Cooling system of liquid cooling charging pile

Abstract

The invention discloses a cooling system of a liquid cooling charging pile, which can realize high system integration, reduce cost and complexity and perform dynamic collaborative management on differentiated cooling requirements, and comprises a variable-frequency cooling liquid cooling unit, an intelligent three-way valve, a first liquid cooling loop, a second liquid cooling loop, a refrigerant storage unit and a PID (Proportion Integration Differentiation) controller, the first pipeline to the seventh pipeline are integrally and circularly connected, a refrigerant output by the frequency conversion heat dissipation liquid cooling unit is divided into the first liquid cooling loop and the second liquid cooling loop through control of the intelligent three-way valve, the first liquid cooling loop dissipates heat for the power output unit of the charging pile, and the second liquid cooling loop dissipates heat for the charging gun. The first pipeline, the second pipeline, the third pipeline, the fourth pipeline, the fifth pipeline and the seventh pipeline are each internally provided with a temperature sensor and a flow sensor, and the PID controller is configured to control the operation frequency of the frequency conversion heat dissipation liquid cooling unit and the interface opening degree of the intelligent three-way valve based on detection data of the temperature sensors and the flow sensors.

Description

Technical Field

[0001] This invention relates to the field of thermal management technology for charging equipment, and specifically to a liquid-cooled charging pile heat dissipation system. Background Technology

[0002] Currently, the heat dissipation system of high-power charging piles faces stringent requirements that are both differentiated and complex, namely, simultaneously meeting the transient high current (e.g., 200A / mm) of the charging gun cable. 2 Rapid cooling of power modules (heat dissipation area ≥ 0.5m²) 2 The temperature uniformity requirement under steady-state high heat flux density (ΔT < 15℃) is as follows. Existing technologies have the following drawbacks: First, they employ independent dual-system designs, such as configuring separate liquid cooling cycles and pump sets for the power module and charging gun respectively. While this approach attempts to meet differentiated needs, it results in bulky equipment, numerous piping interfaces, complex collaborative control, and increased costs and failure rates due to component redundancy, exhibiting significant structural flaws. Second, they use a simplified single cooling loop and fixed flow distribution, attempting to cover all heat dissipation scenarios with a single system. Although this approach improves integration, it cannot dynamically sense and respond to the real-time demands of different loads. It exhibits sluggish response when dealing with transient peaks in the charging gun line and coarse adjustments when maintaining the uniform temperature of the power module, revealing rigid heat management and low energy efficiency. Summary of the Invention

[0003] The purpose of this invention is to overcome the defects in the existing technology and provide a liquid-cooled charging pile heat dissipation system that can achieve high system integration, reduce costs and complexity, and dynamically, accurately and collaboratively manage differentiated heat dissipation needs.

[0004] To achieve the above objectives, the present invention designs a liquid-cooled charging pile heat dissipation system, comprising: Variable frequency liquid cooling unit, intelligent three-way valve, first liquid cooling circuit, second liquid cooling circuit, refrigerant storage unit and PID controller; The refrigerant outlet of the variable frequency liquid cooling unit is connected to the first port of the intelligent three-way valve through a first pipeline. The second port of the intelligent three-way valve is connected to the inlet of the first liquid cooling circuit through a second pipeline. The third port of the intelligent three-way valve is connected to the inlet of the second liquid cooling circuit through a third pipeline. The outlet of the first liquid cooling circuit is connected to a fourth pipeline. The outlet of the second liquid cooling circuit is connected to a fifth pipeline. The fourth and fifth pipelines are connected to a sixth pipeline. The sixth pipeline is connected to the inlet of the refrigerant storage unit. The outlet of the refrigerant storage unit is connected to the refrigerant inlet of the variable frequency liquid cooling unit through a seventh pipeline. The first liquid cooling circuit is used to dissipate heat for the power output unit of the charging pile, and the second liquid cooling circuit is used to dissipate heat for the charging gun. Temperature sensors and flow sensors are provided in the first, second, third, fourth, fifth and seventh pipelines to detect the temperature and flow rate of the refrigerant in the corresponding pipelines. The PID controller is communicatively connected to each temperature sensor, each flow sensor, the variable frequency liquid cooling unit, and the intelligent three-way valve. The PID controller is configured to control the operating frequency of the variable frequency liquid cooling unit and the opening degree of the intelligent three-way valve based on the detection data of each temperature sensor and each flow sensor.

[0005] Furthermore, the system also includes an electrical parameter detection device, and the PID controller is communicatively connected to the electrical parameter detection device; The electrical parameter detection device is used to detect the electrical parameter data of the charging pile power output unit and the charging gun; The PID controller is configured to control the operating frequency of the variable frequency liquid cooling unit and the opening degree of the intelligent three-way valve based on the detection data of each temperature sensor and each flow sensor, and in combination with the electrical parameter data.

[0006] Furthermore, the charging pile power output unit includes multiple power modules, each power module including a rectifier module and a switch module; The first liquid cooling circuit includes multiple microchannel cold plates that correspond one-to-one with the multiple power modules. The multiple microchannel cold plates and the multiple power modules are arranged in an alternating stacked manner to form a sandwich heat dissipation structure.

[0007] Furthermore, the charging station includes multiple charging guns; The second liquid cooling circuit includes multiple parallel heat dissipation branches, which are respectively connected to the liquid cooling heat dissipation devices of each charging gun.

[0008] Furthermore, the system also includes a moisture content detection terminal installed in the refrigerant storage unit for periodically detecting the moisture content in the refrigerant within the refrigerant storage unit; The refrigerant storage unit is equipped with controllable valves at its inlet and outlet; The PID controller is communicatively connected to the moisture content detection terminal and the controllable valve, and is configured to control the controllable valve to close and output an alarm signal when the moisture content in the refrigerant exceeds a preset threshold.

[0009] Furthermore, a flow sensor is provided in the sixth pipeline; The PID controller is configured to calculate the ratio of the flow rate of the sixth pipeline to the sum of the flow rates of the fourth and fifth pipelines based on the flow rate detection data of the fourth, fifth, and sixth pipelines; if the ratio is lower than a preset ratio threshold, it is determined that there is a leak in the system.

