CN-121408883-B - Temperature control method and system for corrosion-resistant refrigeration pipe with intelligent temperature control coating

CN121408883BCN 121408883 BCN121408883 BCN 121408883BCN-121408883-B

Abstract

The invention discloses a temperature control method and a temperature control system of a corrosion-resistant refrigeration pipe with an intelligent temperature control coating, which relate to the technical field of refrigeration process control, and the method is used for collecting the temperature and flow in each electric refrigeration section pipe and the external wind speed and the temperature in the refrigeration process so as to combine a coupling heat balance equation of Fourier law and convection heat transfer and quantify the radial and axial heat transfer to calculate the heat absorption capacity; designing a physical embedded element learning graph neural network to map the heat absorption capacity into target voltage, taking the target voltage of each electric refrigerating section as the aim, cooperatively considering the influence of the power-on voltage of the electric refrigerating section on other electric refrigerating sections to design a voltage coupling conduction function, iteratively determining the optimal power-on voltage through a group intelligent algorithm, and feeding back optimization parameters according to the temperature difference; the system comprises an acquisition module, a regulation and control module and a feedback module, and respectively executes data acquisition, voltage regulation and control and parameter optimization decision, and aims to accurately control temperature, eliminate intersegmental interference and adapt to cooling requirements under multiple working conditions.

Inventors

WANG JIGANG
Yue Junting
WANG DONGSHENG

Assignees

济南明湖制冷空调设备有限公司

Dates

Publication Date: 20260512
Application Date: 20251104

Claims (10)

