A low-temperature heat source refrigeration system and method using waste heat of a vehicle radiator

Abstract

This invention relates to the field of vehicle waste heat utilization technology, and discloses a low-temperature heat source refrigeration system and method utilizing waste heat from a vehicle's radiator. The system includes: a parameter acquisition module for real-time acquisition of engine operating parameters; an engine thermal inertial response model to characterize the coolant temperature hysteresis response characteristics; a refrigeration cycle thermal inertial response model to characterize the dynamic response characteristics of the refrigeration cycle thermal state; a model coupling module for iteratively predicting the return coolant temperature and engine coolant temperature step-by-step within a preset time domain, identifying the feedback inflection point step corresponding to the maximum temperature difference; a constraint screening module for tightening heat dissipation safety constraints and screening effective time steps after the inflection point step; a strategy optimization module for generating candidate waste heat extraction strategies and screening the optimal strategy based on the cumulative cooling capacity value of the effective time steps; and a control command generation module for controlling waste heat extraction according to the optimal strategy.

Description

Technical Field

[0001] This invention relates to the field of vehicle waste heat utilization technology, and in particular to a low-temperature heat source refrigeration system and method that utilizes waste heat from a vehicle's radiator. Background Technology

[0002] With the automotive industry's increasing emphasis on energy conservation and emission reduction, utilizing engine waste heat for cooling has become a crucial research direction in vehicle thermal management. During operation, vehicle engines consume fossil fuels and emit carbon dioxide, while simultaneously generating significant waste heat that is discharged through the radiator. Effectively recovering and utilizing this low-grade heat to drive the cooling cycle can significantly reduce the power consumption of the air conditioning system on the engine, thereby reducing fossil fuel consumption and carbon dioxide emissions, and realizing the resource utilization of waste heat in the carbon utilization field. This aligns perfectly with the technical goals of carbon dioxide capture, utilization, and storage (CCUS), both aiming to reduce carbon emissions and improve energy efficiency. Existing technologies have proposed various vehicle waste heat cooling solutions, typically employing absorption or injection cooling cycles, using engine coolant as the driving heat source to extract waste heat for cooling while meeting the engine's cooling requirements.

[0003] However, existing technologies generally treat the refrigeration cycle as a static energy conversion device, neglecting the thermal inertia of the refrigeration cycle itself and its feedback effect on the engine's thermal balance. In actual operation, the coolant temperature drops significantly after flowing through the refrigeration cycle generator. The low-temperature coolant flows back to the engine, affecting its thermal state and thus altering the subsequently extractable waste heat, forming a closed-loop feedback system. Existing control methods do not model or predict this feedback loop, resulting in waste heat extraction strategies failing to balance engine cooling safety and maximizing cooling capacity under transient conditions, making it difficult to achieve globally optimal waste heat utilization.

[0004] Therefore, this invention proposes a low-temperature heat source refrigeration system and method that utilizes the waste heat from a vehicle's radiator. Summary of the Invention

[0005] This invention provides a low-temperature heat source refrigeration system and method that utilizes the waste heat of a vehicle's radiator. By constructing a dual thermal inertial model of the engine and the refrigeration cycle and performing time-step coupling iteration, the system identifies the feedback inflection point time step that has the strongest impact on the engine's thermal balance feedback and adaptively tightens the safety constraints, thereby maximizing the cumulative value of cooling capacity within a preset time domain while ensuring the engine's heat dissipation safety.

[0006] This invention provides a low-temperature heat source refrigeration system utilizing waste heat from a vehicle's radiator, comprising: The parameter acquisition module is used to collect engine speed, load, engine coolant temperature and coolant flow rate in real time. An engine thermal inertial response model is used to characterize the hysteresis response characteristics of engine coolant temperature under the action of waste heat extraction. A thermal inertial response model for a refrigeration cycle is used to characterize the dynamic response characteristics of the thermal state of a refrigeration cycle under the influence of coolant extraction rate. The model coupling module performs the following operations step-by-step within a preset time domain: It acquires the coolant extraction amount for the current time step, determined by the coolant flow rate and the diversion ratio corresponding to the candidate waste heat extraction strategy; it inputs the coolant extraction amount for the current time step into the refrigeration cycle thermal inertial response model to predict the return coolant temperature for the current time step; it inputs the engine speed, load, and return coolant temperature into the engine thermal inertial response model to predict the engine coolant temperature for the next time step; and during the time step iteration, it identifies the time step corresponding to the maximum temperature difference between the return coolant temperature and the engine coolant temperature, using this as the feedback inflection point time step. The constraint filtering module is used to tighten the engine heat dissipation safety constraint by a preset margin value in the time steps at the feedback inflection point and subsequent time steps based on the coupled simulation results of the model coupling module, and to filter out the effective time steps that simultaneously satisfy the tightened engine heat dissipation safety constraint and the normal operation constraint of the cooling cycle. The strategy selection module is used to generate multiple candidate waste heat extraction strategies. Based on the cumulative cooling capacity of each candidate waste heat extraction strategy at the effective time step, the optimal waste heat extraction strategy is selected from the multiple candidate waste heat extraction strategies. The control command generation module is used to generate control commands based on the optimal waste heat extraction strategy to control the extraction ratio of the waste heat extraction device and the operating parameters of the refrigeration cycle.

[0007] Furthermore, the engine thermal inertial response model is constructed in the following way: On the engine test bench, the engine is fixed at multiple speed-load steady-state operating points. At each steady-state operating point, the engine speed and load are kept constant, and the flow ratio of the flow divider valve is changed from a first steady-state value to a second steady-state value. The complete response curve of the engine coolant temperature from the steady state before the step to the steady state after the step is recorded. For the complete response curve at each steady-state operating point, a first-order inertial plus pure time delay model is used for fitting, and three characteristic parameters are extracted: gain coefficient, time constant, and pure time delay. Using engine speed and load as inputs and gain coefficient, time constant, and pure time delay as outputs, a two-input, three-output radial basis function neural network prediction model is constructed as an engine thermal inertial response model under different operating conditions.

[0008] Furthermore, the thermal inertial response model of the refrigeration cycle is constructed in the following way: On the refrigeration cycle performance test bench, the condensing temperature and evaporation temperature of the refrigeration cycle are fixed, and a step change in hot water temperature and hot water flow rate is applied to the inlet of the generator of the refrigeration cycle. The complete response curve of the refrigeration capacity of the refrigeration cycle from the steady state before the step change to the steady state after the step change is recorded. For each complete response curve, a first-order inertial plus pure time delay model is used for fitting, and the gain coefficient, time constant and pure time delay of the refrigeration cycle are extracted. A thermal inertial response model of the refrigeration cycle is established with the generator inlet hot water temperature and hot water flow rate as inputs and the gain coefficient, time constant and pure time delay of the refrigeration cycle as outputs.

[0009] Furthermore, the tightened engine cooling safety constraint in the constraint filtering module is: the original engine cooling safety upper limit is reduced by a preset margin value that is positively correlated with the maximum temperature difference.

[0010] Furthermore, the strategy optimization module includes a quality matching unit, which stores a refrigeration cycle quality-performance mapping model. The refrigeration cycle quality-performance mapping model is a multi-dimensional lookup table. The input dimensions of the multi-dimensional lookup table include the generator inlet hot water temperature, the hot water flow rate converted from the coolant flow rate and the split ratio, the refrigeration cycle pressure ratio, and the evaporation temperature. The output dimensions are the cooling capacity and the cooling energy efficiency ratio. The quality matching unit is used to match the thermodynamic parameters corresponding to the effective time step with the refrigeration cycle quality-performance mapping model to obtain the cooling capacity of each effective time step. The cooling capacity of each effective time step is multiplied by the strategy forward value decay factor corresponding to the effective time step and then accumulated to obtain the cumulative value of the cooling capacity. The strategy forward value decay factor decreases as the time interval between the effective time step and the current time increases.

[0011] Furthermore, the refrigeration cycle grade-performance mapping model is constructed in the following way: On the refrigeration cycle performance test bench, the range of generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio and evaporation temperature were traversed, and the cooling capacity and refrigeration efficiency ratio were recorded for each combination of generator inlet hot water temperature-hot water flow rate-refrigeration cycle pressure ratio-evaporation temperature. Using the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature as input dimensions, and refrigeration capacity and refrigeration energy efficiency ratio as output dimensions, a multidimensional lookup table is established as a refrigeration cycle quality-performance mapping model.

[0012] Furthermore, it also includes a thermal shock safety correction module, which is used for: Obtain the engine oil temperature and coolant temperature, and calculate the real-time difference between the oil temperature and coolant temperature. When the rate of decrease of coolant temperature per unit time exceeds the preset rate of decrease threshold, and the real-time difference between engine oil temperature and coolant temperature exceeds the preset safety threshold, it is determined that there is a risk of thermal shock. When a risk of thermal shock is determined to exist, the upper limit of engine cooling safety will be reduced by a preset safety margin value; The revised engine cooling safety limit is output to the constraint screening module and the strategy optimization module. The constraint screening module performs effective time step screening, and the strategy optimization module uses it as a constraint condition when selecting the optimal waste heat extraction strategy.

