A high-power charging pile liquid cooling circulation heat dissipation industrial control system

Abstract

The present application relates to the technical field of industrial control system, and particularly relates to a high-power charging pile liquid cooling circulation heat dissipation industrial control system. The system comprises a data acquisition interface, an industrial controller and a control output interface, the data acquisition interface receives charging request current, actual output power, power module temperature, cooling liquid inlet temperature, cooling liquid outlet temperature, circulating pump rotating speed and refrigeration unit operating state; the industrial controller forms a thermal load sequence in a future control cycle based on the charging request current and the actual output power, forms a liquid cooling thermal inertia state queue based on the power module temperature, the cooling liquid inlet and outlet temperature and the circulating pump rotating speed, and inputs the two into a rolling constraint control process together to generate circulating pump rotating speed instructions, refrigeration unit target instructions and charging power smoothing restriction instructions. The system can match heat dissipation control with charging thermal load change and liquid cooling response lag, suppress power module temperature overshoot, and reduce sudden derating and control fluctuation.

Description

Technical Field

[0001] This invention relates to the field of industrial control system technology, and specifically to an industrial control system for liquid cooling circulation heat dissipation of high-power charging piles. Background Technology

[0002] During continuous high-current output, high-power charging piles generate concentrated heat in the power conversion unit, connecting busbars, and related electrical components. Liquid cooling circulation is a common method to maintain stable temperatures. Existing industrial control systems typically acquire data such as power module temperature, coolant inlet temperature, coolant outlet temperature, circulation pump speed, and refrigeration unit operating status via a data acquisition interface. The industrial controller then outputs circulation pump speed control commands and refrigeration unit start / stop commands according to preset temperature or temperature difference ranges. When the power module temperature is below the set value, the circulation pump maintains low-speed operation or the refrigeration unit remains in standby mode. When the power module temperature or coolant outlet temperature reaches the set value, the industrial control system increases the circulation pump speed, increases coolant flow, and activates or enhances the cooling capacity of the refrigeration unit. This approach is straightforward, with control logic primarily relying on the comparison between the currently acquired values ​​and preset boundaries. It is suitable for operating conditions where charging power changes relatively gradually and the heat load increases slowly.

[0003] In relatively recent technologies, to reduce frequent fluctuations in temperature control, some high-power charging pile liquid cooling industrial control systems add segmented speed regulation, temperature difference closed-loop, or hysteresis control on top of temperature threshold control. The industrial controller divides the circulating pump into multiple speed levels based on the changing trend of the power module temperature and the temperature difference between the coolant inlet and outlet. It increases the speed level when the temperature exceeds a higher boundary and decreases it when the temperature drops to a lower boundary. The cooling unit performs graded cooling based on the coolant outlet temperature or return temperature. Other solutions estimate the current heat generation based on the actual output power, increasing the cooling level in advance when the output power increases and decreasing it later when the output power decreases. While these control methods incorporate multiple data points such as power, temperature, and temperature difference, most still rely on the current sampling time or a short historical window as the control basis. Circulating pump commands, cooling unit commands, and charging power limits are typically generated separately, failing to incorporate future heat load, liquid cooling transmission delay, and charging power continuity into the same constraint control process.

[0004] The main technical problem with existing technologies is that during high-power charging, the charging request current and actual output power may change abruptly within a short period. Changes in the heat generated by the power module are detected before the coolant outlet and return temperatures, and the liquid cooling circuit suffers from lag due to heat capacity, flow transport, and heat transfer response. If the industrial control system still triggers control based on the current temperature or temperature difference, the increase in circulation pump speed and enhanced cooling by the refrigeration unit often lag behind the rise in heat load, causing the power module temperature to continue to rise before the control action takes effect. Simply increasing the circulation pump speed or expanding the cooling output in advance makes it difficult to determine whether the advance amount matches the liquid cooling thermal inertia, easily leading to repeated changes in control variables. If charging power limiting is only implemented after the temperature approaches the boundary, the power reduction process and the liquid cooling capacity recovery process cannot be coordinated, easily resulting in sudden derating and temperature fluctuations. Therefore, existing industrial control systems lack a coordinated control mechanism for future heat loads and liquid cooling response lag. Summary of the Invention

[0005] The purpose of this invention is to provide an industrial control system for liquid cooling circulation heat dissipation of high-power charging piles, which can effectively solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] An industrial control system for liquid cooling circulation heat dissipation of a high-power charging pile includes a data acquisition interface, an industrial controller, and a control output interface. The data acquisition interface receives charging request current, actual output power, power module temperature, coolant inlet temperature, coolant outlet temperature, circulation pump speed, and refrigeration unit operating status.

[0008] The industrial controller forms a heat load sequence for the future control cycle based on the charging request current and the actual output power, and forms a liquid-cooled thermal inertia state queue based on the power module temperature, coolant inlet temperature, coolant outlet temperature and circulating pump speed. The heat load sequence and the liquid-cooled thermal inertia state queue are input into the rolling constraint control process to generate circulating pump speed command, refrigeration unit target command and charging power smoothing limit command.