1. The temperature control method of the corrosion-resistant refrigeration pipe with the intelligent temperature control coating is characterized by comprising the following steps of: In the refrigerating process, collecting the temperature and the flow in the tube of each electric refrigerating section and synchronously measuring the external wind speed and the external temperature; Constructing a coupling heat balance equation according to a Fourier law and a convection heat transfer formula, calculating the heat absorption capacity of each electric refrigerating section, mapping the heat absorption capacity conversion of each electric refrigerating section into target voltage through a physical embedded element learning graph neural network, aiming at meeting the target voltage of each electric refrigerating section, designing a voltage coupling conduction function by cooperatively considering the influence of the electrifying voltage of the electric refrigerating section on other electric refrigerating sections, and iteratively determining the optimal electrifying voltage and correspondingly applying control through a group intelligent algorithm; and calculating the absolute temperature difference between the temperature in the tube of each electric refrigerating section and the preset target temperature, and comparing the absolute temperature difference with a temperature difference threshold value to decide whether to simulate and optimize so as to obtain the optimal parameters of the coupling heat balance equation and the physical embedded element learning graph neural network.
2. The method for controlling the temperature of a corrosion-resistant refrigeration tube with an intelligent temperature control coating according to claim 1, wherein the heat absorption capacity is converted into an optimal energizing voltage by combining a physical embedded element learning graph neural network with cooperative calculation of a voltage coupling conduction function, comprising the following steps: inputting the collected heat absorption capacity, the tube internal temperature, the tube internal flow and the external wind speed of each electric refrigerating section and the external temperature into a trained physical embedded element learning graph neural network to obtain target voltage; according to the heat conductivity coefficients of the corrosion-resistant base pipe and the electric cooling coating, the axial length of the electric cooling section, the coating resistance and the axial heat conduction contact area, the voltage coupling coefficient of a certain section of voltage to the temperature of other sections is calculated by combining the Fourier heat conduction law and the electric ohm law, and the voltage coupling conduction function is determined; calculating external coupling partial pressure generated by the energizing voltage of other electric refrigerating sections to the current electric refrigerating section through a voltage coupling conduction function by considering the mutual interference effect of voltages among the electric refrigerating sections; and (3) approximating the target voltage output by the physical embedded element learning graph neural network by using the superposition value of the actual energizing voltage and the external coupling partial pressure as a core target, simultaneously introducing energy consumption constraint based on Joule's law to construct an objective function, combining the upper limit constraint and the lower limit constraint of the safe working voltage of the electric refrigeration coating, guiding voltage iterative updating through a group intelligent algorithm until the objective function tends to be stable and the voltage deviation meets the preset requirement, and obtaining the optimal energizing voltage of each electric refrigeration section.
3. The method for controlling the temperature of a corrosion-resistant refrigeration tube with an intelligent temperature control coating according to claim 1, wherein the method for constructing a coupling heat balance equation and calculating the heat absorption capacity of each electric refrigeration section comprises the following steps: dividing the refrigeration tube into a plurality of independent electric refrigeration sections according to the total length of the refrigeration tube and the axial length of a single electric refrigeration section, wherein the heat of each electric refrigeration section is transmitted in the radial direction and the axial direction; respectively calculating radial total thermal resistance and axial coupling thermal resistance, namely the resistance of heat transfer along radial and axial directions; According to the principle of energy conservation, the heat absorption capacity required to be provided by the electric refrigeration coating is required to completely offset the total heat loss of the section, thereby constructing a coupling heat balance equation and calculating the heat absorption capacity required to be compensated by each electric refrigeration section.
4. The temperature control method of the corrosion-resistant refrigeration pipe with the intelligent temperature control coating according to claim 2, wherein the construction of the physical embedded element learning graph neural network to generate the target voltage comprises the following steps: Integrating the heat absorption capacity, the temperature in the tube, the flow in the tube, the external wind speed and the external temperature of each electric refrigerating section into an input characteristic vector; the coupled heat balance equation is deformed and then used as a physical constraint term, and is added into an input feature vector to generate a physical enhancement feature; Taking each electro-cooling segment as a node of the graph neural network, taking the axial heat conduction capacity between the segments as the edge weight between the nodes, and constructing an adjacent matrix to complete the topology construction of the graph neural network; aiming at the built graph neural network, pre-training is carried out by adopting historical multi-working-condition data covering different in-pipe flow, in-pipe temperature, external wind speed and external temperature scenes, and a commonality rule under each working condition is fitted to generate pre-training element parameters of the network; Inputting the physical enhancement features into a graph neural network of the pre-training element parameters, updating the node features through a three-layer feature propagation mechanism and performing iterative training; and (3) the fitted graph neural network is trained, and the characteristic vector of the last layer is linearly mapped into the target voltage without considering inter-segment coupling through the graph neural network output layer.
5. A method of controlling the temperature of a corrosion resistant coated refrigerant tube with intelligent temperature control coating as set forth in claim 3, wherein heat from each electrically cooled section is transferred in radial and axial directions, wherein the radial transfer is along the unidirectional steady state transfer of fluid in the tube, corrosion resistant basepipe, electrically cooled coating, external air, heat from inside the tube diffuses outwardly or permeates from outside into the tube, and the axial transfer is through the corrosion resistant basepipe and electrically cooled coating, the transfer direction being from the higher temperature section to the lower temperature section between adjacent two electrically cooled sections.
6. The temperature control method of the corrosion-resistant refrigeration tube with the intelligent temperature control coating according to claim 3, wherein the radial total thermal resistance is formed by connecting four parts of an in-tube convection thermal resistance, a corrosion-resistant base tube radial conduction thermal resistance, an electric refrigeration coating radial conduction thermal resistance and an external convection thermal resistance in series, the total resistance of the series thermal resistances is the sum of all the partial resistance values, wherein the in-tube convection thermal resistance is calculated by adopting a Dittus-Boelter empirical formula, the external convection thermal resistance is calculated by adopting a Qiul-Bernstein empirical formula, the corrosion-resistant base tube radial conduction thermal resistance is calculated according to the heat conduction characteristic of a cylinder wall, and the electric refrigeration coating radial conduction thermal resistance is calculated according to the simplified heat conduction characteristic of a flat wall because the thickness of the coating is far smaller than the outer diameter of the base tube.
7. A temperature control method of an anti-corrosion refrigeration tube with an intelligent temperature control coating according to claim 3, wherein the axial coupling thermal resistance is formed by connecting two parts of the axial thermal resistance of the anti-corrosion base tube and the axial thermal resistance of the electric refrigeration coating in parallel, the reciprocal of the total resistance of the parallel thermal resistances is equal to the sum of the reciprocal of each partial resistance, and the axial coupling thermal resistances of all adjacent electric refrigeration sections are equal due to the consistent structure.
8. A method of controlling the temperature of a corrosion resistant refrigeration tube with an intelligent temperature control coating according to claim 3, wherein the radial heat transfer is obtained by dividing the temperature difference between the temperature in the tube and the external temperature by the radial total thermal resistance, and if the temperature in the tube is higher than the external temperature, the heat is released, otherwise the heat is absorbed, the axial net heat transfer is the axial heat transfer of a certain electric refrigeration segment, which is equal to the axial outgoing heat of a right adjacent segment minus the axial incoming heat of a left adjacent segment, wherein the first segment has no left adjacent segment, the tail segment has no right adjacent segment, and the corresponding direction has no heat transfer.
9. The method for controlling the temperature of the corrosion-resistant cooling tube with the intelligent temperature control coating according to claim 1, wherein the corrosion-resistant cooling tube with the intelligent temperature control coating sequentially comprises a corrosion-resistant base tube, an electric cooling coating and an insulating protective layer from inside to outside, the electric cooling coating is uniformly adhered to the outer wall of the corrosion-resistant base tube, electrode groups are arranged at preset distances along the axial direction of the cooling tube, and the electric cooling section is formed between the adjacent electrode groups.
10. The temperature control system of the corrosion-resistant refrigeration pipe with the intelligent temperature control coating is characterized by comprising an acquisition module, a regulation and control module and a feedback module; The acquisition module acquires the temperature and the flow in the tube of each electric refrigerating section in each period and synchronously measures the external wind speed and the external temperature; the regulation and control module builds a coupling heat balance equation according to Fourier's law and a convection heat transfer formula in each period, calculates the heat absorption capacity of each electric refrigerating section, converts and maps the heat absorption capacity of each electric refrigerating section into target voltage through a physical embedded element learning graph neural network, aims at meeting the target voltage of each electric refrigerating section, designs a voltage coupling conduction function by cooperatively considering the influence of the energizing voltage of the electric refrigerating section on other electric refrigerating sections, iteratively determines optimal energizing voltage through a group intelligent algorithm and correspondingly applies control; And the feedback module compares the absolute temperature difference between the temperature of each electric cooling section and the preset target temperature with a temperature difference threshold value in each period to decide whether to simulate and optimize in the period so as to obtain the optimal parameters of the coupling heat balance equation and the element learning graph neural network.