[0013] Furthermore, the refrigeration cycle grade-performance mapping model stored in the grade matching unit has a self-evolutionary update mechanism, which includes: Each refrigeration system terminal will upload data collected during actual operation, including generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, evaporation temperature, and corresponding actual cooling capacity, to the cloud server. The cloud server cleans, denoises, and times-aligns data from multiple refrigeration system terminals; The cloud server uses the processed data to correct the refrigeration cycle grade-performance mapping model stored in the cloud; The cloud server will wirelessly distribute the revised refrigeration cycle quality-performance mapping model to each refrigeration system terminal. Each refrigeration system terminal receives the corrected refrigeration cycle grade-performance mapping model, and the grade matching unit loads the corrected refrigeration cycle grade-performance mapping model to replace the original refrigeration cycle grade-performance mapping model.

[0014] Furthermore, the extraction ratio of the waste heat extraction device generated by the control command generation module is executed by a diversion valve connected to the coolant circulation pipeline of the vehicle's radiator. The diversion valve extracts high-temperature coolant from upstream of the radiator according to the diversion ratio value as the driving heat source for the refrigeration cycle.

[0015] This invention provides a low-temperature heat source cooling method utilizing waste heat from a vehicle radiator, comprising: Real-time data collection of engine speed, load, engine coolant temperature, and coolant flow rate; The engine thermal inertial response model is obtained, which is used to characterize the hysteresis response characteristics of engine coolant temperature under the action of waste heat extraction. Obtain the thermal inertial response model of the refrigeration cycle. The thermal inertial response model of the refrigeration cycle is used to characterize the dynamic response characteristics of the thermal state of the refrigeration cycle under the action of coolant extraction amount. Within a preset time domain, perform the following operations step by step: obtain the coolant extraction amount for the current time step, which is determined by the coolant flow rate and the diversion ratio corresponding to the candidate waste heat extraction strategy; input the coolant extraction amount for the current time step into the refrigeration cycle thermal inertial response model to predict the return coolant temperature for the current time step; input the engine speed, load, and return coolant temperature into the engine thermal inertial response model to predict the engine coolant temperature for the next time step; during the time step iteration, identify the time step corresponding to the maximum temperature difference between the return coolant temperature and the engine coolant temperature, and use it as the feedback inflection point time step; In the feedback inflection point and subsequent time steps, the engine cooling safety constraint is tightened by a preset margin value, and the effective time steps that simultaneously satisfy the tightened engine cooling safety constraint and the normal operation constraint of the cooling cycle are selected. Multiple candidate waste heat extraction strategies are generated. Based on the cumulative cooling capacity of each candidate waste heat extraction strategy at the effective time step, the optimal waste heat extraction strategy is selected from the multiple candidate waste heat extraction strategies. Control commands are generated based on the optimal waste heat extraction strategy to control the extraction ratio of the waste heat extraction device and the operating parameters of the refrigeration cycle.

[0016] The beneficial effects of this invention compared to existing technologies are as follows: This invention solves the fundamental defect in existing vehicle waste heat cooling technologies, which treat the cooling cycle as a static energy conversion device and completely ignore its own thermal inertia and the feedback influence of the return coolant on the engine thermal balance, resulting in the inability to simultaneously achieve optimal heat dissipation safety and global cooling capacity under transient conditions. By constructing an engine thermal inertia response model and a cooling cycle thermal inertia response model, and using a model coupling module to iteratively predict the dynamic coupling relationship between the return coolant temperature and the engine coolant temperature step by step within a preset time domain, this invention identifies the feedback inflection point time step with the strongest feedback influence of the return coolant on the engine thermal balance. Then, after the inflection point, it adaptively tightens the heat dissipation safety constraints to perform bidirectional screening of effective time steps. Combined with a strategy optimization module, it globally optimizes the candidate waste heat extraction strategy, achieving precise coordinated control of the bidirectional thermal state of two independent heat capacities under complex transient conditions. This ensures sufficient engine heat dissipation safety while maximizing the cumulative cooling capacity within the preset time domain. This invention reduces the power consumption of the vehicle's air conditioning system by efficiently utilizing engine waste heat to drive the refrigeration cycle, thereby reducing the consumption of fossil fuels and carbon dioxide emissions, and achieving effective control of carbon emissions.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a diagram illustrating the architecture of a low-temperature heat source refrigeration system utilizing waste heat from a vehicle's radiator, as described in an embodiment of the present invention. Figure 2 This is a time-step iterative logic diagram of the model coupling module in an embodiment of the present invention. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] refer to Figure 1 and Figure 2 This invention provides an embodiment of a low-temperature heat source refrigeration system utilizing waste heat from a vehicle's radiator, comprising: The parameter acquisition module is used to collect engine speed, load, engine coolant temperature and coolant flow rate in real time. An engine thermal inertial response model is used to characterize the hysteresis response characteristics of engine coolant temperature under the action of waste heat extraction. A thermal inertial response model for a refrigeration cycle is used to characterize the dynamic response characteristics of the thermal state of a refrigeration cycle under the influence of coolant extraction rate. The model coupling module performs the following operations step-by-step within a preset time domain: It acquires the coolant extraction amount for the current time step, determined by the coolant flow rate and the diversion ratio corresponding to the candidate waste heat extraction strategy; it inputs the coolant extraction amount for the current time step into the refrigeration cycle thermal inertial response model to predict the return coolant temperature for the current time step; it inputs the engine speed, load, and return coolant temperature into the engine thermal inertial response model to predict the engine coolant temperature for the next time step; and during the time step iteration, it identifies the time step corresponding to the maximum temperature difference between the return coolant temperature and the engine coolant temperature, using this as the feedback inflection point time step. The constraint filtering module is used to tighten the engine heat dissipation safety constraint by a preset margin value in the time steps at the feedback inflection point and subsequent time steps based on the coupled simulation results of the model coupling module, and to filter out the effective time steps that simultaneously satisfy the tightened engine heat dissipation safety constraint and the normal operation constraint of the cooling cycle. The strategy selection module is used to generate multiple candidate waste heat extraction strategies. Based on the cumulative cooling capacity of each candidate waste heat extraction strategy at the effective time step, the optimal waste heat extraction strategy is selected from the multiple candidate waste heat extraction strategies. The control command generation module is used to generate control commands based on the optimal waste heat extraction strategy to control the extraction ratio of the waste heat extraction device and the operating parameters of the refrigeration cycle.

[0022] In this embodiment, real-time acquisition of engine speed, load, engine coolant temperature, and coolant flow rate refers to obtaining the real-time crankshaft speed through a speed sensor installed on the engine, reading the current engine load percentage through the engine electronic control unit, obtaining the real-time engine coolant temperature through a temperature sensor installed at the engine coolant outlet, and obtaining the real-time total coolant flow rate upstream of the radiator through a flow meter installed on the engine coolant circulation main. These four parameters together constitute the real-time operating condition input data required for the system to make subsequent thermal inertia prediction and waste heat extraction decisions.

[0023] In this embodiment, the engine thermal inertial response model is used to characterize the hysteresis response characteristics of the engine coolant temperature under the action of waste heat extraction. The engine thermal inertial response model is constructed as follows: On an engine test bench, the engine is fixed at multiple speed-load steady-state operating points. At each steady-state operating point, keeping the engine speed and load constant, the flow splitting ratio of the diverter valve is changed stepwise from a first steady-state value to a second steady-state value, and the complete response curve of the engine coolant temperature from the steady state before the step to the steady state after the step is recorded. For the response curve at each operating point, a first-order inertial plus pure hysteresis model is used for fitting, and three feature parameters are extracted: gain coefficient, time constant, and pure hysteresis time. With engine speed and load as inputs and gain coefficient, time constant, and pure hysteresis time as outputs, a two-input-three-output radial basis function neural network prediction model is constructed. During system operation, the radial basis function neural network prediction model uses three characteristic parameters corresponding to the real-time collected engine speed and load output. The first-order inertial plus pure hysteresis model uses these three characteristic parameters combined with the input return coolant temperature change to calculate the engine coolant temperature at the next time step, thereby characterizing the hysteresis response characteristics of the engine coolant temperature under the action of waste heat extraction.

[0024] In this embodiment, engine coolant refers to the liquid cooling medium circulating in the vehicle engine cooling system, used to absorb heat generated during engine operation and transfer it to the radiator for heat dissipation. The engine coolant absorbs heat and its temperature rises as it flows through the engine water jacket, and releases heat and its temperature decreases as it flows through the radiator. In this system, a portion of the high-temperature coolant is diverted and extracted before entering the radiator, serving as the driving heat source for the refrigeration cycle. After releasing heat and cooling down, it flows back into the coolant circulation pipeline.