[0009] The control output interface outputs the circulating pump speed command, the refrigeration unit target command, and the charging power smoothing limit command to the corresponding execution objects, respectively.

[0010] Preferably, the industrial controller is configured with a load sequence generation process, which divides the charging request current into request current segments according to the control cycle, divides the actual output power into power segments according to the same control cycle, and extracts the change slope of adjacent request current segments and adjacent power segments respectively.

[0011] The industrial controller will align the requested current change slope with the power change slope in a time sequence, and use the actual output power as the current heat reference to form a future heat load sequence that includes the heat reference, the power change direction, and the power change amplitude.

[0012] Preferably, the industrial controller is equipped with a liquid-cooled thermal inertia modeling process. The liquid-cooled thermal inertia modeling process establishes a state record consisting of the power module temperature rise slope, the temperature difference between the inlet and outlet of the coolant, the change in the speed of the circulating pump, and the change in the return liquid temperature according to the acquisition time sequence, and writes multiple continuous state records into the liquid-cooled thermal inertia state queue.

[0013] The liquid-cooled thermal inertial state queue is set up according to the order in which heat is transferred from the power module to the coolant and reflected by the outlet temperature, so that the temperature rise information, temperature difference information and flow information at different acquisition times maintain the same transmission delay reference.

[0014] Preferably, the industrial controller is configured with a rolling constraint control process. In each control cycle, the rolling constraint control process reads the heat load sequence and the liquid cooling thermal inertia state queue, and uses the power module temperature safety boundary, the circulating pump speed change boundary, the refrigeration unit target change boundary, and the charging power decrease continuity boundary as common constraints to generate candidate circulating pump speed commands, candidate refrigeration unit target commands, and candidate charging power limit commands, and selects a combination command that satisfies the constraint relationship of the same control cycle from the candidate commands.

[0015] Preferably, the load sequence generation process further includes power request reliability correction. The industrial controller performs delay matching between the charging request current segment and the actual output power segment. During the control period when the requested current rises but the actual output power does not change synchronously, the controller marks the request lag state. During the control period when the requested current falls but the actual output power remains at the output level, the controller marks the release lag state. The controller then redistributes the adjacent heat generation increments in the future heat load sequence based on the request lag state and the release lag state.

[0016] Preferably, the load sequence generation process further includes heat load step identification, wherein the industrial controller divides the future heat load sequence into a stable segment, an increasing segment, and a decreasing segment, and extracts the initial control cycle, the number of consecutive increasing cycles, and the cumulative heat increment within the increasing segment;

[0017] The industrial controller writes the initial control cycle, the number of consecutive rising cycles, and the cumulative heat increment into the rolling constraint control process, so that the circulating pump speed command and the refrigeration unit target command are included in the candidate command calculation before the control cycle corresponding to the heat load step.

[0018] Preferably, the liquid cooling thermal inertia modeling process further includes transmission delay calibration. The industrial controller extracts the peak time of the power module temperature rise slope, the start time of the coolant outlet temperature change, and the peak time of the coolant inlet and outlet temperature difference during the period when the actual output power changes continuously. It then updates the queue length, queue advance step size, and state record weight of the liquid cooling thermal inertia state queue according to the time interval between the three, so that the liquid cooling thermal inertia state queue corresponds to the current liquid cooling cycle response process.

[0019] Preferably, the rolling constraint control process further includes smoothing limit command generation. When neither the candidate circulating pump speed command nor the candidate refrigeration unit target command can keep the predicted power module temperature within the temperature safety boundary, the industrial controller generates a charging power smoothing limit command based on the portion of the future heat load sequence that exceeds the heat dissipation capacity.

[0020] The charging power smoothing limit command includes the limit start control cycle, the allowable power value for each control cycle, and the liquid-cooled thermal inertia state queue recovery conditions required to release the limit.

[0021] Preferably, the transmission delay calibration further includes inertial state validity screening. Before updating the queue length, queue advance step size, and state record weight, the industrial controller removes state records where the circulating pump speed changes abruptly and the coolant outlet temperature does not change accordingly, removes state records where the charging request current drops abruptly and the actual output power has entered a limited state, and generates validity markers for the remaining state records according to the degree of correspondence between the power module temperature rise slope and the coolant inlet and outlet temperature difference, and updates the liquid-cooled thermal inertial state queue based on the validity markers.

[0022] Preferably, the liquid cooling thermal inertia state queue recovery conditions required to lift the restriction include temperature rise recovery conditions, temperature difference recovery conditions, and control quantity stability conditions. When the power module temperature rise slope enters the preset recovery range, the coolant inlet and outlet temperature difference enters the preset heat exchange range, and the circulating pump speed command and the refrigeration unit target command maintain convergence in the same direction within the continuous control cycle, the industrial controller recalculates the allowable power value for each control cycle according to the future heat load sequence, and writes the recalculated allowable power value into the charging power smoothing limit command.