[0010] Furthermore, the system also includes a cloud-based operation and maintenance platform; The PID controller is communicatively connected to the cloud-based operation and maintenance platform; The PID controller is configured to upload the detection data from each temperature sensor and each flow sensor, as well as the electrical parameter data, to the cloud-based operation and maintenance platform. The cloud-based operation and maintenance platform is configured to generate a first control strategy for the variable frequency liquid cooling unit and a second control strategy for the intelligent three-way valve based on the detection data of each temperature sensor, each flow sensor and the electrical parameter data, and send the first control strategy and the second control strategy to the PID controller. The PID controller is configured to control the operating frequency of the variable frequency cooling liquid chiller and the interface opening of the intelligent three-way valve based on the detection data of each temperature sensor and each flow sensor, and in combination with the electrical parameter data. Specifically, it is configured to execute the first control strategy to control the operating frequency of the variable frequency cooling liquid chiller and execute the second control strategy to control the interface opening of the intelligent three-way valve.

[0011] Furthermore, when the cloud-based operation and maintenance platform is configured to generate the first control strategy for the variable frequency liquid cooling unit and the second control strategy for the intelligent three-way valve based on the detection data from each temperature sensor, each flow sensor, and the electrical parameter data, it is specifically configured as follows: The heat loss of the second pipeline is calculated based on the temperature difference and corresponding flow rate of the refrigerant in the second pipeline and the first pipeline; the heat loss of the third pipeline is calculated based on the temperature difference and corresponding flow rate of the refrigerant in the third pipeline and the first pipeline. Based on the temperature difference and corresponding flow rate of the refrigerant in the fourth and second pipes, the actual heat dissipation of the first liquid cooling circuit is calculated; based on the temperature difference and corresponding flow rate of the refrigerant in the fifth and third pipes, the actual heat dissipation of the second liquid cooling circuit is calculated. Based on the actual heat dissipation of the first and second liquid cooling circuits, the heat loss of the second and third pipelines, and in combination with the temperature and flow rate of the refrigerant in the seventh pipeline, the target outlet temperature that the output refrigerant of the variable frequency cooling liquid chiller unit needs to reach is calculated. Based on the target outlet temperature and the temperature of the refrigerant in the first pipeline, a first control strategy for the variable frequency liquid cooling unit is generated; and Based on the heat dissipation requirements of the first liquid cooling circuit and the second liquid cooling circuit, and combined with the heat loss of the second pipeline and the third pipeline, a second control strategy for the intelligent three-way valve is generated. The temperature differences between the second and first pipelines, the third and first pipelines, the fourth and second pipelines, the fifth and third pipelines, and the temperatures of the refrigerant in the first and seventh pipelines are determined by the detection data of the temperature sensors in each pipeline. The flow rate in each pipeline is determined by the detection data of the flow sensors in each pipeline. The heat dissipation requirement of the first liquid cooling circuit is determined based on the electrical parameter data of the charging pile power output unit detected by the electrical parameter detection device. The heat dissipation requirement of the second liquid cooling circuit is determined based on the electrical parameter data of the charging gun detected by the electrical parameter detection device.

[0012] Furthermore, the cloud-based operation and maintenance platform is also configured to store the historical detection data and historical electrical parameter data of each temperature sensor and each flow sensor in time sequence to generate a first training dataset. The pre-trained time series prediction model is trained using the first training dataset to obtain the heat load prediction model; The heat dissipation requirements of the first liquid cooling circuit and the second liquid cooling circuit are determined based on the heat load prediction model and the electrical parameter data.

[0013] Furthermore, the cloud-based operation and maintenance platform is also configured to generate the first control policy and the second control policy in the following manner: The real-time received detection data and electrical parameter data are input into the strategy formulation model, and the strategy formulation model outputs the first control strategy and the second control strategy. The strategy formulation model is trained in the following way: Construct a second training dataset, which includes historical detection data, historical electrical parameter data, and corresponding historical control strategy data arranged in time series. The historical control strategy data includes a first historical control parameter for controlling the operating frequency of the variable frequency cooling liquid chiller and a second historical control parameter for controlling the interface opening of the intelligent three-way valve. The long short-term memory network model is trained based on the second training dataset to obtain the policy formulation model; During the training process of the strategy formulation model, the heat load prediction results generated by the heat load prediction model for the corresponding historical time period in the second training dataset are input as additional features into the long short-term memory network model.

[0014] The advantages and beneficial effects of this invention are as follows: The integrated architecture of the liquid-cooled charging pile heat dissipation system of this invention solves the defects of complex structure, difficult coordination and large space occupation. By acquiring full-dimensional data on the refrigerant state in each part of the system through sensors distributed in each pipeline, and based on this, the PID controller synchronously adjusts the cooling output of the variable frequency cooling liquid cooling unit and the flow distribution of the intelligent three-way valve, thereby realizing the on-demand precise control of the overall cooling capacity and the branch flow ratio. This enables the system to perform coordinated heat dissipation management according to the different thermal load characteristics of the power unit and the charging gun. It can cope with the heat dissipation requirements of the charging pile power output unit in steady state high density and the charging gun in transient high peak value. At the same time, it allows the output of the variable frequency unit to match the actual heat load in real time, reducing the energy consumption of the system under partial load and improving the reliability and maintainability of operation. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of a liquid-cooled charging pile heat dissipation system according to the present invention; Figure 2 This is a schematic diagram of the structure of another liquid-cooled charging pile heat dissipation system of the present invention; Figure 3 This is a structural diagram showing the position of the microchannel cold plate in the first liquid cooling circuit of the present invention; Figure 4 This is a schematic diagram of the structure of another liquid-cooled charging pile heat dissipation system of the present invention; Among them, 1. Variable frequency liquid cooling unit; 2. Intelligent three-way valve; 3. First liquid cooling circuit; 4. Second liquid cooling circuit; 5. Refrigerant storage unit; 6. PID controller; 7. Cloud operation and maintenance platform; 11. First pipeline; 12. Second pipeline; 13. Third pipeline; 14. Fourth pipeline; 15. Fifth pipeline; 16. Sixth pipeline; 17. Seventh pipeline; 21, 22, 23, 24, 25, and 27 are temperature sensors; 31, 32, 33, 34, 35, 36, and 37 are flow sensors; 41. Electrical parameter detection device; 42. Microchannel cold plate; 43. Moisture content detection terminal. Detailed Implementation