[0025] In this embodiment, the waste heat extraction operation refers to the process of adjusting the diversion ratio of the diversion valve installed on the coolant circulation pipeline of the vehicle's radiator to draw out a portion of the high-temperature coolant that originally flowed to the radiator for heat dissipation from upstream of the radiator and deliver it to the generator of the refrigeration cycle as a driving heat source. The diversion ratio determines the proportion of the extracted high-temperature coolant flow rate to the total coolant flow rate; the larger the diversion ratio, the larger the extracted high-temperature coolant flow rate, and the more heat is used to drive the refrigeration cycle.

[0026] It should be noted that in this specification, the diversion ratio and extraction ratio have the same technical meaning, both referring to the proportion of the high-temperature coolant flow rate extracted and delivered to the refrigeration cycle generator to the total coolant flow rate.

[0027] In this embodiment, the thermal state of the refrigeration cycle refers to the real-time values ​​of thermodynamic state parameters such as the temperature and concentration of the solution within the generator. The thermal state of the refrigeration cycle directly affects its cooling capacity and efficiency. When the solution temperature within the generator reaches or exceeds the minimum driving temperature threshold of the refrigeration cycle, the cycle can operate normally and output cooling capacity; when the solution temperature is below this threshold, the refrigeration cycle cannot effectively cool. The thermal state of the refrigeration cycle changes dynamically with variations in the input generator inlet hot water temperature and hot water flow rate.

[0028] In this embodiment, the hysteresis response characteristic refers to the dynamic characteristic that, after the waste heat extraction action occurs, the engine coolant temperature does not change immediately, but gradually transitions to a new steady-state value with a certain delay time and rate of change. This hysteresis response is mainly caused by the heat capacity effect of metal components such as the engine block and cylinder head, as well as the transmission delay of the coolant in the circulation pipeline. Specifically, the hysteresis response characteristic is manifested as follows: when the split ratio changes abruptly, the engine coolant temperature needs a pure hysteresis time before it begins to respond, and then gradually approaches the new steady-state temperature at a rate determined by the time constant of the first-order inertial element.

[0029] In this embodiment, the coolant extraction amount refers to the flow rate of coolant extracted at each time step within a preset time domain, determined based on the coolant flow rate and the diversion ratio corresponding to the candidate waste heat extraction strategy. It is equal to the coolant flow rate multiplied by the diversion ratio. The coolant extraction amount serves as the input to the refrigeration cycle thermal inertial response model, directly affecting the thermal state changes of the refrigeration cycle and the temperature of the return coolant.

[0030] In this embodiment, a thermal inertial response model of the refrigeration cycle is used to characterize the dynamic response characteristics of the thermal state of the refrigeration cycle under the influence of coolant extraction rate. The thermal inertial response model of the refrigeration cycle is constructed as follows: On a refrigeration cycle performance test bench, the condensing and evaporating temperatures of the refrigeration cycle are fixed. A step change in hot water temperature and flow rate is applied to the generator inlet of the refrigeration cycle, and the complete response curve of the refrigeration capacity transitioning from the steady state before the step change to the steady state after the step change is recorded. For each response curve, a first-order inertial plus pure time delay model is used for fitting, and the gain coefficient, time constant, and pure time delay of the refrigeration cycle are extracted. A model is established with the generator inlet hot water temperature and flow rate as inputs and the gain coefficient, time constant, and pure time delay of the refrigeration cycle as outputs. During system operation, the coolant extraction rate and the corresponding engine coolant temperature at the current time step are used as inputs. The model uses the gain coefficient, time constant, and pure time delay to calculate the return coolant temperature at the current time step, thereby characterizing the dynamic response characteristics of the thermal state of the refrigeration cycle under the influence of coolant extraction rate.

[0031] In this embodiment, the dynamic response characteristic refers to the dynamic characteristic that when the coolant extraction rate or hot water temperature of the input refrigeration cycle changes, the thermal state of the refrigeration cycle and the return coolant temperature do not change immediately, but gradually transition to a new steady-state value with a certain time constant. This dynamic response is mainly caused by the heat capacity effect of the solution in the generator and the heat transfer delay of the heat exchange process. Specifically, the dynamic response characteristic is manifested as follows: when the coolant extraction rate or hot water temperature undergoes a step change, the return coolant temperature needs a pure time lag before it begins to respond, and then gradually approaches the new steady-state temperature at a rate determined by the time constant of the first-order inertial element.

[0032] In this embodiment, the preset time domain refers to a calibrated predictive control time window extending into the future from the current moment. The length of the preset time domain is determined comprehensively based on the thermal inertia time constant of the engine coolant temperature and the thermal inertia time constant of the refrigeration cycle thermal state, ensuring that the preset time domain can cover the complete dynamic process of the engine coolant temperature and the refrigeration cycle thermal state transitioning from the current state to a new steady state. The preset time domain contains multiple equally spaced time steps, and the interval length of the time steps is set according to the dynamic response speed of the system and the limitations of computing resources.

[0033] In this embodiment, the coolant extraction amount for the current time step is obtained. The coolant extraction amount is determined by the coolant flow rate and the diversion ratio value corresponding to the candidate waste heat extraction strategy. This means that at the beginning of the iterative calculation of each time step, the system reads the real-time coolant flow rate value from the parameter acquisition module, reads the diversion ratio value corresponding to the candidate waste heat extraction strategy generated by the strategy optimization module, and multiplies the two to calculate the coolant extraction amount for the current time step.

[0034] In this embodiment, the coolant flow rate refers to the real-time value of the total coolant flow rate upstream of the radiator, collected in real time by a flow meter installed on the engine coolant circulation main line. The coolant flow rate varies with engine speed and the operating status of the coolant pump, and is one of the basic parameters for the system to obtain the coolant extraction amount.

[0035] In this embodiment, the diversion ratio value corresponding to the candidate waste heat extraction strategy refers to the target value of the diversion ratio maintained within a preset time domain for each candidate strategy among the multiple candidate waste heat extraction strategies generated by the strategy optimization module. The diversion ratio value determines the proportion of high-temperature coolant extracted from the total coolant flow rate to drive the refrigeration cycle.

[0036] In this embodiment, the coolant extraction amount of the current time step is input into the refrigeration cycle thermal inertial response model to predict the return coolant temperature of the current time step. This means that in the iterative calculation of each time step, the system uses the coolant extraction amount of the current time step and the corresponding engine coolant temperature as input to the refrigeration cycle thermal inertial response model. The model calculates and outputs the predicted value of the return coolant temperature at the end of the current time step based on the first-order inertial plus pure time delay model.

[0037] In this embodiment, the return coolant temperature refers to the temperature of the high-temperature coolant extracted from upstream of the radiator, after releasing heat through the generator of the refrigeration cycle, and returning to the engine coolant circulation pipe. The return coolant temperature is typically lower than the engine coolant temperature; the temperature difference reflects the amount of heat extracted from the coolant by the refrigeration cycle. Changes in the return coolant temperature are fed back through the engine thermal inertia response model, influencing the predicted engine coolant temperature for the next time step.

[0038] In this embodiment, the engine speed, load, and return coolant temperature are input into the engine thermal inertial response model to predict the engine coolant temperature at the next time step. This means that in each iterative calculation at each time step, the system acquires the real-time engine speed and load, inputs the speed and load into the radial basis function neural network prediction model in the engine thermal inertial response model, and outputs the gain coefficient, time constant, and pure time delay corresponding to the current operating condition. Then, the return coolant temperature predicted at this time step is used as the input change of the first-order inertial plus pure time delay model, combined with the real-time collected engine coolant temperature as the initial state, to calculate and output the predicted engine coolant temperature value for the next time step.

[0039] In this embodiment, engine coolant temperature refers to the temperature of the engine coolant at the engine outlet, which is a core parameter for measuring the engine's thermal load and heat dissipation requirements. The real-time value of the engine coolant temperature is acquired by the parameter acquisition module through a temperature sensor. During the time step iteration process, the engine coolant temperature of subsequent time steps is predicted by the engine thermal inertial response model based on the feedback effects of engine speed, load, and return coolant temperature.

[0040] In this embodiment, during the time-step iteration process, the time step corresponding to the maximum temperature difference between the return coolant temperature and the engine coolant temperature is identified as the feedback inflection point time step. This refers to the time step where, after completing the iterative calculation of all time steps within a preset time domain, the system compares the temperature difference between the return coolant temperature and the engine coolant temperature in each time step, finds the time step where the temperature difference reaches its maximum value, and marks this time step as the feedback inflection point time step. The feedback inflection point time step marks the moment when the feedback influence of the return coolant on the engine thermal balance reaches its strongest. When the temperature difference between the return coolant and the engine coolant reaches its maximum, it indicates that the cooling disturbance intensity caused by the return low-temperature coolant to the engine thermal capacity system per unit time reaches its peak. After this, the disturbance begins to decay. Therefore, at this time step and thereafter, the engine heat dissipation safety constraints are tightened to prevent excessive engine cooling.

