N (k + i)|k 4. The temperature control method of the liquid nitrogen quick-freezing machine according to claim 1, characterized in that After the step of sending the target value corresponding to the optimal valve opening degree to the execution module in the liquid nitrogen quick-freezing machine and controlling the liquid nitrogen regulating valve in the execution module to be adjusted to the target value until the difference between the real-time temperature at each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold, it includes: Respond to the timing monitoring instruction, detect whether the difference between the real-time temperature in the freezing space and the preset temperature exceeds the preset temperature difference threshold. If it exceeds, send the target value corresponding to the optimal valve opening degree to the execution module in the liquid nitrogen quick-freezing machine, and control the liquid nitrogen regulating valve in the execution module to be adjusted to the target value.

5. The temperature control method of the liquid nitrogen quick-freezing machine according to any one of claims 1 to 4, characterized in that, After the step of collecting the historical state data in the freezing space of the liquid nitrogen quick-freezing machine, it includes: Respond to the data preprocessing instruction, and determine the temperature values at each fixed point in the freezing space, the difference between the temperature values at each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve; Perform normalization processing on the temperature values at each fixed point, the difference between the temperature values at each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve based on a preset normalization processing algorithm. The normalization processing algorithm is: where x normalized is the normalized data, x min is the minimum value of the data, and x max is the maximum value of the data.

6. A temperature control device for a liquid nitrogen quick-freezing machine, characterized in that, Include: A data acquisition module, configured to respond to the temperature control instruction of the liquid nitrogen quick-freezing machine and collect the historical state data in the freezing space of the liquid nitrogen quick-freezing machine. The historical state data includes the temperature values at each fixed point in the freezing space, the difference between the temperature values at each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve; The control model construction module is configured to construct a liquid nitrogen regulating valve control model based on the temperature values of each fixed point in the freezing space, the difference between the temperature values of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve. Among them, the liquid nitrogen regulating valve control model is constructed by training an RBF neural network with a particle swarm optimization algorithm. The basic network architecture of the liquid nitrogen regulating valve control model is an RBF neural network. The RBF neural network includes an input layer, a hidden layer, and an output layer. The input layer is the state vector x(t) = (x1, x2,..., x m ) T of the current time step that affects the temperature distribution in the freezing space. The number of input vectors is m, and the state vector of the current time step is the temperature values of each fixed point in the freezing space, the difference between the temperature values of each fixed point and the preset temperature, and the corresponding opening degree of the liquid nitrogen regulating valve; The hidden layer uses a Gaussian function, expressed as: where c i is the center of the i-th node, β i is the width of the i-th node, ||x - c i || is the Euclidean distance between them, and n is the number of hidden layer nodes; The state variable at the next time step predicted by the output layer, its expression is: In the formula, w is the weight, and the number of grid outputs is 2; A regulating valve opening degree prediction module, configured to predict the temperature at each fixed point at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve based on the pre-trained liquid nitrogen regulating valve control model in the MPC controller according to the real-time temperature at each fixed point and the opening degree of the liquid nitrogen regulating valve at the current time step; An optimal valve opening degree determination module, configured to use the MPC controller to optimize the temperature at the next time step and the corresponding opening degree of the liquid nitrogen regulating valve according to the real-time temperature and the opening degree of the liquid nitrogen regulating valve at the current time step to determine the optimal valve opening degree; A temperature control module, configured to send the target value corresponding to the optimal valve opening degree to the execution module in the liquid nitrogen quick-freezing machine, and control the liquid nitrogen regulating valve in the execution module to be adjusted to the target value until the difference between the real-time temperature at each fixed point in the freezing space and the preset temperature reaches the preset temperature difference threshold to complete the temperature control of the liquid nitrogen quick-freezing machine.

7. An electronic device, comprising a central processing unit and a memory, characterized in that, The central processing unit is used to call and run the computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, It stores in the form of computer-readable instructions a computer program implemented according to the method according to any one of claims 1 to 5. When the computer program is called and run by the computer, it executes the steps included in the corresponding method.

Images