[0016] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and examples. The following examples are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0017] according to Figures 1-4 As shown, the present invention is a liquid-cooled charging pile heat dissipation system, comprising: 1. Variable frequency heat dissipation liquid cooling unit; 2. Intelligent three-way valve; 3. First liquid cooling circuit; 4. Second liquid cooling circuit; 5. Refrigerant storage unit; and 6. PID controller. The refrigerant outlet of the variable frequency heat dissipation liquid cooling unit 1 is connected to the first port of the intelligent three-way valve 2 through the first pipe 11. The second port of the intelligent three-way valve 2 is connected to the inlet of the first liquid cooling circuit 3 through the second pipe 12. The third port of the intelligent three-way valve 2 is connected to the inlet of the second liquid cooling circuit 4 through the third pipe 13. The outlet of the first liquid cooling circuit 3 is connected to the fourth pipe 14. The outlet of the second liquid cooling circuit 4 is connected to the fifth pipe 15. The fourth pipe 14 and the fifth pipe 15 are connected to the sixth pipe 16. The sixth pipe 16 is connected to the inlet of the refrigerant storage unit 5. The outlet of the refrigerant storage unit 5 is connected to the refrigerant inlet of the variable frequency heat dissipation liquid cooling unit 1 through the seventh pipe 17. The first liquid cooling circuit 3 is used to dissipate heat for the power output unit of the charging pile, and the second liquid cooling circuit 4 is used to dissipate heat for the charging gun. Temperature sensors and flow sensors are installed in the first pipe 11, the second pipe 12, the third pipe 13, the fourth pipe 14, the fifth pipe 15, and the seventh pipe 17. Figure 1 Temperature sensors 21, 22, 23, 24, 25, and 27, and flow sensors 31, 32, 33, 34, 35, and 37 are used to detect the temperature and flow rate of the refrigerant in the corresponding pipelines. The PID controller 6 is communicatively connected to each temperature sensor, each flow sensor, the variable frequency heat dissipation liquid cooling unit 1, and the intelligent three-way valve 2. The PID controller 6 is configured to control the operating frequency of the variable frequency liquid cooling unit 1 and the opening degree of the interface of the intelligent three-way valve 2 based on the detection data of each temperature sensor and each flow sensor.

[0018] The liquid-cooled charging pile heat dissipation system of this invention is an intelligent precision temperature control system based on load dynamic prediction and multi-loop coordination. It decomposes the traditional single cooling loop into multiple sub-loops for different heat load characteristics and cooling requirements through an intelligent three-way valve, and uses cloud big data and PID algorithm for forward control, thereby achieving extremely high energy efficiency ratio, thermal management accuracy and system reliability.

[0019] The variable frequency liquid chiller unit acts as the refrigeration engine of the entire system, providing cooling and circulation power for the refrigerant (such as coolant). Variable frequency drive means that the compressor and water pump speeds can be dynamically adjusted according to the actual cooling capacity required by the system, rather than simple start-stop control. When the heat load is low, it automatically reduces the frequency, significantly saving energy. The variable frequency liquid chiller unit uses a built-in plate heat exchanger and cooling fan to ultimately transfer the heat generated by the equipment to the ambient air. In essence, this unit is a temperature control source with full variable frequency regulation capabilities. Its core function is to dynamically adjust its operating state according to the system's thermal management commands, including implementing the necessary cooling or heating operations on the refrigerant to ensure that the refrigerant enters the circulation at a precise target temperature. Example parameters of the variable frequency liquid chiller unit are as follows: Cooling capacity range: 5kW - 60kW; Temperature control accuracy: ±0.5°C; Compressor type: Fully enclosed scroll variable frequency compressor; Water pump: Variable frequency centrifugal pump, head: 15-40m. Communication interface protocol: Standard Modbus-RTU / MQTT protocol for integration with upper-layer controllers. Communication interfaces: RS485 / CAN / LAN.

[0020] The intelligent three-way valve is the system's flow control center. It receives instructions from the PID controller and dynamically and proportionally distributes refrigerant flow to the first and second liquid cooling circuits. For example, at the start of high-power charging at the charging station, priority is given to ensuring flow in the charging gun branch; during charging intervals or standby, flow is directed to the heat storage unit for "cold storage" or simply to maintain basic cooling of the main circuit. Example parameters of the intelligent three-way valve are as follows: Valve type: Three-way electric regulating valve (continuously adjustable, non-on / off). Control signal: 4-20mA / 0-10V DC or industrial bus signal (such as Profibus-DP).

[0021] The first liquid cooling circuit and the second liquid cooling circuit can be heat dissipation devices installed in the power output unit of the charging pile and heat dissipation devices installed in the charging gun, respectively.

[0022] Temperature and flow sensors form the system's sensing neural network, distributed across various key monitoring points. They collect high-precision temperature data in real time and upload it to the PID intelligent controller as the basis for control decisions. Monitoring points include the inlet and outlet of the liquid chiller unit and the inlet and outlet of each branch line. Additionally, sensors can be installed inside the refrigerant storage unit and on pipeline surfaces to detect ambient temperature.

[0023] The PID controller, as the control center of the system, performs real-time decision-making and algorithm execution. The PID controller can implement multi-loop PID control, with multiple independent PID algorithms built-in, each used to control the frequency of the liquid chiller, the opening of the three-way valve, etc., and can be decoupled to prevent mutual interference. In some embodiments, the PID controller can achieve feedforward control, combining load forecast data from the cloud (such as upcoming charging orders for charging piles) to adjust the three-way valve and liquid chiller in advance, achieving "demand-responsive" cooling and eliminating control lag. Furthermore, the PID controller can also perform fault diagnosis and protection, such as analyzing data like temperature, pressure, and flow rate, and executing degraded operation or safe shutdown strategies when anomalies occur (such as pipe blockage, leakage, or sensor failure). Examples of PID controller parameters are as follows: Processor: Industrial-grade multi-core ARM or x86 processor. Control Algorithm: Adaptive fuzzy PID with parameter self-tuning function. I / O Interface: Digital / analog input / output interface, supporting multi-channel communication. Predictive Control Response Time: <1 second.