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0024] 1. This invention forms a heat load sequence within the future control cycle from the charging request current and actual output power, and forms a liquid cooling thermal inertia state queue from the power module temperature, coolant inlet temperature, coolant outlet temperature, and circulating pump speed. Within the industrial controller, a rolling constraint control process uniformly generates circulating pump speed commands, cooling unit target commands, and charging power smoothing limit commands. Through these technical means, the industrial control system's heat dissipation control is shifted from judging current temperature exceedances to a joint judgment based on future heat load and liquid cooling response. The heat load sequence provides the heating trend resulting from charging power changes in subsequent control cycles, while the liquid cooling thermal inertia state queue characterizes the temporal relationship of heat transfer from the power module to the coolant and reflected by the outlet temperature. The rolling constraint control process incorporates both types of information into the command calculation, ensuring that the circulating pump speed increase and cooling unit target adjustment enter the control process before a significant temperature rise, and maintaining a continuous correspondence between charging power limits and liquid cooling capacity recovery. This suppresses power module temperature surges and reduces sudden derating caused by control lag.

[0025] 2. This invention further improves the stability of the liquid cooling cycle heat dissipation control process by employing techniques such as load sequence generation, power request credibility correction, thermal load step identification, transmission delay calibration, inertial state validity screening, and smoothing limit command generation. Delay matching between the requested current segment and the actual output power segment reduces misjudgments of the thermal load sequence when charging request changes are asynchronous with actual heat generation. The initial control cycle, number of consecutive rising cycles, and cumulative heat increment of the thermal load step are written into the rolling constraint control process, allowing candidate circulating pump speed commands and candidate refrigeration unit target commands to participate in calculations before the thermal load step arrives. The time interval between the peak value of the power module temperature rise slope, the start time of the coolant outlet temperature change, and the peak value of the inlet and outlet temperature difference is used to update the liquid cooling thermal inertial state queue, reducing control deviations caused by inconsistencies between the fixed delay model and the actual liquid cooling response. When the heat dissipation capacity is insufficient, the charging power smoothing limit command is updated according to the allowable power value and recovery conditions of each control cycle, ensuring that the derating and delimitation processes correspond to temperature rise recovery, temperature difference recovery, and control quantity convergence states. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating the overall closed-loop control of the high-power charging pile liquid-cooled circulating heat dissipation industrial control system of the present invention.

[0027] Figure 2 This is a flowchart of the future heat load sequence generation and power request credibility correction process of the present invention.

[0028] Figure 3 This is a flowchart of the liquid-cooled thermal inertial state queue construction and transmission delay calibration of the present invention;

[0029] Figure 4This is a flowchart of the rolling constraint control and charging power smoothing limit of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Please refer to Figure 1 This embodiment provides an industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile. The system includes a data acquisition interface, an industrial controller, and a control output interface. The data acquisition interface receives charging request current, actual output power, power module temperature, coolant inlet temperature, coolant outlet temperature, circulating pump speed, and refrigeration unit operating status. The industrial controller generates a heat load sequence for the future control cycle based on the charging request current and actual output power, and generates a liquid-cooled thermal inertia state queue based on the power module temperature, coolant inlet temperature, coolant outlet temperature, and circulating pump speed. The heat load sequence and the liquid-cooled thermal inertia state queue are input into a rolling constraint control process to generate a circulating pump speed command, a refrigeration unit target command, and a charging power smoothing limit command. The control output interface outputs the circulating pump speed command, the refrigeration unit target command, and the charging power smoothing limit command to their respective execution objects. In specific implementation... The data acquisition interface samples and registers the operating quantities from the charging control link, power conversion link, and liquid cooling circulation link under the same clock reference, so that the charging request current and the actual output power can correspond to the same control cycle, and the power module temperature, coolant inlet temperature, coolant outlet temperature, and circulation pump speed can correspond to the same heat transfer process. In each control cycle, the industrial controller first forms a heat load sequence that can represent subsequent heat generation changes, then forms a liquid cooling thermal inertia state queue that can represent coolant heat absorption and return response lag, and then calculates the combined instructions of circulation pump, refrigeration unit, and charging power limit in the same constraint control process. The control output interface converts the combined instructions into control messages or control quantities that match the execution objects. In this embodiment, heat dissipation control no longer depends on single-point actions after the current temperature exceeds the limit, but instead makes heat generation prediction, heat transfer lag, and execution control quantities correspond in the same industrial control closed loop.

[0032] In this embodiment, the data acquisition interface adds a control cycle identifier to the received charging request current, actual output power, power module temperature, coolant inlet temperature, coolant outlet temperature, circulating pump speed, and refrigeration unit operating status, and writes the data within the same control cycle into the operating status cache. After reading the operating status cache, the industrial controller performs time alignment on the charging request current and actual output power, thermal state alignment on the power module temperature and coolant inlet and outlet temperatures, and execution state alignment on the circulating pump speed and refrigeration unit operating status. The heat load sequence is formed by the current actual output power, the direction of future request current changes, and the magnitude of power changes. The liquid cooling thermal inertia state queue is formed by the power module temperature rise slope, the coolant inlet and outlet temperature difference, the circulating pump speed change, and the return liquid temperature change. The rolling constraint control process retains the instruction record of the previous control cycle after each output control instruction, so that the candidate instructions of the next control cycle can be constrained by the continuity of the control quantity changes of the previous control cycle. In this embodiment, the heating changes on the charging side, the heat absorption changes on the liquid cooling side, and the control changes on the execution side have the same time reference, reducing the control deviation caused by the inconsistency of sampling times of different data sources.