[0041] In this embodiment, based on the coupled simulation results of the model coupling module, the engine cooling safety constraint is tightened by a preset margin value in the time steps from the feedback inflection point to the subsequent time steps. Valid time steps that simultaneously satisfy both the tightened engine cooling safety constraint and the normal operation constraint of the cooling cycle are selected. This means that after identifying the feedback inflection point, the system applies more stringent safety constraints to all time steps from the feedback inflection point to the subsequent time steps, i.e., the original engine cooling safety upper limit is reduced by a preset margin value to obtain the tightened safety upper limit. Then, step-by-step, it is determined whether the predicted engine coolant temperature does not exceed the tightened safety upper limit and whether the thermal state of the cooling cycle meets the normal operation constraint of the cooling cycle. Time steps that simultaneously satisfy both conditions are determined as valid time steps.

[0042] In this embodiment, the preset margin value refers to the downward adjustment value of the engine heat dissipation safety upper limit preset by the system, which is used to form a more stringent safety constraint after the feedback inflection point step. The preset margin value is positively correlated with the maximum temperature difference, that is, the larger the maximum temperature difference, the larger the preset margin value, and the tighter the safety constraint.

[0043] In this embodiment, the engine cooling safety constraint refers to the upper limit of the engine coolant temperature set to ensure normal engine operation and prevent engine overheating. The engine coolant temperature must not exceed this upper limit, otherwise it may cause engine damage. This upper limit is determined by the engine manufacturer based on the heat resistance performance of the engine cylinder head and cylinder block.

[0044] In this embodiment, the tightened engine cooling safety constraint refers to a more stringent upper limit for coolant temperature obtained by lowering the original engine cooling safety upper limit by a preset margin value. Using the tightened safety constraint at the feedback inflection point and subsequent time steps is to prevent the engine coolant temperature from dropping excessively under the continuous feedback influence of the returning coolant, ensuring stable engine operation within a safe temperature range.

[0045] In this embodiment, the normal operation constraint of the refrigeration cycle refers to the constraint conditions set to ensure the safe and efficient operation of the refrigeration cycle. The normal operation constraint of the refrigeration cycle includes: the solution temperature inside the generator is not lower than the minimum driving temperature threshold of the refrigeration cycle, and the rate of change of the solution temperature inside the generator per unit time does not exceed the thermal stress safety threshold of the refrigeration cycle. The minimum driving temperature threshold of the refrigeration cycle refers to the lowest inlet hot water temperature of the generator that allows the refrigeration cycle to generate effective cooling capacity; below this temperature, the refrigeration cycle cannot operate normally. The thermal stress safety threshold of the refrigeration cycle refers to the highest permissible rate of temperature change to prevent thermal stress damage to the generator due to excessively rapid temperature changes.

[0046] In this embodiment, generating multiple candidate waste heat extraction strategies means that the strategy selection module uniformly selects multiple diversion ratio values ​​within the diversion ratio adjustment range of the diversion valve at a preset step size, and sets each selected diversion ratio value to be maintained within a preset time domain, thus forming a candidate waste heat extraction strategy. The number of candidate waste heat extraction strategies depends on the diversion ratio adjustment range and the size of the preset step size.

[0047] In this embodiment, the cumulative cooling capacity of each candidate waste heat extraction strategy for each effective time step refers to the cumulative cooling capacity of the candidate waste heat extraction strategy within a preset time domain. This is achieved by the system matching the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature corresponding to each effective time step with the refrigeration cycle grade-performance mapping model through the grade matching unit. The cooling capacity of each effective time step is then multiplied by the strategy prospective value decay factor corresponding to the effective time step and accumulated to obtain the cumulative cooling capacity of the candidate waste heat extraction strategy within a preset time domain.

[0048] In this embodiment, the optimal waste heat extraction strategy is selected from multiple candidate waste heat extraction strategies based on the cumulative cooling capacity of the effective time step corresponding to each candidate waste heat extraction strategy. This means that the system calculates the cumulative cooling capacity of each candidate waste heat extraction strategy, compares the cumulative cooling capacity of all candidate waste heat extraction strategies, and selects the candidate waste heat extraction strategy with the largest cumulative cooling capacity as the optimal waste heat extraction strategy.

[0049] In this embodiment, the control command generated based on the optimal waste heat extraction strategy controls the extraction ratio of the waste heat extraction device and the operating parameters of the refrigeration cycle. This means that the control command generation module obtains the diversion ratio value corresponding to the optimal waste heat extraction strategy, generates a first control command to control the opening of the diversion valve, and causes the diversion valve to deliver the corresponding flow rate of high-temperature coolant to the generator of the refrigeration cycle according to the diversion ratio value. At the same time, a second control command is generated to control the pressure ratio and evaporation temperature of the refrigeration cycle, so that the operating parameters of the refrigeration cycle are matched with the current hot water temperature and hot water flow rate, so as to achieve efficient operation of the refrigeration cycle.

[0050] In this embodiment, the waste heat extraction device refers to a device used to extract high-temperature coolant from the vehicle's radiator coolant circulation pipeline, including a diversion valve connected to the vehicle's radiator coolant circulation pipeline. The diversion valve extracts high-temperature coolant from upstream of the radiator according to a diversion ratio value as the driving heat source for the refrigeration cycle.

[0051] In this embodiment, the extraction ratio of the waste heat extraction device refers to the diversion ratio value of the diversion valve, that is, the proportion of the high-temperature coolant flow rate extracted and delivered to the refrigeration cycle generator to the total coolant flow rate. The size of the extraction ratio directly determines the amount of heat used to drive the refrigeration cycle; the larger the extraction ratio, the larger the flow rate of high-temperature coolant entering the refrigeration cycle, and the more heat is used to drive the refrigeration cycle.

[0052] In this embodiment, the operating parameters of the refrigeration cycle refer to the controllable parameters that affect the refrigeration effect and energy efficiency ratio of the refrigeration cycle, including the refrigeration cycle pressure ratio and evaporation temperature. The refrigeration cycle pressure ratio is the ratio of the absolute pressure on the high-pressure side to the absolute pressure on the low-pressure side of the refrigeration cycle, and the evaporation temperature is the temperature at which the refrigerant evaporates in the evaporator during the refrigeration cycle. By adjusting the refrigeration cycle pressure ratio and evaporation temperature, the operating conditions of the refrigeration cycle can be matched with the inlet hot water temperature and flow rate of the generator, thereby obtaining the optimal cooling capacity and refrigeration energy efficiency ratio.

[0053] Furthermore, the engine thermal inertial response model is constructed in the following way: On the engine test bench, the engine is fixed at multiple speed-load steady-state operating points. At each steady-state operating point, the engine speed and load are kept constant, and the flow ratio of the flow divider valve is changed from a first steady-state value to a second steady-state value. The complete response curve of the engine coolant temperature from the steady state before the step to the steady state after the step is recorded. For the complete response curve at each steady-state operating point, a first-order inertial plus pure time delay model is used for fitting, and three characteristic parameters are extracted: gain coefficient, time constant, and pure time delay. Using engine speed and load as inputs and gain coefficient, time constant, and pure time delay as outputs, a two-input, three-output radial basis function neural network prediction model is constructed as an engine thermal inertial response model under different operating conditions.

[0054] In this embodiment, fixing the engine at multiple speed-load steady-state operating points on the engine bench means that, in a bench test environment, the engine is sequentially and stably operated at multiple pre-selected combinations of speed and load operating points. These operating points uniformly cover the engine's normal operating speed range and load range. For example, multiple sets of speed values ​​are selected, evenly distributed at preset intervals, from idle speed to rated speed. At the same time, multiple sets of load values ​​are selected, evenly distributed at preset intervals, from low load to high load, under each set of speed values. The speed values ​​and load values ​​are combined one by one to form the speed-load steady-state operating point matrix to be tested, ensuring that the subsequently constructed radial basis function neural network prediction model can cover all operating conditions that the engine may encounter in actual vehicle operation.

[0055] In this embodiment, the first steady-state value refers to the initial steady-state value maintained by the flow splitting ratio of the flow splitting valve before a step test is performed at each speed-load steady-state operating point. The first steady-state value is usually set to zero or a low initial extraction ratio, indicating that the engine is in a baseline state where no waste heat extraction is performed or only a small amount of waste heat extraction is performed.

[0056] In this embodiment, the second steady-state value refers to the target steady-state value reached after the flow split ratio of the flow split valve changes abruptly from the first steady-state value during a step test at each speed-load steady-state operating point. The second steady-state value differs from the first steady-state value and is typically set to an intermediate or larger value within the flow split ratio adjustment range to simulate a sudden increase in waste heat extraction. The specific value of the second steady-state value is selected within the flow split ratio adjustment range according to the test requirements.