[0024] In the specific control process, taking the first liquid cooling circuit (heat dissipation of the charging pile power output unit) as an example, the PID controller reads the temperature T2 and flow rate Q2 of the second pipeline (inlet of the first liquid cooling circuit) and the temperature T4 and flow rate Q4 of the fourth pipeline (outlet of the first liquid cooling circuit) in real time. By calculating the temperature rise ΔT_1 = T4 - T2, and combining the specific heat capacity c, density ρ, and flow rate Q4 of the refrigerant, the actual heat dissipation of the circuit can be estimated as Φ1 = c * ρ * Q4. This is a direct feedback on the heat dissipation demand. The calculation principle of the heat dissipation demand of the second liquid cooling circuit is similar and will not be elaborated here.

[0025] For controlling the operating frequency of the variable frequency liquid-cooled chiller, the PID controller sets a basic target: maintaining the inlet temperature T2 of the power module (first liquid-cooling circuit) within an ideal range (e.g., 35°C). When T2 rises due to insufficient heat dissipation, the controller calculates the deviation from the set value and uses a built-in PID algorithm (e.g., proportional-integral-derivative calculation) to determine the required increase in cooling capacity. This requirement is translated into instructions to increase the operating frequency of the compressor and water pump in the variable frequency liquid-cooled chiller, thereby enhancing cooling capacity and causing T2 to drop.

[0026] Meanwhile, for controlling the opening of the intelligent three-way valve's control interface, the PID controller monitors the flow balance of the entire system. Assume the total flow Q1 (first pipeline) is provided by the variable frequency unit. The controller allocates flow based on the real-time heat dissipation demand ratio of the two loops (estimated by multiplying their respective temperature rise ΔT by their flow rate). For example, if calculations show that Φ1 is significantly greater than the heat dissipation Φ2 of the second liquid cooling loop, the PID algorithm will output a command to appropriately increase the opening of the second port of the intelligent three-way valve (leading to the first liquid cooling loop) while correspondingly decreasing the opening of the third port, thus achieving dynamic flow rebalancing and ensuring that both heat sources receive refrigerant flow matching their needs.

[0027] The integrated architecture of the liquid-cooled charging pile heat dissipation system in this invention solves the defects of complex structure, difficult coordination, and large space occupation. By acquiring full-dimensional data on the refrigerant status in each part of the system through sensors distributed in each pipeline, and based on this, the PID controller synchronously adjusts the cooling output of the variable frequency liquid cooling unit and the flow distribution of the intelligent three-way valve, thereby realizing on-demand precise control of the overall cooling capacity and the branch flow ratio. This enables the system to perform coordinated heat dissipation management according to the different thermal load characteristics of the power unit and the charging gun. It can cope with the heat dissipation requirements of the charging pile power output unit in steady state high density and the charging gun in transient high peak value. At the same time, it allows the output of the variable frequency unit to match the actual heat load in real time, reducing the energy consumption of the system under partial load and improving the reliability and maintainability of operation.

[0028] To improve the response speed and accuracy of system control, a preferred embodiment of the present invention further includes an electrical parameter detection device 41, and the PID controller 6 is communicatively connected to the electrical parameter detection device 41. The electrical parameter detection device 41 is used to detect the electrical parameter data of the charging pile power output unit and the charging gun. The PID controller 6 is configured to control the operating frequency of the variable frequency cooling liquid cooling unit 1 and the interface opening of the intelligent three-way valve 2 based on the detection data of each temperature sensor and each flow sensor, and in combination with the electrical parameter data.

[0029] This invention further introduces electrical parameters (current, voltage, and power) as feedforward signals. These electrical parameters directly reflect the heat generation rate of the heat source (charging pile power output module and charging gun), and their changes are much faster than temperature changes. By deploying electrical parameter detection devices such as Hall current sensors at the power output unit and charging gun, the system can acquire current (I) and voltage (U) in real time and calculate instantaneous power (P). The PID controller integrates temperature, flow rate, and power data, and can adjust the cooling strategy in advance based on the power surge trend before the actual occurrence of heat load (temperature rise), achieving feedforward control, greatly eliminating control lag, and improving heat dissipation accuracy and response speed.

[0030] For example, consider the feedforward regulation of a variable frequency liquid-cooled chiller: when the charging gun begins high-power charging, the electrical parameter detection device immediately detects a step increase in power P2. Even though the outlet temperature T5 of the charging gun circuit has not yet significantly increased, the PID controller has already calculated the impending huge heat load based on the sudden increase in P2. The controller no longer waits for feedback from T5, but immediately outputs a command to significantly increase the operating frequency of the variable frequency liquid-cooled chiller based on the preset "power-cooling capacity" mapping model. This achieves cooling-first regulation and can suppress temperature spikes.

[0031] For the feedforward control of the intelligent three-way valve: At the instant the charging pile's power output unit is in standby mode and a charging gun is about to start charging, the controller predicts the rapid shift in heat dissipation demand based on cloud orders or standby current signals from electrical parameter detection devices. Before the charging gun is actually connected, the PID controller instructs the intelligent three-way valve to gradually adjust its opening, preemptively directing the flow to the branch of the second liquid cooling circuit. When charging actually begins, sufficient refrigerant is already in place, completely eliminating temperature fluctuation delays caused by flow redistribution.

[0032] The PID control system of this invention improves the system's response speed and control accuracy in the face of transient and high-power loads by incorporating feedforward information such as electrical parameters.

[0033] To improve the heat dissipation effect of the charging pile power module, a preferred embodiment of the present invention is as follows: Figure 3 As shown, the charging pile power output unit includes multiple power modules, each power module including a rectifier module and a switch module; the first liquid cooling circuit includes multiple microchannel cold plates 42 that correspond one-to-one with the multiple power modules, wherein the multiple microchannel cold plates and the multiple power modules are arranged in an alternating stacked manner to form a sandwich heat dissipation structure.