[0033] In this embodiment, the industrial control system uses the following data organization method to merge the control inputs. The data processing purpose in Table 1 does not change the physical source of the corresponding data, but only describes the processing location after entering the industrial controller and the control calculation content involved.

[0034] Table 1. Correspondence between liquid cooling cycle heat dissipation control data and processing applications.

[0035] Charging request current Charging control link Request current fragment cache Power variation direction of future heat load sequence Actual output power Power conversion link Output power clip buffer Current calorific value and calorific value increment Power module temperature Power module temperature sampling link Temperature rise slope cache Liquid-cooled thermal inertia state queue and temperature safety boundary judgment Coolant inlet temperature Liquid-cooled cyclic sampling link Inlet temperature cache Judging the temperature difference between inlet and outlet and the heat exchange status Coolant outlet temperature Liquid-cooled cyclic sampling link Export temperature buffer Transmission delay calibration and liquid return response judgment Circulating pump speed Circulating pump feedback link Rotation speed change buffer Execution residuals and candidate pump speed command boundaries Refrigeration unit operating status Refrigeration control feedback link Cooling status cache Candidate Refrigeration Unit Target Command Boundary

[0036] When the industrial controller reads data according to the processing positions shown in Table 1, it uses the same control cycle number for the request current segment cache and the output power segment cache, the same heat transfer number for the temperature rise slope cache, the inlet temperature cache and the outlet temperature cache, and the same execution number for the speed change cache and the refrigeration state cache. The control cycle number is used to form the future heat load sequence, the heat transfer number is used to form the liquid cooling thermal inertia state queue, and the execution number is used to limit the continuity of changes in the circulating pump speed command and the refrigeration unit target command. The three types of numbers are merged according to the control cycle during the rolling constraint control process. In this embodiment, a fixed data merging relationship is used to avoid the controller from repeatedly interpreting the same operating data or omitting key operating data when calculating heat load, thermal inertia and control quantities.

[0037] In one embodiment, the future heat load sequence is established as follows, wherein the heat load amount for each control cycle is determined by the normalized actual output power, the change in charging request current, and the change in actual output power:

[0038] ;

[0039] Where k represents the control cycle number. This represents the heat load corresponding to the k-th control cycle. This represents the normalized actual output power in the k-th control cycle. This represents the normalized change in charging request current. This represents the normalized charging request current in the k-th control cycle. This represents the normalized change in actual output power. This indicates the weight of the requested change in current with respect to future changes in heat generation. This represents the weight of the actual output power change with the current heat generation change. Take positive value and When the heat load is close to zero, the upward trend of the requested current is written into future cycles in advance. Take the negative value and When the value is still positive or close to zero, the heat load retains the heating reference corresponding to the actual output power. and When the heat load changes in the same direction, the heat load forms a continuous heating trend according to the common direction of change of the two. In this embodiment, the future heat load sequence reflects both the feedforward change of the charging request and is constrained by the current heating state of the actual output power.

[0040] In a preferred embodiment, reference Figure 2The industrial controller is configured with a load sequence generation process. This process divides the charging request current into request current segments according to the control cycle, and divides the actual output power into power segments according to the same control cycle. It also extracts the change slopes of adjacent request current segments and adjacent power segments. The industrial controller aligns the request current change slopes with the power change slopes in time, and uses the actual output power as the current heating reference to form a future heat load sequence that includes the heating reference, the power change direction, and the power change amplitude. Specifically, the request current segments are described using the average, final, and change direction of the request current within the start and end times of the control cycle, and the power segments are described using... The power change is described by the average actual output power, the final value, and the direction of change within the same control cycle start and end times. The slope of change is obtained by the ratio of the difference between the average values ​​of adjacent control cycles to the length of the control cycle. The timing alignment is accomplished by the control cycle number and the charging control link feedback delay mark. The heating reference is determined by the current actual output power segment. The power change direction is determined by the same or opposite relationship between the requested current segment and the power segment. The power change amplitude is defined by the slope of the requested current change and the slope of the power change. In this embodiment, the charging request change and the actual power change are treated separately to avoid directly equating the request change that has not yet been converted into heat generation with the heat generation change of the power module.

[0041] Furthermore, the load sequence generation process includes power request reliability correction, whereby the industrial controller uses the following reliability expression to determine the degree of consistency between the requested current change and the actual output power change:

[0042] ;

[0043] in, This indicates the reliability of the power request in the k-th control cycle. This indicates a positive number used to avoid a denominator of zero. and When they are in the same direction and the difference is small, Approaching a high-reliability state, the future heat load sequence distributes the heat increment according to the changes in requested current and power. For positive value When it is close to zero, When entering a low-confidence state, the industrial controller marks the request as delayed and allocates the corresponding heat increment to subsequent adjacent control cycles. For negative values When the value remains positive or close to zero, the industrial controller marks the release hysteresis state and retains the heating reference in the adjacent control cycle. In this embodiment, the request credibility is used to avoid sudden changes in the heat load sequence when the charging request current is not synchronized with the actual output power.