[0057] In this embodiment, at each steady-state operating point, while keeping the engine speed and load constant, the flow splitting ratio of the diverter valve is abruptly changed from a first steady-state value to a second steady-state value. The complete response curve of the engine coolant temperature transitioning from the steady-state before the step change to the steady-state after the step change is recorded. This means that during the bench step test, the engine is first stabilized at a selected combination of speed and load until the engine coolant temperature reaches the steady-state temperature corresponding to the first steady-state value. Then, the flow splitting ratio of the diverter valve is instantaneously adjusted from the first steady-state value to the second steady-state value and maintained constant. Simultaneously, the data acquisition system continuously records the complete temperature change process of the engine coolant temperature from the steady-state value before the step change, through the transition process, until it stabilizes again at the new steady-state value. This results in a complete response curve with time as the horizontal axis and engine coolant temperature as the vertical axis. This curve fully reflects the entire dynamic response process of the engine coolant temperature to the step change in the flow splitting ratio under this specific operating condition.

[0058] In this embodiment, a first-order inertial plus pure time delay model is used to fit the complete response curve at each steady-state operating point, extracting three characteristic parameters: gain coefficient, time constant, and pure time delay. This involves systematically identifying each complete response curve. Specifically, the pure time delay is first determined based on the initial delay period of the response curve. The pure time delay represents the time interval between the step change in the flow ratio and the start of the engine coolant temperature response. Then, on the response curve segment after removing the pure time delay, the step response function of the first-order inertial plus pure time delay model is fitted using least squares to identify the gain coefficient and time constant. The gain coefficient represents the ratio of the steady-state change in coolant temperature to the change in flow ratio, reflecting the sensitivity of coolant temperature to the change in flow ratio under this operating condition. The time constant represents the time required for the coolant temperature to reach approximately two-thirds of its steady-state change from the start of the response, reflecting the rate of change of coolant temperature under this operating condition. These three characteristic parameters together quantitatively describe the hysteresis response characteristics of the engine coolant temperature to the waste heat extraction action under this operating condition.

[0059] In this embodiment, a two-input, three-output radial basis function neural network prediction model is constructed using engine speed and load as inputs and gain coefficient, time constant, and pure time delay as outputs. This model serves as the engine's thermal inertial response model under different operating conditions. Specifically, it uses three sets of feature parameters extracted from all speed-load steady-state operating points as training data, engine speed and load as the two-dimensional input vector of the radial basis function neural network, and the corresponding gain coefficient, time constant, and pure time delay as the three-dimensional output vector. By training the radial basis function neural network, the center position, width parameters, and connection weights of each radial basis function in the network are determined. This allows the trained radial basis function neural network to accurately output the corresponding gain coefficient, time constant, and pure time delay through interpolation calculations under any given engine speed and load input. Combining these prediction parameters with a first-order inertial plus pure time delay model constitutes an engine thermal inertial response model that characterizes the hysteresis response characteristics of the engine's waste heat extraction action under any operating condition.

[0060] Furthermore, the thermal inertial response model of the refrigeration cycle is constructed in the following way: On the refrigeration cycle performance test bench, the condensing temperature and evaporation temperature of the refrigeration cycle are fixed, and a step change in hot water temperature and hot water flow rate is applied to the inlet of the generator of the refrigeration cycle. The complete response curve of the refrigeration capacity of the refrigeration cycle from the steady state before the step change to the steady state after the step change is recorded. For each complete response curve, a first-order inertial plus pure time delay model is used for fitting, and the gain coefficient, time constant and pure time delay of the refrigeration cycle are extracted. A thermal inertial response model of the refrigeration cycle is established with the generator inlet hot water temperature and hot water flow rate as inputs and the gain coefficient, time constant and pure time delay of the refrigeration cycle as outputs.

[0061] In this embodiment, the refrigeration cycle performance test bench is a special experimental device for testing the dynamic performance of the refrigeration cycle. It includes a supply system that can control the hot water temperature and flow rate, a cooling water circulation system, and various sensors. It can accurately simulate the hot water conditions at the generator inlet and record the cooling capacity response data of the refrigeration cycle in real time.

[0062] In this embodiment, the condensing temperature and evaporating temperature of the refrigeration cycle refer to the temperatures at which the refrigerant condenses in the condenser and the temperatures at which the refrigerant evaporates in the evaporator, respectively. In the model building experiment, these are fixed to the design operating conditions to eliminate their interference with the test results.

[0063] In this embodiment, the generator of the refrigeration cycle is a heat exchange device in the circulation system that receives heat from the external high-temperature coolant, heats the internal working fluid solution, and separates the refrigerant vapor. Its inlet receives the extracted high-temperature coolant, and its outlet discharges the cooled refrigerated coolant.

[0064] In this embodiment, the step change in hot water temperature and hot water flow rate refers to the excitation method in which the hot water temperature and hot water flow rate input to the refrigeration cycle generator instantly jump from one steady-state value to another steady-state value, which is used to stimulate the dynamic response of the refrigeration cycle.

[0065] In this embodiment, the steady state before the step is the state in which the refrigeration cycle operates under fixed input conditions until all state parameters remain stable before the step test, and the steady state after the step is the stable state that the refrigeration cycle reaches again under new input conditions after the step occurs.

[0066] In this embodiment, on a refrigeration cycle performance test bench, the condensing and evaporating temperatures of the refrigeration cycle are fixed. A step change in the hot water temperature and flow rate is applied to the generator inlet of the refrigeration cycle, and the complete response curve of the refrigeration capacity transitioning from the steady state before the step change to the steady state after the step change is recorded. By changing the step amplitude of the hot water temperature and the step amplitude of the hot water flow rate, multiple sets of complete response curves under different step conditions can be obtained.

[0067] In this embodiment, a first-order inertial plus pure time delay model is used to fit each complete response curve, extracting the gain coefficient, time constant, and pure time delay of the refrigeration cycle. The gain coefficient is the ratio of the steady-state change in cooling capacity to the change in hot water temperature or hot water flow rate; the time constant is the time required for the cooling capacity to reach approximately two-thirds of its steady-state change; and the pure time delay is the time interval between the input step change and the start of the cooling capacity response. Using the generator inlet hot water temperature and hot water flow rate as indices, the three extracted characteristic parameters are organized into a lookup table, thus forming the thermal inertial response model of the refrigeration cycle. During system operation, the corresponding gain coefficient, time constant, and pure time delay are obtained by querying this model based on the current hot water temperature and hot water flow rate, which is used to predict the dynamic change of the return coolant temperature under the influence of coolant extraction.

[0068] Furthermore, the tightened engine cooling safety constraint in the constraint filtering module is: the original engine cooling safety upper limit is reduced by a preset margin value that is positively correlated with the maximum temperature difference.

[0069] In this embodiment, the original engine heat dissipation safety limit refers to the maximum allowable value of engine coolant temperature determined by the engine manufacturer based on the heat resistance performance of the engine cylinder head and cylinder block. It is a safety boundary that the coolant temperature cannot be exceeded during normal engine operation.

[0070] In this embodiment, the maximum temperature difference refers to the value with the largest temperature difference found by the model coupling module after iterative calculation of all time steps within the preset time domain, comparing the temperature difference between the return coolant temperature and the engine coolant temperature in each time step. The time step corresponding to this value is the feedback inflection point time step.

[0071] In this embodiment, the preset margin value positively correlated with the maximum temperature difference refers to the magnitude value preset in the constraint screening module for lowering the original engine heat dissipation safety upper limit. This margin value increases as the maximum temperature difference increases. Specifically, a temperature difference-margin value correspondence table is pre-stored in the constraint screening module. This table defines the margin values ​​corresponding to different maximum temperature difference ranges; the larger the maximum temperature difference, the larger the corresponding margin value. After identifying the maximum temperature difference, the system determines the preset margin value by looking up this correspondence table. The temperature difference-margin value correspondence table can be pre-established based on engine thermal balance bench tests by calibrating the safety margin requirements under different temperature difference levels.

[0072] In this embodiment, lowering the original engine cooling safety limit by a preset margin value positively correlated with the maximum temperature difference means that in the time steps at and after the feedback inflection point, the constraint filtering module subtracts the preset margin value from the original engine cooling safety limit to obtain the tightened engine cooling safety constraint. The larger the maximum temperature difference, the larger the preset margin value, and the stricter the tightened safety constraint, thereby limiting residual heat extraction earlier to prevent excessive cooling of the engine coolant.

[0073] Furthermore, the strategy optimization module includes a quality matching unit, which stores a refrigeration cycle quality-performance mapping model. The refrigeration cycle quality-performance mapping model is a multi-dimensional lookup table. The input dimensions of the multi-dimensional lookup table include the generator inlet hot water temperature, the hot water flow rate converted from the coolant flow rate and the split ratio, the refrigeration cycle pressure ratio, and the evaporation temperature. The output dimensions are the cooling capacity and the cooling energy efficiency ratio. The quality matching unit is used to match the thermodynamic parameters corresponding to the effective time step with the refrigeration cycle quality-performance mapping model to obtain the cooling capacity of each effective time step. The cooling capacity of each effective time step is multiplied by the strategy forward value decay factor corresponding to the effective time step and then accumulated to obtain the cumulative value of the cooling capacity. The strategy forward value decay factor decreases as the time interval between the effective time step and the current time increases.