[0034] Power modules and microchannel cold plates are alternately stacked to form a compact sandwich structure, such as... Figure 3 As shown, each microchannel cold plate is tightly fitted between two adjacent power modules, utilizing the surface of the cold plate and the microchannel flow channels to rapidly dissipate the heat generated by the power modules.

[0035] To meet the concurrent heat dissipation requirements of multi-gun charging piles, a preferred embodiment of the present invention is that the charging pile includes multiple charging guns; the second liquid cooling circuit includes multiple parallel heat dissipation branches, which are respectively connected to the liquid cooling heat dissipation devices of each charging gun.

[0036] Each charging gun in the charging station is equipped with an independent heat dissipation branch, and the branches are connected in parallel. A smart three-way valve distributes refrigerant to the second liquid cooling circuit, simultaneously cooling multiple charging guns through the parallel branches. In some embodiments, each branch can be equipped with an independent valve to independently fine-tune the flow rate according to the real-time operating status of the corresponding charging gun (via electrical parameter sensing), ensuring that each gun receives sufficient cooling capacity in high-concurrency charging scenarios and preventing localized overheating from affecting charging power or safety.

[0037] To achieve online monitoring and predictive maintenance of refrigerant quality, a preferred embodiment of the present invention further includes a moisture content detection terminal 43 installed in the refrigerant storage unit 5 for periodically detecting the moisture content in the refrigerant within the refrigerant storage unit 5; controllable valves are provided at the inlet and outlet of the refrigerant storage unit 5; the PID controller 6 is communicatively connected to the moisture content detection terminal 43 and the controllable valves, and is configured to control the controllable valves to close and output an alarm signal when the moisture content in the refrigerant exceeds a preset threshold.

[0038] The moisture content detection terminal can use a sensor based on the dielectric constant principle to periodically detect the moisture content in the refrigerant. When the moisture content is detected to exceed a preset threshold (such as the upper limit of moisture content corresponding to a freezing point of -20℃) due to refrigerant degradation or external seepage, the PID controller immediately closes the controllable valves at the inlet and outlet of the refrigerant storage unit, isolating the degraded refrigerant within the storage unit and preventing it from entering the main circulation and damaging the system. At the same time, it issues a refrigerant replacement and maintenance alarm through an audible and visual alarm or a cloud platform to achieve preventative maintenance.

[0039] To diagnose system sealing in real time and prevent leakage, a preferred embodiment of the present invention includes a flow sensor 36 in the sixth pipeline; the PID controller 6 is configured to calculate the ratio of the flow rate of the sixth pipeline to the sum of the flow rates of the fourth and fifth pipelines based on the flow detection data of the fourth, fifth, and sixth pipelines; if the ratio is lower than a preset ratio threshold, it is determined that there is a leakage in the system.

[0040] During system operation, the flow rate Q6 of the summative return line (sixth line) is compared in real time with the sum of the flow rates (Q4+Q5) of the two branch return lines (fourth and fifth lines). Under ideal leak-free conditions, Q6 should stably approach (Q4+Q5). The PID controller continuously calculates the ratio R = Q6 / (Q4+Q5). When the ratio consistently falls below a preset proportional threshold (which can be determined based on the refrigerant flow rate intercepted by the first and second liquid cooling circuits in the system), a leakage risk is identified. The controller can immediately trigger an alarm and execute safety strategies, such as reducing system pressure or preparing for shutdown, thereby improving system safety.

[0041] To improve system computing power without increasing component costs, thereby achieving global optimization and intelligent decision-making, a preferred embodiment of the present invention further includes a cloud-based operation and maintenance platform 7; the PID controller 6 is communicatively connected to the cloud-based operation and maintenance platform 7; the PID controller 6 is configured to upload the detection data from each temperature sensor and each flow sensor, as well as the electrical parameter data, to the cloud-based operation and maintenance platform 7; the cloud-based operation and maintenance platform 7 is configured to generate a first control strategy for the variable frequency liquid cooling unit 1 and a second control strategy for the intelligent three-way valve 2 based on the detection data from each temperature sensor and each flow sensor, and the electrical parameter data, and send the first control strategy and the second control strategy to the PID controller 6; the PID controller 6 is configured to, based on the detection data from each temperature sensor and each flow sensor, and in conjunction with the electrical parameter data, control the operating frequency of the variable frequency liquid cooling unit 1 and the interface opening of the intelligent three-way valve 2, specifically execute the first control strategy to control the operating frequency of the variable frequency liquid cooling unit and execute the second control strategy to control the interface opening of the intelligent three-way valve.

[0042] This invention constructs a cloud-edge collaborative intelligent control architecture. The local PID controller is responsible for high-frequency data acquisition, command reception, and reliable execution, while the cloud-based operation and maintenance platform utilizes its powerful computing and storage capabilities to perform complex model calculations and strategy optimization. This division of labor enables the system to meet both the low-latency requirements of real-time control and achieve globally optimal decision-making based on big data and AI.

[0043] To generate precise control strategies, a preferred embodiment of the present invention is that the cloud-based operation and maintenance platform 7 is configured to generate the first control strategy for the variable frequency liquid cooling unit 1 and the second control strategy for the intelligent three-way valve 2 based on the detection data from each temperature sensor, each flow sensor, and the electrical parameter data. Specifically, it is configured as follows: The heat loss of the second pipeline is calculated based on the temperature difference and corresponding flow rate of the refrigerant in the second pipeline and the first pipeline; the heat loss of the third pipeline is calculated based on the temperature difference and corresponding flow rate of the refrigerant in the third pipeline and the first pipeline. Based on the temperature difference and corresponding flow rate of the refrigerant in the fourth and second pipes, the actual heat dissipation of the first liquid cooling circuit is calculated; based on the temperature difference and corresponding flow rate of the refrigerant in the fifth and third pipes, the actual heat dissipation of the second liquid cooling circuit is calculated. Based on the actual heat dissipation of the first and second liquid cooling circuits, the heat loss of the second and third pipelines, and in combination with the temperature and flow rate of the refrigerant in the seventh pipeline, the target outlet temperature that the output refrigerant of the variable frequency cooling liquid chiller unit needs to reach is calculated. Based on the target outlet temperature and the temperature of the refrigerant in the first pipeline, a first control strategy for the variable frequency liquid cooling unit is generated; and Based on the heat dissipation requirements of the first liquid cooling circuit and the second liquid cooling circuit, and combined with the heat loss of the second pipeline and the third pipeline, a second control strategy for the intelligent three-way valve is generated. The temperature differences between the second and first pipelines, the third and first pipelines, the fourth and second pipelines, the fifth and third pipelines, and the temperatures of the refrigerant in the first and seventh pipelines are determined by the detection data of the temperature sensors in each pipeline. The flow rate in each pipeline is determined by the detection data of the flow sensors in each pipeline. The heat dissipation requirement of the first liquid cooling circuit is determined based on the electrical parameter data of the charging pile power output unit detected by the electrical parameter detection device. The heat dissipation requirement of the second liquid cooling circuit is determined based on the electrical parameter data of the charging gun detected by the electrical parameter detection device.