[0044] In a preferred embodiment, the load sequence generation process further includes heat load step identification. The industrial controller divides the future heat load sequence into a stable segment, an increasing segment, and a decreasing segment. Within the increasing segment, it extracts the initial control cycle, the number of consecutive increasing cycles, and the cumulative heat increment. The industrial controller writes the initial control cycle, the number of consecutive increasing cycles, and the cumulative heat increment into a rolling constraint control process, so that the circulating pump speed command and the refrigeration unit target command are included in the candidate command calculation before the control cycle corresponding to the heat load step. Specifically, the stable segment consists of adjacent... The rising segment consists of a continuous cycle with alternating directions of change and amplitudes within the control noise band. The downward segment is defined as a segment that continues to rise in an upward direction and whose cumulative heat generation exceeds the absorbable amount given by the current heat dissipation capacity. The system is formed when the temperature rise slope of the current power module no longer increases and the downward trend continues. The initial control cycle is used to determine the starting point of the pre-calculated parameters. The number of consecutive rising cycles is used to determine the control span maintained by the candidate command. The cumulative heat increment is used to determine the calculation intensity of the candidate circulating pump speed command and the candidate refrigeration unit target command. In this embodiment, the heat load step is separated from the ordinary fluctuation, so that the subsequent rolling constraint control can form the control candidate quantity in advance based on the time position of the step and the heat increment.

[0045] In one embodiment, reference Figure 3 The industrial controller is equipped with a liquid-cooled thermal inertia modeling process. This process establishes a state record based on the acquisition sequence, consisting of the power module temperature rise slope, the coolant inlet and outlet temperature difference, the circulating pump speed change, and the return liquid temperature change. Multiple consecutive state records are written into a liquid-cooled thermal inertia state queue. The queue positions are set according to the order in which heat is transferred from the power module to the coolant and reflected by the outlet temperature, ensuring that the temperature rise, temperature difference, and flow information at different acquisition times maintain the same transmission delay reference. Specifically, the power module temperature rise slope is used to represent the heat storage change of the power module within the current control cycle. The inlet and outlet temperature difference of the coolant is used to represent the heat removal status of the power module by the liquid cooling cycle. The change in the speed of the circulating pump is used to represent the change in the liquid cooling flow capacity. The change in the return liquid temperature is used to represent the lag change in the return of heat to the cooling side through the liquid cooling circuit. The status record enters the tail of the queue according to the acquisition time and moves to the head of the queue according to the queue advancement rules. The status record at the head of the queue participates in the rolling constraint control of the current control cycle. The status records in the middle and tail of the queue are used to describe the heat that has not yet been fully reflected in the outlet temperature. In this embodiment, the liquid cooling thermal inertia is transformed from a static delay constant into a time-series queue that can be updated with the operating status.

[0046] Furthermore, a single state record in the liquid-cooled thermal inertial state queue can be expressed in the following form:

[0047] ;

[0048] Where j represents the state record number in the liquid-cooled thermal inertial state queue. This represents the inertial state quantity of the j-th state record. Indicates the slope of the power module temperature rise. Indicates the temperature difference between the inlet and outlet of the coolant. This indicates the change in the speed of the circulating pump. This indicates the change in return liquid temperature. , , and These represent the weights of the corresponding state variables. All weights are dimensionless operational coefficients and maintain a fixed meaning under the same queue update rule. Take the rising state. When they have not yet risen synchronously, the status record indicates that the heat storage of the power module changes before the heat transfer of the coolant. It has already risen and When the change is still lagging, the status record indicates that the circulation pump control action has not yet been fully transmitted to the return liquid temperature. and When all states enter the fallback state, the state record indicates that the residual heat in the liquid cooling circuit is being released. In this embodiment, the temperature rise, temperature difference, flow and liquid return changes are included in a comparable queue record using the same inertial state quantity.

[0049] In a preferred embodiment, the liquid cooling thermal inertia modeling process further includes transmission delay calibration. The industrial controller extracts the peak time of the power module temperature rise slope, the start time of the coolant outlet temperature change, and the peak time of the coolant inlet and outlet temperature difference during the period of continuous change in actual output power. It then updates the queue length, queue advance step size, and state record weight of the liquid cooling thermal inertia state queue according to the time interval between these three parameters, ensuring that the liquid cooling thermal inertia state queue corresponds to the current liquid cooling cycle response process. The transmission delay can be obtained in the following way:

[0050]

[0051]

[0052] ;

[0053] Where m represents the number of a single continuous power change period. This represents the transmission delay corresponding to the m-th consecutive power change period. This indicates taking the median. This indicates the peak time of the power module's temperature rise slope. Indicates the starting time of the coolant outlet temperature change. This indicates the peak time of the temperature difference between the coolant inlet and outlet. Indicates the length of the control cycle. This indicates the queue length of the liquid-cooled thermal inertial state queue. This represents the advance step size of the liquid-cooled thermal inertial state queue, when and All later than At that time, the queue length expands according to the heat transfer lag time. near but If there is still a lag, the queue advancement step size retains the delay relationship corresponding to the temperature difference response. In this embodiment, the queue length and advancement rhythm are determined by the actual thermal response timing.