[0074] In this embodiment, the hot water flow rate obtained by converting the coolant flow rate and the diversion ratio value refers to the hot water flow rate entering the refrigeration cycle generator by multiplying the coolant flow rate collected in real time by the parameter acquisition module by the diversion ratio value corresponding to the candidate waste heat extraction strategy.

[0075] In this embodiment, the refrigeration cycle pressure ratio refers to the ratio of the absolute pressure on the high-pressure side to the absolute pressure on the low-pressure side of the refrigeration cycle, which is an important operating parameter affecting the refrigeration effect and energy efficiency ratio of the refrigeration cycle.

[0076] In this embodiment, the evaporation temperature refers to the temperature at which the refrigerant evaporates in the evaporator during the refrigeration cycle, which directly affects the refrigeration cycle's ability to absorb heat from the object being cooled.

[0077] In this embodiment, cooling capacity refers to the heat absorbed by the cooling cycle from the object being cooled per unit time, and is the core indicator for measuring the cooling output capacity of the cooling cycle.

[0078] In this embodiment, the cooling energy efficiency ratio refers to the ratio of the cooling capacity of the cooling cycle to the driving heat consumed, which is a core indicator for measuring the energy utilization efficiency of the cooling cycle.

[0079] In this embodiment, the thermodynamic parameters corresponding to the effective time step refer to the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature corresponding to the effective time step selected by the constraint screening module. These four parameters together serve as the input conditions for the refrigeration cycle quality-performance mapping model.

[0080] In this embodiment, the thermodynamic parameters corresponding to the effective time step are matched with the refrigeration cycle grade-performance mapping model to obtain the refrigeration capacity of each effective time step. This means that the grade matching unit takes the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio and evaporation temperature corresponding to the effective time step as input, queries the refrigeration cycle grade-performance mapping model, and looks up the table to obtain the corresponding refrigeration capacity output value under the input conditions.

[0081] In this embodiment, the strategy forward-looking value decay factor corresponding to the effective time step decreases as the time interval between the effective time step and the current time increases. This means that the quality matching unit assigns a weighting coefficient to each effective time step, and the value of this weighting coefficient decreases as the effective time step's position in the preset time domain becomes further away from the current time. For example, an exponential decay form can be used, and the decay coefficient is preset according to the system's prediction uncertainty. The more distant the time step is from the current time, the lower the reliability of the cooling capacity prediction value. Therefore, a smaller weight is assigned to the time step, so that the cumulative cooling capacity value better reflects the contribution of recent prediction results.

[0082] In this embodiment, the cumulative cooling capacity is obtained by multiplying the cooling capacity of each effective time step by the strategy prospective value decay factor corresponding to the effective time step and then summing them. This means that the grade matching unit multiplies the cooling capacity obtained by looking up the table for each effective time step by the strategy prospective value decay factor corresponding to that effective time step to obtain the weighted cooling capacity, and then sums the weighted cooling capacities of all effective time steps to obtain the cumulative cooling capacity of the candidate waste heat extraction strategy within the preset time domain.

[0083] Furthermore, the refrigeration cycle grade-performance mapping model is constructed in the following way: On the refrigeration cycle performance test bench, the range of generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio and evaporation temperature were traversed, and the cooling capacity and refrigeration efficiency ratio were recorded for each combination of generator inlet hot water temperature-hot water flow rate-refrigeration cycle pressure ratio-evaporation temperature. Using the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature as input dimensions, and refrigeration capacity and refrigeration energy efficiency ratio as output dimensions, a multidimensional lookup table is established as a refrigeration cycle quality-performance mapping model.

[0084] In this embodiment, the generator inlet hot water temperature range refers to the range of hot water temperatures that the hot water supply system can provide to the generator inlet on the refrigeration cycle performance test bench. The lower limit of the temperature range is lower than the minimum driving temperature threshold of the refrigeration cycle, and the upper limit is higher than the highest outlet temperature of the engine coolant under normal operating conditions, to ensure that the established lookup table can cover all hot water temperature conditions that may occur during actual vehicle operation.

[0085] In this embodiment, the hot water flow range refers to the range of hot water flow entering the generator inlet that the hot water supply system can provide on the refrigeration cycle performance test bench. The lower limit of the flow range corresponds to the product of the minimum diversion ratio of the diversion valve and the coolant flow rate at engine idle speed, and the upper limit corresponds to the product of the maximum diversion ratio of the diversion valve and the coolant flow rate at engine rated speed, to ensure that the established lookup table can cover all hot water flow conditions that may occur during actual vehicle operation.

[0086] In this embodiment, the refrigeration cycle pressure ratio range refers to the range of changes in the ratio of the absolute pressures on the high-pressure side to the low-pressure side during the performance test of the refrigeration cycle. The lower limit of the pressure ratio range corresponds to the lowest pressure ratio at which the refrigeration cycle can generate effective cooling capacity, and the upper limit corresponds to the highest pressure ratio at which the refrigeration cycle can operate safely, ensuring that the established lookup table can cover the entire normal operating pressure ratio range of the refrigeration cycle.

[0087] In this embodiment, the evaporation temperature range refers to the range of refrigerant evaporation temperature variation within the evaporator during performance testing of the refrigeration cycle. The lower limit of the evaporation temperature range is the minimum evaporation temperature designed for the refrigeration cycle, and the upper limit is the highest evaporation temperature required to meet the temperature requirements of the object being cooled, ensuring that the established lookup table can cover all evaporation temperature conditions that may occur during actual vehicle operation.

[0088] In this embodiment, recording the cooling capacity and cooling efficiency ratio (RER) for each generator inlet hot water temperature-hot water flow rate-refrigeration cycle pressure ratio-evaporation temperature combination means that during bench testing, multiple test points are selected at preset intervals within the generator inlet hot water temperature range, hot water flow rate range, refrigeration cycle pressure ratio range, and evaporation temperature range. The values ​​of each dimension's test points are combined to form all the test condition combinations. The refrigeration cycle is stably operated under each set of condition combinations. After the cooling capacity stabilizes, the measured cooling capacity and calculated RER for that set of condition combinations are recorded. The RER is the measured cooling capacity under that set of condition combinations divided by the input heat of the generator inlet hot water.

[0089] In this embodiment, a multi-dimensional lookup table is established as the refrigeration cycle grade-performance mapping model, using the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature as input dimensions, and refrigeration capacity and refrigeration efficiency ratio as output dimensions. This means that the test data recorded under all operating conditions are organized into a multi-dimensional lookup table with four input parameters as indexes and two output parameters as query results. During system operation, the grade matching unit searches for the corresponding refrigeration capacity and refrigeration efficiency ratio in this lookup table based on the current generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature, thereby completing the rapid matching calculation of refrigeration cycle grade and performance.

[0090] Furthermore, it also includes a thermal shock safety correction module, which is used for: Obtain the engine oil temperature and coolant temperature, and calculate the real-time difference between the oil temperature and coolant temperature. When the rate of decrease of coolant temperature per unit time exceeds the preset rate of decrease threshold, and the real-time difference between engine oil temperature and coolant temperature exceeds the preset safety threshold, it is determined that there is a risk of thermal shock. When a risk of thermal shock is determined to exist, the upper limit of engine cooling safety will be reduced by a preset safety margin value; The revised engine cooling safety limit is output to the constraint screening module and the strategy optimization module. The constraint screening module performs effective time step screening, and the strategy optimization module uses it as a constraint condition when selecting the optimal waste heat extraction strategy.

[0091] In this embodiment, the engine oil temperature and coolant temperature are obtained. The real-time difference between the engine oil temperature and coolant temperature refers to the real-time oil temperature value collected by the oil temperature sensor installed at the engine oil pan and the real-time coolant temperature value collected by the coolant temperature sensor installed at the engine coolant outlet, which are read by the thermal shock safety correction module from the parameter acquisition module. Then, the real-time oil-coolant temperature difference is obtained by subtracting the engine oil temperature from the coolant temperature.

[0092] In this embodiment, the preset rate of temperature drop threshold refers to the maximum allowable rate of temperature drop of the coolant per unit time, preset by the system. When the actual rate of temperature drop of the coolant exceeds this threshold, it indicates that the coolant temperature is decreasing rapidly, which may pose a risk of thermal shock. The specific value of the preset rate of temperature drop threshold is determined based on the results of engine bench thermal shock tests, for example, set to a value between three degrees Celsius and five degrees Celsius per minute.

[0093] In this embodiment, the preset safety threshold refers to the maximum permissible temperature difference between the engine oil temperature and the coolant temperature preset by the system. When the oil-coolant temperature difference exceeds this threshold, it indicates that the temperature difference between the internal metal parts of the engine and the coolant is too large, which may cause damage to the parts due to thermal stress.