[0044] After receiving real-time data, the cloud platform executes the perception-modeling-prediction-decision processing steps to generate the control strategy for the PID controller. Specifically: S1: Global Data Perception and Basic Computing.

[0045] Calculation of environmental heat loss: Compare the temperature differences between the first pipeline (T1, Q1) and the second pipeline (T2) and the third pipeline (T3) (ΔT_env21 = T2 - T1, ΔT_env31 = T3 - T1), and combine the flow rate and pipeline heat capacity parameters to estimate the heat change Φ_env (environmental heat loss of the pipeline) caused by environmental heat exchange during the process of the coolant flowing from the outlet of the liquid chiller unit through the distribution pipeline to the inlet of each heat dissipation circuit.

[0046] Actual heat dissipation performance evaluation: Compare the temperature difference between the second pipe (T2) and the fourth pipe (T4) (ΔT_loop1 = T4 - T2), and the temperature difference between the third pipe (T3) and the fifth pipe (T5) (ΔT_loop2 = T5 - T3). Combining the flow rate data of the corresponding pipes (Q2, Q4, Q3, Q5), and applying the thermodynamic formula Q = c * m * ΔT (where c is the specific heat capacity of the refrigerant and m is the mass flow rate), the heat removed by the first liquid cooling loop (power output unit) and the second liquid cooling loop (charging gun) per unit time (i.e., the actual heat dissipation Φ1, Φ2) can be accurately calculated.

[0047] System inventory and balance monitoring: By comparing the sum of the flow rates of the fourth and fifth pipes (Q4+Q5) with the flow rate of the sixth pipe (Q6), it is determined whether there is a leak in the system. At the same time, by combining the data of the seventh pipe (T7, Q7), the status of the refrigerant returning to the cold source is known.

[0048] S2: Cooling Source Demand Modeling and Target Setting Total heat load calculation: The total heat load of the system Φ_total is the sum of the heat dissipation of the two loops, and the heat exchange between the pipes and the environment is taken into account: Φ_total = Φ1 + Φ2 + Φ_env.

[0049] Cold source target temperature setting: Based on energy conservation, the total heat that the cold source (variable frequency liquid cooling unit) needs to remove is equal to Φ_total. Combining the return temperature T7 of the seventh pipe and the total flow rate Q7, the target outlet temperature T_target that the coolant needs to be cooled to can be calculated to meet the heat dissipation requirements of the next cycle. The calculation formula can be simplified to: T_target = T7 - ​​(Φ_total) / (c * ρ * Q7) (ρ is the refrigerant density).

[0050] S3: Control strategy generation.

[0051] Control strategy of the variable frequency liquid cooling unit: The first control strategy is a frequency adjustment scheme for the variable frequency liquid cooling unit to make the refrigerant temperature T1 in the first pipeline (i.e., the refrigerant temperature at the outlet of the variable frequency liquid cooling unit) approach the target outlet temperature T_target. For example, the first control strategy includes the parameter target outlet temperature T_target, which instructs the PID controller to compare the calculated target temperature T_target with the refrigerant temperature T1 in the first pipeline. The difference is used as the input of the PID controller for the control loop of the variable frequency liquid cooling unit. The compressor and water pump frequencies of the variable frequency liquid cooling unit are dynamically adjusted through an (adaptive fuzzy) PID algorithm so that T1 tracks T_target accurately and quickly.

[0052] Intelligent three-way valve interface opening control strategy: The second control strategy is a control scheme for the intelligent three-way valve interface opening generated so that the refrigerant flow distributed to the second and third pipelines can meet the heat dissipation requirements of their respective sides (i.e., the first liquid cooling circuit and the second liquid cooling circuit).

[0053] First, demand-side weighting is calculated, and the heat dissipation requirements of the two circuits are analyzed in real time. The heat load of the first liquid-cooled circuit (power output unit) is relatively stable but the power density is extremely high, requiring stringent temperature stability (ΔT < 3℃); the second liquid-cooled circuit (charging gun) may face transient high current (200A / mm). 2 The dramatic temperature rise caused by this can be mitigated by combining the electrical parameter data (P1, P2) of the power output unit and the charging gun. This allows for the early detection of power change trends, and the heat dissipation requirements of the first and second liquid cooling circuits can be predicted based on these trends.

[0054] Then, a dynamic allocation decision is made. Based on the real-time heat load ratio of the two circuits, the heat dissipation efficiency of each circuit (the heat dissipation efficiency assessment calculated by S1), and the preset priority strategy (e.g., prioritizing the cooling of the charging cable at the moment of charging start-up), the ideal flow distribution ratio of the second and third pipelines is calculated.

[0055] Finally, a second control strategy is generated. In one implementation, a second control strategy can be directly generated, including the ideal flow distribution ratio of the second and third pipelines as control parameters. This strategy instructs the PID controller to compare the refrigerant flow rates of the second and third pipelines and adjust the opening degrees of the second and third ports in the intelligent three-way valve, so that the refrigerant flow ratio of the second and third pipelines approximates the ideal flow distribution ratio. In another implementation, valve control commands can be generated based on the ideal flow distribution ratio of the second and third pipelines. That is, the ideal flow distribution ratio is converted into precise control commands for the opening degrees of the two output branch ports of the intelligent three-way valve, serving as the second control strategy to instruct the PID controller to control the intelligent three-way valve to achieve dynamic and proportional flow scheduling.