[0054] In a preferred embodiment, the transmission delay calibration further includes inertial state validity screening. Before updating the queue length, queue advance step size, and state record weight, the industrial controller removes state records where the circulating pump speed changes abruptly but the coolant outlet temperature does not change accordingly, and removes state records where the charging request current drops abruptly but the actual output power has entered a limited state. The remaining state records are then assigned validity markers based on the correspondence between the power module temperature rise slope and the coolant inlet / outlet temperature difference. The liquid-cooled thermal inertial state queue is updated based on these validity markers. Specifically, the circulating pump speed changes abruptly but the coolant outlet temperature does not change accordingly. The corresponding change in the state record is identified as an execution change that has not been transmitted to the observable heat transfer state. The state record in which the charging request current suddenly drops and the actual output power has entered the limited state is identified as a power limit intervention record for the thermal response. The degree of correspondence between the power module temperature rise slope and the temperature difference between the inlet and outlet of the coolant is jointly judged by the duration of the same direction change, the peak timing interval, and the consistency of the change direction. The validity mark is written into the state record and used to determine the degree of participation of the corresponding record in the queue length, queue advance step size, and state record weight update. In this embodiment, the interference of abnormal execution changes and power limit intervention on the transmission delay calibration is reduced.

[0055] In one embodiment, reference Figure 4The industrial controller is configured with a rolling constraint control process. In each control cycle, the rolling constraint control process reads the heat load sequence and the liquid cooling thermal inertia state queue, and uses the power module temperature safety boundary, the circulating pump speed change boundary, the refrigeration unit target change boundary, and the charging power decrease continuity boundary as common constraints to generate candidate circulating pump speed commands, candidate refrigeration unit target commands, and candidate charging power limit commands. It then selects a combination command from the candidate commands that satisfies the constraint relationship within the same control cycle. In specific implementation, the industrial controller will consider the future heat load sequence... In the liquid-cooled thermal inertial state queue Corresponding to the same control cycle window, the power module temperature is calculated and predicted based on the current power module temperature and heat load. Several candidate circulating pump speed commands are generated based on the circulating pump speed change boundary, several candidate cooling unit target commands are generated based on the cooling unit target change boundary, and several candidate charging power limit commands are generated based on the charging power decrease continuity boundary. The candidate commands are combined and verified within the same control cycle. Combinations that cannot simultaneously satisfy the temperature safety boundary and the control quantity change boundary are excluded. In this embodiment, the circulating pump, cooling unit, and charging power limit no longer form mutually unverified commands, but are coordinated and selected within the same constraint window.

[0056] Furthermore, the rolling constraint control process can use the following evaluation function to filter candidate combination instructions:

[0057] ;

[0058] in, This represents the evaluation value of the candidate combination in the k-th control period. This represents the predicted power module temperature during the k-th control cycle. Indicates the temperature safety boundary of the power module. This represents the candidate circulating pump speed command for the k-th control cycle. This indicates the circulating pump speed command from the previous control cycle. This represents the target instruction for the candidate refrigeration unit in the k-th control cycle. This indicates the target command for the refrigeration unit in the previous control cycle. This represents the candidate allowable charging power value for the k-th control cycle. This indicates the allowable charging power value of the previous control cycle. , , and Indicates dimensionless evaluation weight. This means that when the predicted temperature exceeds the safety boundary, the excess amount is taken, and when it does not exceed the safety boundary, the value is zero. Below The evaluation metric is mainly determined by the continuity of instruction changes. Higher than The evaluation metric will include the temperature exceeding the threshold in the candidate combination screening. , and Compared to the previous control cycle where the evaluation quantity increased synchronously due to excessive changes, this embodiment incorporates the temperature boundary and the continuity of the control quantity into the same candidate instruction screening process by unifying the evaluation quantity.

[0059] In a preferred embodiment, when calculating the combined instructions, the rolling constraint control process uses the heat load step start control period in the future heat load sequence as the advance calculation position of the candidate circulating pump speed instruction and the candidate refrigeration unit target instruction, uses the transmission delay in the liquid cooling thermal inertia state queue as the delayed position of the candidate instruction applied to the predicted power module temperature, and uses the power request credibility as the allocation ratio for the heat load increment entering the prediction window. Specifically, when the future heat load sequence is in a stable section, the industrial controller only allows the candidate circulating pump speed instruction and the candidate refrigeration unit target instruction to be adjusted within a small continuous change range. When the future heat load sequence is in an upward section and the liquid cooling thermal inertia state queue shows a lagging outlet temperature response, the industrial controller adds the candidate values ​​of the circulating pump speed and the candidate values ​​of the refrigeration unit target to the evaluation function in advance. When the future heat load sequence is in a downward section and the liquid cooling thermal inertia state queue still shows that the return liquid temperature is in a lagging release state, the industrial controller delays the release of the candidate refrigeration unit target instruction and maintains the continuous recovery of the charging power allowable value. In this embodiment, advance control, delayed action, and power smoothing limitation are all in the same rolling control logic.