[0094] In this embodiment, the engine cooling safety upper limit refers to the maximum permissible value of the engine coolant temperature set to ensure normal engine operation and prevent engine overheating. This upper limit is determined by the engine manufacturer based on the heat resistance performance of the engine cylinder head and cylinder block. In the thermal shock safety correction module, this upper limit is the original, uncorrected upper limit.

[0095] In this embodiment, the preset safety margin value refers to the single downward adjustment value of the engine heat dissipation safety upper limit preset by the system. This value is fixed and is independent of the coolant temperature drop rate and the oil-water temperature difference.

[0096] In this embodiment, reducing the engine cooling safety upper limit by a preset safety margin means that when the thermal shock safety correction module determines that there is a risk of thermal shock, it subtracts the preset safety margin from the current engine cooling safety upper limit to obtain a smaller corrected engine cooling safety upper limit. The corrected upper limit is more stringent than the original upper limit, causing the system to tend to choose a more conservative diversion ratio value when selecting effective time steps and optimizing strategies.

[0097] In this embodiment, the corrected engine heat dissipation safety upper limit is output to both the constraint screening module and the strategy optimization module. The constraint screening module uses this corrected upper limit for effective time step selection, and the strategy optimization module uses it as a constraint condition when selecting the optimal waste heat extraction strategy. This means that the thermal shock safety correction module simultaneously sends the corrected engine heat dissipation safety upper limit to both the constraint screening module and the strategy optimization module. The constraint screening module uses this corrected upper limit instead of the original heat dissipation safety upper limit for effective time step selection, and the strategy optimization module uses this corrected upper limit as one of the constraints during global optimization selection.

[0098] Furthermore, the refrigeration cycle grade-performance mapping model stored in the grade matching unit has a self-evolutionary update mechanism, which includes: Each refrigeration system terminal will upload data collected during actual operation, including generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, evaporation temperature, and corresponding actual cooling capacity, to the cloud server. The cloud server cleans, denoises, and times-aligns data from multiple refrigeration system terminals; The cloud server uses the processed data to correct the refrigeration cycle grade-performance mapping model stored in the cloud; The cloud server will wirelessly distribute the revised refrigeration cycle quality-performance mapping model to each refrigeration system terminal. Each refrigeration system terminal receives the corrected refrigeration cycle grade-performance mapping model, and the grade matching unit loads the corrected refrigeration cycle grade-performance mapping model to replace the original refrigeration cycle grade-performance mapping model.

[0099] In this embodiment, each refrigeration system terminal uploads data collected during actual operation, including generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, evaporation temperature, and corresponding actual cooling capacity, to the cloud server. This means that each vehicle equipped with this refrigeration system uploads real-time data on generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, evaporation temperature, and the actual cooling capacity measured by the cooling capacity sensor, to the cloud server via the vehicle's wireless communication module. The uploaded data includes timestamp information to identify the collection time of each set of data.

[0100] In this embodiment, the cloud server cleans, denoises, and times-aligns the data from multiple refrigeration system terminals. This means that after receiving data uploaded from multiple vehicle terminals, the cloud server first removes obviously abnormal values, such as invalid values ​​exceeding the sensor's range or data records with missing timestamps. Then, it smooths random fluctuations caused by sensor noise or communication interference, using a moving average method to remove high-frequency noise components. Finally, the data from different vehicle terminals are unified to the same time base according to their timestamps, enabling data collected by different terminals under similar operating conditions to be compared and aggregated.

[0101] In this embodiment, the cloud server uses processed data to correct the refrigeration cycle quality-performance mapping model stored in the cloud. This means the cloud server uses data that has undergone cleaning, noise reduction, and time alignment as correction samples to update the multidimensional lookup table in the refrigeration cycle quality-performance mapping model stored in the cloud. The correction method is as follows: for each set of uploaded data containing the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature combination, the corresponding grid point is found in the multidimensional lookup table. A weighted moving average is then calculated between the original cooling capacity value of that grid point and the uploaded actual cooling capacity value. The calculated weighted average replaces the original cooling capacity value. The weighting coefficient of the weighted moving average is determined based on the data acquisition time; data acquired more recently is assigned a larger weight.

[0102] In this embodiment, the cloud server wirelessly distributes the corrected refrigeration cycle quality-performance mapping model to each refrigeration system terminal. This means that after the cloud server completes the correction of the multidimensional lookup table, it sends the updated complete multidimensional lookup table to the refrigeration system terminals installed on each vehicle via a mobile communication network in a wireless transmission manner.

[0103] In this embodiment, each refrigeration system terminal receives the corrected refrigeration cycle grade-performance mapping model. The grade matching unit loads the corrected refrigeration cycle grade-performance mapping model, replacing the original refrigeration cycle grade-performance mapping model. This means that after the refrigeration system terminals on each vehicle receive the corrected multidimensional lookup table sent from the cloud server via the vehicle-mounted wireless communication module, the grade matching unit writes the newly received multidimensional lookup table into the storage unit, replacing the previously stored old version of the multidimensional lookup table. Subsequently, when the grade matching unit performs the matching operation between the thermodynamic parameters of the effective time step and the model, it uses the updated multidimensional lookup table to query the cooling capacity and cooling energy efficiency ratio.

[0104] Furthermore, the extraction ratio of the waste heat extraction device generated by the control command generation module is executed by a diversion valve connected to the coolant circulation pipeline of the vehicle's radiator. The diversion valve extracts high-temperature coolant from upstream of the radiator according to the diversion ratio value as the driving heat source for the refrigeration cycle.

[0105] In this embodiment, the extraction ratio of the waste heat extraction device generated by the control command generation module is executed by a diversion valve connected to the coolant circulation pipeline of the vehicle's radiator. The diversion valve extracts high-temperature coolant from upstream of the radiator according to the diversion ratio value as the driving heat source for the refrigeration cycle. This means that the control command generation module converts the diversion ratio value corresponding to the optimal waste heat extraction strategy into an opening control signal for the diversion valve. The diversion valve is installed on the coolant circulation main pipeline upstream of the vehicle's radiator. Based on the received opening control signal, the valve core position is adjusted, dividing the high-temperature coolant flowing there into two paths according to the diversion ratio value: one path is delivered to the generator of the refrigeration cycle as the driving heat source according to the extraction ratio, and the other path continues to flow to the radiator for normal heat dissipation. High-temperature coolant refers to coolant flowing out from the engine water jacket outlet that has not yet been cooled by the radiator, and its temperature is within the normal operating temperature range of the engine.

[0106] This invention provides an embodiment of a low-temperature heat source cooling method utilizing waste heat from a vehicle's radiator, comprising: Real-time data collection of engine speed, load, engine coolant temperature, and coolant flow rate; The engine thermal inertial response model is obtained, which is used to characterize the hysteresis response characteristics of engine coolant temperature under the action of waste heat extraction. Obtain the thermal inertial response model of the refrigeration cycle. The thermal inertial response model of the refrigeration cycle is used to characterize the dynamic response characteristics of the thermal state of the refrigeration cycle under the action of coolant extraction amount. Within a preset time domain, perform the following operations step by step: obtain the coolant extraction amount for the current time step, which is determined by the coolant flow rate and the diversion ratio corresponding to the candidate waste heat extraction strategy; input the coolant extraction amount for the current time step into the refrigeration cycle thermal inertial response model to predict the return coolant temperature for the current time step; input the engine speed, load, and return coolant temperature into the engine thermal inertial response model to predict the engine coolant temperature for the next time step; during the time step iteration, identify the time step corresponding to the maximum temperature difference between the return coolant temperature and the engine coolant temperature, and use it as the feedback inflection point time step; In the feedback inflection point and subsequent time steps, the engine cooling safety constraint is tightened by a preset margin value, and the effective time steps that simultaneously satisfy the tightened engine cooling safety constraint and the normal operation constraint of the cooling cycle are selected. Multiple candidate waste heat extraction strategies are generated. Based on the cumulative cooling capacity of each candidate waste heat extraction strategy at the effective time step, the optimal waste heat extraction strategy is selected from the multiple candidate waste heat extraction strategies. Control commands are generated based on the optimal waste heat extraction strategy to control the extraction ratio of the waste heat extraction device and the operating parameters of the refrigeration cycle.