[0056] In one implementation, if the performance of the PID controller allows, the calculation process of the control strategy generated by the cloud platform in this embodiment of the invention can also be executed by the PID controller. That is, the calculation process of generating the control strategy of the variable frequency heat dissipation liquid cooling unit and the intelligent three-way valve is compiled into a corresponding algorithm or program and configured in the PID controller.

[0057] To make the calculation of heat dissipation requirements more accurate, a preferred embodiment of the present invention is that the cloud-based operation and maintenance platform 7 is further configured to store the historical detection data and historical electrical parameter data of each temperature sensor and each flow sensor in a time sequence to generate a first training dataset; use the first training dataset to train a pre-trained time series prediction model to obtain a heat load prediction model; and determine the heat dissipation requirements of the first liquid cooling circuit and the second liquid cooling circuit based on the heat load prediction model and the electrical parameter data.

[0058] The cloud-based operations and maintenance platform has a built-in time-series database that continuously stores historical data on temperature, flow rate, and power output uploaded by all charging piles. This data is timestamped to form a first training dataset containing multivariate time series data. The platform uses time-series prediction algorithms such as Long Short-Term Memory (LSTM) networks or Autoregressive Integrated Moving Average (ARIMA) models to train this dataset. The training objective is to enable the model to learn from historical data sequences to predict the thermal load trend curves of the power output unit and charging gun circuit over a future period (such as the next 5 minutes). After the model training is complete, when formulating control strategies, the cloud platform not only relies on the current real-time electrical parameter data but also calls upon this thermal load prediction model to obtain predictions of future heat dissipation needs, thereby developing more proactive control strategies. For example, it can increase the cooling capacity in advance before high-power charging is predicted to begin.

[0059] To make the generation of control strategies more intelligent, a preferred embodiment of the present invention is that the cloud-based operation and maintenance platform 7 is further configured to generate the first control strategy and the second control strategy in the following manner: inputting the real-time received detection data and electrical parameter data into the strategy formulation model, and the strategy formulation model outputting the first control strategy and the second control strategy; The strategy formulation model is trained in the following way: Construct a second training dataset, which includes historical detection data, historical electrical parameter data, and corresponding historical control strategy data arranged in time series. The historical control strategy data includes a first historical control parameter for controlling the operating frequency of the variable frequency cooling liquid chiller and a second historical control parameter for controlling the interface opening of the intelligent three-way valve. The long short-term memory network model is trained based on the second training dataset to obtain the policy formulation model; During the training process of the strategy formulation model, the heat load prediction results generated by the heat load prediction model for the corresponding historical time period in the second training dataset are input as additional features into the long short-term memory network model.

[0060] The purpose of this invention is to train an AI model (strategy formulation model) that can automatically learn the optimal control strategy, and to formulate a control strategy for the PID controller through the AI ​​model.

[0061] When training the strategy formulation model, historical data is first collected to construct a second training dataset. This second training dataset not only includes historical environmental states (all T, Q, P data), but more importantly, it includes historical control parameters that have been verified as excellent or optimal under those states (e.g., the operating frequency setpoint of the variable frequency cooling unit and the valve opening value of the three-way valve). Then, a Long Short-Term Memory (LSTM) network is used as the initial model for training. The LSTM model receives a sequence of system states over the past N time steps as input. During training, in addition to historical state data, the heat load prediction results for the next M steps within the same time period are also input as additional contextual features. In this way, the strategy formulation model can learn the control decision-making logic based on the heat load prediction results from historical data. After training, the strategy formulation model can directly map and output current and future variable frequency unit frequency control commands (first control strategy) and three-way valve opening allocation commands (second control strategy) based on the real-time system state data, achieving end-to-end intelligent control.

[0062] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0063] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A liquid-cooled charging pile heat dissipation system, characterized in that, include: Variable frequency liquid cooling unit, intelligent three-way valve, first liquid cooling circuit, second liquid cooling circuit, refrigerant storage unit and PID controller; The refrigerant outlet of the variable frequency liquid cooling unit is connected to the first port of the intelligent three-way valve through a first pipeline. The second port of the intelligent three-way valve is connected to the inlet of the first liquid cooling circuit through a second pipeline. The third port of the intelligent three-way valve is connected to the inlet of the second liquid cooling circuit through a third pipeline. The outlet of the first liquid cooling circuit is connected to a fourth pipeline. The outlet of the second liquid cooling circuit is connected to a fifth pipeline. The fourth and fifth pipelines are connected to a sixth pipeline. The sixth pipeline is connected to the inlet of the refrigerant storage unit. The outlet of the refrigerant storage unit is connected to the refrigerant inlet of the variable frequency liquid cooling unit through a seventh pipeline. The first liquid cooling circuit is used to dissipate heat for the power output unit of the charging pile, and the second liquid cooling circuit is used to dissipate heat for the charging gun. Temperature sensors and flow sensors are provided in the first, second, third, fourth, fifth and seventh pipelines to detect the temperature and flow rate of the refrigerant in the corresponding pipelines. The PID controller is communicatively connected to each temperature sensor, each flow sensor, the variable frequency liquid cooling unit, and the intelligent three-way valve. The PID controller is configured to control the operating frequency of the variable frequency liquid cooling unit and the opening degree of the intelligent three-way valve based on the detection data of each temperature sensor and each flow sensor.

2. The liquid-cooled charging pile heat dissipation system according to claim 1, characterized in that, The system also includes an electrical parameter detection device, and the PID controller is communicatively connected to the electrical parameter detection device; The electrical parameter detection device is used to detect the electrical parameter data of the charging pile power output unit and the charging gun; The PID controller is configured to control the operating frequency of the variable frequency liquid cooling unit and the opening degree of the intelligent three-way valve based on the detection data of each temperature sensor and each flow sensor, and in combination with the electrical parameter data.