Description

Temperature control method and system for corrosion-resistant refrigeration pipe with intelligent temperature control coating Technical Field The invention relates to the technical field of refrigeration process control, in particular to a temperature control method and a temperature control system of a corrosion-resistant refrigeration pipe with an intelligent temperature control coating. Background The method has the advantages that on one hand, the calculation mode of heat loss of the refrigeration tube by the existing temperature control method does not integrate key factors such as medium flow in the tube, external environment wind speed, temperature and the like, so that the calculation deviation of heat loss is large, the actual heat absorption capacity cannot be accurately reflected, on the other hand, although partial technologies introduce electric refrigeration sections, a fixed algorithm or a simple neural network is adopted when the electrified voltage is determined, the mutual influence of the voltages of all electric refrigeration sections is not quantized, the voltage and the heat absorption capacity are not matched easily due to coupling interference among sections, partial supercooling or insufficient refrigerating capacity appears, the high-precision scene requirements such as precision instrument cooling and the like are difficult to meet, and the application boundary of the corrosion-resistant refrigeration tube is restricted. Disclosure of Invention The invention aims to provide a temperature control method and a temperature control system for an anti-corrosion refrigeration pipe with an intelligent temperature control coating, which calculate the heat absorption capacity through a coupling heat balance equation, map voltage of a physical embedded element learning graph neural network and influence between quantized segments of a voltage coupling transfer function to determine the optimal power-on voltage, thereby achieving the purposes of accurately controlling the temperature, eliminating interference between segments and adapting to multi-working-condition cooling requirements. The technical scheme for realizing the purpose of the invention is as follows: In one aspect, the invention provides a temperature control method of a corrosion-resistant refrigeration pipe with an intelligent temperature control coating, comprising the following steps: In the refrigerating process, collecting the temperature and the flow in the tube of each electric refrigerating section and synchronously measuring the external wind speed and the external temperature; Constructing a coupling heat balance equation according to a Fourier law and a convection heat transfer formula, calculating the heat absorption capacity of each electric refrigerating section, mapping the heat absorption capacity conversion of each electric refrigerating section into target voltage through a physical embedded element learning graph neural network, aiming at meeting the target voltage of each electric refrigerating section, designing a voltage coupling conduction function by cooperatively considering the influence of the electrifying voltage of the electric refrigerating section on other electric refrigerating sections, and iteratively determining the optimal electrifying voltage and correspondingly applying control through a group intelligent algorithm; and calculating the absolute temperature difference between the temperature in the tube of each electric refrigerating section and the preset target temperature, and comparing the absolute temperature difference with a temperature difference threshold value to decide whether to simulate and optimize so as to obtain the optimal parameters of the coupling heat balance equation and the physical embedded element learning graph neural network. Further, through the cooperative calculation of the physical embedded element learning graph neural network and the voltage coupling transfer function, the heat absorption capacity is converted into the optimal power-on voltage, and the method comprises the following steps of: the heat absorption capacity, the in-tube temperature, the in-tube flow and the external wind speed and the external temperature of each electric cooling section calculated based on the coupling heat balance equation are input into a physical embedded element learning graph neural network pre-trained by multi-working-condition historical data, the neural network has completed pre-training by covering operation data of different in-tube flow, in-tube temperature, external wind speed and external temperature scenes, a physical constraint term obtained by deformation of the coupling heat balance equation is embedded in the network construction, and finally, a network output layer linearly maps a node characteristic vector into a target voltage of each electric cooling section without considering section-to-section coupling; According to the heat conductivity coefficient of