[0107] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A low-temperature heat source refrigeration system utilizing waste heat from a vehicle radiator, characterized by, include: The parameter acquisition module is used to collect engine speed, load, engine coolant temperature and coolant flow rate in real time. An engine thermal inertial response model is used to characterize the hysteresis response characteristics of engine coolant temperature under the action of waste heat extraction. A thermal inertial response model for a refrigeration cycle is used to characterize the dynamic response characteristics of the thermal state of a refrigeration cycle under the influence of coolant extraction rate. The model coupling module performs the following operations step-by-step within a preset time domain: It acquires the coolant extraction amount for the current time step, determined by the coolant flow rate and the diversion ratio corresponding to the candidate waste heat extraction strategy; it inputs the coolant extraction amount for the current time step into the refrigeration cycle thermal inertial response model to predict the return coolant temperature for the current time step; it inputs the engine speed, load, and return coolant temperature into the engine thermal inertial response model to predict the engine coolant temperature for the next time step; and during the time step iteration, it identifies the time step corresponding to the maximum temperature difference between the return coolant temperature and the engine coolant temperature, using this as the feedback inflection point time step. The constraint filtering module is used to tighten the engine heat dissipation safety constraint by a preset margin value in the time steps at the feedback inflection point and subsequent time steps based on the coupled simulation results of the model coupling module, and to filter out the effective time steps that simultaneously satisfy the tightened engine heat dissipation safety constraint and the normal operation constraint of the cooling cycle. The strategy selection module is used to generate multiple candidate waste heat extraction strategies. Based on the cumulative cooling capacity of each candidate waste heat extraction strategy at the effective time step, the optimal waste heat extraction strategy is selected from the multiple candidate waste heat extraction strategies. The control command generation module is used to generate control commands based on the optimal waste heat extraction strategy to control the extraction ratio of the waste heat extraction device and the operating parameters of the refrigeration cycle.

2. The low-temperature heat source refrigeration system utilizing the waste heat of the radiator of a vehicle according to claim 1, characterized in that, The engine thermal inertial response model is constructed in the following way: On the engine test bench, the engine is fixed at multiple speed-load steady-state operating points. At each steady-state operating point, the engine speed and load are kept constant, and the flow ratio of the flow divider valve is changed from a first steady-state value to a second steady-state value. The complete response curve of the engine coolant temperature from the steady state before the step to the steady state after the step is recorded. For the complete response curve at each steady-state operating point, a first-order inertial plus pure time delay model is used for fitting, and three characteristic parameters are extracted: gain coefficient, time constant, and pure time delay. Using engine speed and load as inputs and gain coefficient, time constant, and pure time delay as outputs, a two-input, three-output radial basis function neural network prediction model is constructed as an engine thermal inertial response model under different operating conditions.

3. The low-temperature heat source refrigeration system utilizing the waste heat of the radiator of a vehicle according to claim 1, characterized in that, The thermal inertial response model of the refrigeration cycle is constructed in the following way: On the refrigeration cycle performance test bench, the condensing temperature and evaporation temperature of the refrigeration cycle are fixed, and a step change in hot water temperature and hot water flow rate is applied to the inlet of the generator of the refrigeration cycle. The complete response curve of the refrigeration capacity of the refrigeration cycle from the steady state before the step change to the steady state after the step change is recorded. For each complete response curve, a first-order inertial plus pure time delay model is used for fitting, and the gain coefficient, time constant and pure time delay of the refrigeration cycle are extracted. A thermal inertial response model of the refrigeration cycle is established with the generator inlet hot water temperature and hot water flow rate as inputs and the gain coefficient, time constant and pure time delay of the refrigeration cycle as outputs.

4. The low temperature heat source refrigeration system utilizing the waste heat of the radiator of a vehicle according to claim 1, characterized in that, The tightened engine cooling safety constraint in the constraint filtering module is: the original engine cooling safety upper limit is reduced by a preset margin value that is positively correlated with the maximum temperature difference.

5. The low temperature heat source refrigeration system utilizing the waste heat of the radiator of a vehicle according to claim 1, characterized in that, The strategy optimization module includes a quality matching unit, which stores a refrigeration cycle quality-performance mapping model. The refrigeration cycle quality-performance mapping model is a multi-dimensional lookup table. The input dimensions of the multi-dimensional lookup table include the generator inlet hot water temperature, the hot water flow rate converted from the coolant flow rate and the split ratio, the refrigeration cycle pressure ratio, and the evaporation temperature. The output dimensions are the cooling capacity and the cooling efficiency ratio. The quality matching unit is used to match the thermodynamic parameters corresponding to the effective time step with the refrigeration cycle quality-performance mapping model to obtain the cooling capacity of each effective time step. The cooling capacity of each effective time step is multiplied by the strategy look-ahead value decay factor corresponding to the effective time step and then accumulated to obtain the cumulative cooling capacity value. The strategy look-ahead value decay factor decreases as the time interval between the effective time step and the current time increases.

6. The low-temperature heat source refrigeration system utilizing waste heat from a vehicle radiator according to claim 5, characterized in that, The refrigeration cycle grade-performance mapping model is constructed in the following way: On the refrigeration cycle performance test bench, the range of generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio and evaporation temperature were traversed, and the cooling capacity and refrigeration efficiency ratio were recorded for each combination of generator inlet hot water temperature-hot water flow rate-refrigeration cycle pressure ratio-evaporation temperature. Using the generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, and evaporation temperature as input dimensions, and refrigeration capacity and refrigeration energy efficiency ratio as output dimensions, a multidimensional lookup table is established as a refrigeration cycle quality-performance mapping model.

7. The low-temperature heat source refrigeration system utilizing waste heat from a vehicle radiator according to claim 1, characterized in that, It also includes a thermal shock safety correction module, which is used for: Obtain the engine oil temperature and coolant temperature, and calculate the real-time difference between the oil temperature and coolant temperature. When the rate of decrease of coolant temperature per unit time exceeds the preset rate of decrease threshold, and the real-time difference between engine oil temperature and coolant temperature exceeds the preset safety threshold, it is determined that there is a risk of thermal shock. When a risk of thermal shock is determined to exist, the upper limit of engine cooling safety will be reduced by a preset safety margin value; The revised engine cooling safety limit is output to the constraint screening module and the strategy optimization module. The constraint screening module performs effective time step screening, and the strategy optimization module uses it as a constraint condition when selecting the optimal waste heat extraction strategy.

8. The low-temperature heat source refrigeration system utilizing waste heat from a vehicle radiator according to claim 5, characterized in that, The refrigeration cycle grade-performance mapping model stored in the grade matching unit has a self-evolutionary update mechanism, which includes: Each refrigeration system terminal will upload data collected during actual operation, including generator inlet hot water temperature, hot water flow rate, refrigeration cycle pressure ratio, evaporation temperature, and corresponding actual cooling capacity, to the cloud server. The cloud server cleans, denoises, and times-aligns data from multiple refrigeration system terminals; The cloud server uses the processed data to correct the refrigeration cycle grade-performance mapping model stored in the cloud; The cloud server will wirelessly distribute the revised refrigeration cycle quality-performance mapping model to each refrigeration system terminal. Each refrigeration system terminal receives the corrected refrigeration cycle grade-performance mapping model, and the grade matching unit loads the corrected refrigeration cycle grade-performance mapping model to replace the original refrigeration cycle grade-performance mapping model.

9. The low-temperature heat source refrigeration system utilizing waste heat from a vehicle radiator according to claim 1, characterized in that, The extraction ratio of the waste heat extraction device generated by the control command generation module is executed by a diversion valve connected to the coolant circulation pipeline of the vehicle's radiator. The diversion valve extracts high-temperature coolant from upstream of the radiator according to the diversion ratio value as the driving heat source for the refrigeration cycle.

10. A low-temperature heat source cooling method utilizing waste heat from a vehicle radiator, characterized in that, include: Real-time data collection of engine speed, load, engine coolant temperature, and coolant flow rate; The engine thermal inertial response model is obtained, which is used to characterize the hysteresis response characteristics of engine coolant temperature under the action of waste heat extraction. Obtain the thermal inertial response model of the refrigeration cycle. The thermal inertial response model of the refrigeration cycle is used to characterize the dynamic response characteristics of the thermal state of the refrigeration cycle under the action of coolant extraction amount. Within a preset time domain, perform the following operations step by step: obtain the coolant extraction amount for the current time step, which is determined by the coolant flow rate and the diversion ratio corresponding to the candidate waste heat extraction strategy; input the coolant extraction amount for the current time step into the refrigeration cycle thermal inertial response model to predict the return coolant temperature for the current time step; input the engine speed, load, and return coolant temperature into the engine thermal inertial response model to predict the engine coolant temperature for the next time step; during the time step iteration, identify the time step corresponding to the maximum temperature difference between the return coolant temperature and the engine coolant temperature, and use it as the feedback inflection point time step; In the feedback inflection point and subsequent time steps, the engine cooling safety constraint is tightened by a preset margin value, and the effective time steps that simultaneously satisfy the tightened engine cooling safety constraint and the normal operation constraint of the cooling cycle are selected. Multiple candidate waste heat extraction strategies are generated. Based on the cumulative cooling capacity of each candidate waste heat extraction strategy at the effective time step, the optimal waste heat extraction strategy is selected from the multiple candidate waste heat extraction strategies. Control commands are generated based on the optimal waste heat extraction strategy to control the extraction ratio of the waste heat extraction device and the operating parameters of the refrigeration cycle.

Images