3. The liquid-cooled charging pile heat dissipation system according to claim 1, characterized in that, The charging pile power output unit includes multiple power modules, and each power module includes a rectifier module and a switch module. The first liquid cooling circuit includes multiple microchannel cold plates that correspond one-to-one with the multiple power modules. The multiple microchannel cold plates and the multiple power modules are arranged in an alternating stacked manner to form a sandwich heat dissipation structure.

4. The liquid-cooled charging pile heat dissipation system according to claim 1, characterized in that, The charging station includes multiple charging guns; The second liquid cooling circuit includes multiple parallel heat dissipation branches, which are respectively connected to the liquid cooling heat dissipation devices of each charging gun.

5. The liquid-cooled charging pile heat dissipation system according to claim 1, characterized in that, The system also includes a moisture content detection terminal installed in the refrigerant storage unit for periodically detecting the moisture content in the refrigerant within the refrigerant storage unit; The refrigerant storage unit is equipped with controllable valves at its inlet and outlet; The PID controller is communicatively connected to the moisture content detection terminal and the controllable valve, and is configured to control the controllable valve to close and output an alarm signal when the moisture content in the refrigerant exceeds a preset threshold.

6. The liquid-cooled charging pile heat dissipation system according to claim 1, characterized in that, A flow sensor is installed in the sixth pipeline; The PID controller is configured to calculate the ratio of the flow rate of the sixth pipeline to the sum of the flow rates of the fourth and fifth pipelines based on the flow rate detection data of the fourth, fifth, and sixth pipelines; if the ratio is lower than a preset ratio threshold, it is determined that there is a leak in the system.

7. The liquid-cooled charging pile heat dissipation system according to claim 2, characterized in that, The system also includes a cloud-based operation and maintenance platform; The PID controller is communicatively connected to the cloud-based operation and maintenance platform; The PID controller is configured to upload the detection data from each temperature sensor and each flow sensor, as well as the electrical parameter data, to the cloud-based operation and maintenance platform. The cloud-based operation and maintenance platform is configured to generate a first control strategy for the variable frequency liquid cooling unit and a second control strategy for the intelligent three-way valve based on the detection data of each temperature sensor, each flow sensor and the electrical parameter data, and send the first control strategy and the second control strategy to the PID controller. The PID controller is configured to control the operating frequency of the variable frequency cooling liquid chiller and the interface opening of the intelligent three-way valve based on the detection data of each temperature sensor and each flow sensor, and in combination with the electrical parameter data. Specifically, it is configured to execute the first control strategy to control the operating frequency of the variable frequency cooling liquid chiller and execute the second control strategy to control the interface opening of the intelligent three-way valve.

8. The liquid-cooled charging pile heat dissipation system according to claim 7, characterized in that, When the cloud-based operation and maintenance platform is configured to generate a first control strategy for the variable frequency liquid cooling unit and a second control strategy for the intelligent three-way valve based on the detection data from each temperature sensor, each flow sensor, and the electrical parameter data, it is specifically configured as follows: The heat loss of the second pipeline is calculated based on the temperature difference and corresponding flow rate of the refrigerant in the second pipeline and the first pipeline; the heat loss of the third pipeline is calculated based on the temperature difference and corresponding flow rate of the refrigerant in the third pipeline and the first pipeline. Based on the temperature difference and corresponding flow rate of the refrigerant in the fourth and second pipes, the actual heat dissipation of the first liquid cooling circuit is calculated; based on the temperature difference and corresponding flow rate of the refrigerant in the fifth and third pipes, the actual heat dissipation of the second liquid cooling circuit is calculated. Based on the actual heat dissipation of the first and second liquid cooling circuits, the heat loss of the second and third pipelines, and in combination with the temperature and flow rate of the refrigerant in the seventh pipeline, the target outlet temperature that the output refrigerant of the variable frequency cooling liquid chiller unit needs to reach is calculated. Based on the target outlet temperature and the temperature of the refrigerant in the first pipeline, a first control strategy for the variable frequency heat dissipation liquid cooling unit is generated. as well as Based on the heat dissipation requirements of the first liquid cooling circuit and the second liquid cooling circuit, and combined with the heat loss of the second pipeline and the third pipeline, a second control strategy for the intelligent three-way valve is generated. The temperature differences between the second and first pipelines, the third and first pipelines, the fourth and second pipelines, the fifth and third pipelines, and the temperatures of the refrigerant in the first and seventh pipelines are determined by the detection data of the temperature sensors in each pipeline. The flow rate in each pipeline is determined by the detection data of the flow sensors in each pipeline. The heat dissipation requirement of the first liquid cooling circuit is determined based on the electrical parameter data of the charging pile power output unit detected by the electrical parameter detection device. The heat dissipation requirement of the second liquid cooling circuit is determined based on the electrical parameter data of the charging gun detected by the electrical parameter detection device.

9. The liquid-cooled charging pile heat dissipation system according to claim 8, characterized in that, The cloud-based operation and maintenance platform is also configured to store the historical detection data and historical electrical parameter data of each temperature sensor and each flow sensor in time sequence to generate the first training dataset. The time series prediction model is trained using the first training dataset to obtain the heat load prediction model; The heat dissipation requirements of the first liquid cooling circuit and the second liquid cooling circuit are determined based on the heat load prediction model and the electrical parameter data.

10. The liquid-cooled charging pile heat dissipation system according to claim 9, characterized in that, The cloud-based operations and maintenance platform is also configured to generate the first control policy and the second control policy in the following ways: The real-time received detection data and electrical parameter data are input into the strategy formulation model, and the strategy formulation model outputs the first control strategy and the second control strategy. The strategy formulation model is trained in the following way: Construct a second training dataset, which includes historical detection data, historical electrical parameter data, and corresponding historical control strategy data arranged in time series. The historical control strategy data includes a first historical control parameter for controlling the operating frequency of the variable frequency cooling liquid chiller and a second historical control parameter for controlling the interface opening of the intelligent three-way valve. The long short-term memory network model is trained based on the second training dataset to obtain the policy formulation model; During the training process of the strategy formulation model, the heat load prediction results generated by the heat load prediction model for the corresponding historical time period in the second training dataset are input as additional features into the long short-term memory network model.

Images