[0060] In a preferred embodiment, the rolling constraint control process further includes the generation of a smoothing limit instruction. When neither the candidate circulating pump speed instruction nor the candidate cooling unit target instruction can keep the predicted power module temperature within the temperature safety boundary, the industrial controller generates a charging power smoothing limit instruction based on the portion of the future heat load sequence that exceeds the heat dissipation capacity. The charging power smoothing limit instruction includes the limit start control period, the allowable power value for each control period, and the liquid cooling thermal inertia state queue recovery conditions required to release the limit. In specific implementation, the industrial controller first calculates the corresponding heat dissipation based on the candidate circulating pump speed instruction and the candidate cooling unit target instruction. The load capacity is then used to mark the portion of the future heat load sequence that exceeds the heat dissipation capacity as heat loads to be reduced. The initial control cycle of the restriction is determined by the control cycle of the first occurrence of the heat load to be reduced and the transmission delay. The allowable power value of each control cycle is generated according to the distribution of the heat load to be reduced in subsequent control cycles. The liquid cooling thermal inertia state queue recovery condition required to lift the restriction is determined by the convergence state of the power module temperature rise slope, the temperature difference between the inlet and outlet of the coolant, and the circulating pump speed command and the refrigeration unit target command. In this embodiment, the power restriction is transformed from a single burst action into a continuous control command with an initial cycle, process allowable value and recovery condition.

[0061] Furthermore, the liquid-cooled thermal inertia state queue recovery conditions required to lift the restrictions include temperature rise recovery conditions, temperature difference recovery conditions, and control quantity stability conditions. When the power module temperature rise slope enters the preset recovery range, the coolant inlet and outlet temperature difference enters the preset heat exchange range, and the circulating pump speed command and the refrigeration unit target command maintain convergence in the same direction within the continuous control cycle, the industrial controller recalculates the allowable power value for each control cycle according to the future heat load sequence, and writes the recalculated allowable power value into the charging power smoothing limit command. The allowable power value can be updated in the following manner:

[0062]

[0063] ;

[0064] in, This represents the recovery determination value in the k-th control cycle. This represents a decision function that takes the value 1 when the temperature recovery condition is met and 0 when it is not met. This represents a decision function that takes the value 1 when the temperature difference recovery condition is met and 0 when it is not met. This represents a decision function that takes the value 1 when the stability condition of the control quantity is met and 0 when it is not met. This indicates the allowable charging power value for the next control cycle. This indicates the requested power value for the next control cycle. This indicates the allowable step size for power value recovery. This indicates taking the smaller value when the temperature rise recovery condition, temperature difference recovery condition, and control quantity stability condition are all simultaneously met. Take 1 and allow Recovery is performed under the requested power value constraint; if any recovery condition is not met... Take 0 and make To maintain the current allowable power value, this embodiment binds the unrestriction process to the recovery state of the thermal inertia state queue to prevent the allowable power value from rising before the liquid cooling circuit recovers.

[0065] In this embodiment, the industrial control system cyclically performs data acquisition, merging, sequence generation, queue updating, candidate instruction filtering, and control output according to the control cycle during its complete operation. The data acquisition interface receives and marks the charging request current, actual output power, power module temperature, coolant inlet temperature, coolant outlet temperature, circulating pump speed, and refrigeration unit operating status. The industrial controller forms a future heat load sequence based on the requested current and power segments, and determines the time distribution of heat generation increments through power request reliability correction and heat load step identification. The industrial controller forms a liquid cooling thermal inertia state queue based on the power module temperature rise slope, coolant inlet and outlet temperature difference, circulating pump speed change, and return liquid temperature change, and performs transmission delay calibration and inertia state effective... The industrial controller filters and updates the queue length, queue advance step size, and state record weight. It inputs the future heat load sequence and the liquid cooling thermal inertia state queue into the rolling constraint control process. Based on the temperature safety boundary, the circulation pump speed change boundary, the refrigeration unit target change boundary, and the charging power decrease continuity boundary, it filters and combines instructions. When the heat dissipation capacity is insufficient, it generates a charging power smoothing limit instruction that includes a limit start control cycle, an allowable power value for each control cycle, and a recovery condition. The control output interface outputs the circulation pump speed instruction, the refrigeration unit target instruction, and the charging power smoothing limit instruction to the circulation pump, refrigeration unit, and charging power control links, respectively. In this embodiment, a closed-loop control path is formed from future heat load prediction to liquid cooling thermal inertia compensation and then to rolling constraint output.

Claims

1. An industrial control system for liquid-cooled circulating heat dissipation of high-power charging piles, characterized in that, It includes a data acquisition interface, an industrial controller, and a control output interface. The data acquisition interface receives charging request current, actual output power, power module temperature, coolant inlet temperature, coolant outlet temperature, circulating pump speed, and refrigeration unit operating status. The industrial controller forms a heat load sequence for the future control cycle based on the charging request current and the actual output power, and forms a liquid-cooled thermal inertia state queue based on the power module temperature, coolant inlet temperature, coolant outlet temperature and circulating pump speed. The heat load sequence and the liquid-cooled thermal inertia state queue are input into the rolling constraint control process to generate circulating pump speed command, refrigeration unit target command and charging power smoothing limit command. The control output interface outputs the circulating pump speed command, the refrigeration unit target command, and the charging power smoothing limit command to the corresponding execution objects, respectively.

2. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 1, characterized in that, The industrial controller is equipped with a load sequence generation process, which divides the charging request current into request current segments according to the control cycle, divides the actual output power into power segments according to the same control cycle, and extracts the change slope of adjacent request current segments and adjacent power segments respectively. The industrial controller will align the requested current change slope with the power change slope in a time sequence, and use the actual output power as the current heat reference to form a future heat load sequence that includes the heat reference, the power change direction, and the power change amplitude.

3. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 1, characterized in that, The industrial controller is equipped with a liquid-cooled thermal inertia modeling process. The liquid-cooled thermal inertia modeling process establishes a state record consisting of the power module temperature rise slope, the temperature difference between the inlet and outlet of the coolant, the change in the speed of the circulating pump, and the change in the return liquid temperature according to the acquisition time sequence, and writes multiple continuous state records into the liquid-cooled thermal inertia state queue. The liquid-cooled thermal inertial state queue is set up according to the order in which heat is transferred from the power module to the coolant and reflected by the outlet temperature, so that the temperature rise information, temperature difference information and flow information at different acquisition times maintain the same transmission delay reference.

4. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 1, characterized in that, The industrial controller is equipped with a rolling constraint control process. In each control cycle, the rolling constraint control process reads the heat load sequence and the liquid cooling thermal inertia state queue, and uses the power module temperature safety boundary, the circulating pump speed change boundary, the refrigeration unit target change boundary, and the charging power decrease continuity boundary as common constraints to generate candidate circulating pump speed commands, candidate refrigeration unit target commands, and candidate charging power limit commands. Then, it selects a combination command that satisfies the constraint relationship of the same control cycle from the candidate commands.

5. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 2, characterized in that, The load sequence generation process further includes power request reliability correction. The industrial controller performs delay matching between the charging request current segment and the actual output power segment. During the control period when the requested current rises but the actual output power does not change synchronously, the request lag state is marked. During the control period when the requested current falls but the actual output power remains at the output, the release lag state is marked. The adjacent heat generation increments in the future heat load sequence are redistributed according to the request lag state and the release lag state.

6. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 2, characterized in that, The load sequence generation process further includes heat load step identification. The industrial controller divides the future heat load sequence into a stable segment, an increasing segment, and a decreasing segment, and extracts the initial control cycle, the number of consecutive increasing cycles, and the cumulative heat increment within the increasing segment. The industrial controller writes the initial control cycle, the number of consecutive rising cycles, and the cumulative heat increment into the rolling constraint control process, so that the circulating pump speed command and the refrigeration unit target command are included in the candidate command calculation before the control cycle corresponding to the heat load step.

7. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 3, characterized in that, The liquid cooling thermal inertia modeling process further includes transmission delay calibration. The industrial controller extracts the peak moment of the power module temperature rise slope, the start moment of the coolant outlet temperature change, and the peak moment of the coolant inlet and outlet temperature difference during the period when the actual output power changes continuously. It then updates the queue length, queue advance step size, and state record weight of the liquid cooling thermal inertia state queue according to the time interval between the three, so that the liquid cooling thermal inertia state queue corresponds to the current liquid cooling cycle response process.

8. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 4, characterized in that, The rolling constraint control process further includes smoothing limit command generation. When neither the candidate circulating pump speed command nor the candidate refrigeration unit target command can keep the predicted power module temperature within the temperature safety boundary, the industrial controller generates a charging power smoothing limit command based on the portion of the future heat load sequence that exceeds the heat dissipation capacity. The charging power smoothing limit command includes the limit start control cycle, the allowable power value for each control cycle, and the liquid-cooled thermal inertia state queue recovery conditions required to release the limit.

9. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 7, characterized in that, The transmission delay calibration further includes inertial state validity screening. Before updating the queue length, queue advance step size and state record weight, the industrial controller removes state records where the circulating pump speed changes abruptly and the coolant outlet temperature does not change accordingly, removes state records where the charging request current drops abruptly and the actual output power has entered the limited state, and generates validity markers for the remaining state records according to the degree of correspondence between the power module temperature rise slope and the temperature difference between the coolant inlet and outlet, and updates the liquid-cooled thermal inertial state queue according to the validity markers.

10. The industrial control system for liquid-cooled circulating heat dissipation of a high-power charging pile according to claim 8, characterized in that, The liquid-cooled thermal inertia state queue recovery conditions required to lift the restrictions include temperature rise recovery conditions, temperature difference recovery conditions, and control quantity stability conditions. When the temperature rise slope of the power module enters the preset recovery range, the temperature difference between the inlet and outlet of the coolant enters the preset heat exchange range, and the circulating pump speed command and the refrigeration unit target command remain in the same direction and converge in the continuous control cycle, the industrial controller recalculates the allowable power value for each control cycle according to the future heat load sequence, and writes the recalculated allowable power value into the charging power smoothing limit command.

Images