Self-adaptive temperature difference adjusting method for liquid cooling converter
The temperature feedback control parameters are calculated through the particle swarm optimization algorithm, and the proportional valve opening of the liquid-cooled converter is adjusted, which solves the problem of temperature difference between the liquid-cooled converter module, reduces the simulation and actual measurement workload, and improves the temperature difference adjustment accuracy and efficiency.
Patent Information
- Application Number
- CN202411172034.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-26
- Publication Date
- 2024-11-29
AI Technical Summary
In the high power density, the existing liquid-cooled converters have different consistency of energy storage converters due to the local temperature difference between the IGBT and the inductor, and the life span is uneven and the power derating curve is inconsistent. The traditional method needs to determine the temperature difference through thermal simulation and testing. The relationship curve with traffic is heavy and complex.
The particle swarm optimization algorithm is used to calculate the optimal control parameters of temperature feedback control, adjust the proportional valve opening according to the real-time temperature and target temperature value of IGBT, adjust the flow rate of the liquid-cooled pipeline through the PWM duty cycle, reduce the temperature difference, and pass the fitting curve function Accelerate the iteration process.
There is no need to know the relationship curve of the liquid-cooled pipeline flow rate and IGBT temperature difference in advance, which reduces the simulation and actual measurement workload, improves the accuracy and efficiency of temperature difference adjustment, and achieves higher implementation and accuracy.
Smart Images

Figure CN119045567A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of converters, and in particular to a method for self-adapting temperature difference regulation of a liquid-cooled converter. Background Art
[0002] As the power of PCS (Power Conversion System) energy storage converters increases, the size of the cabinet is limited, and air cooling can no longer meet the high power density requirements. More and more energy storage converters use liquid cooling with high heat dissipation efficiency and good heat dissipation effect. Liquid-cooled energy storage converters use an integrated solution of liquid cooling plate, channel radiator, liquid cooling pipe, and water pump. The IGBT (Insulated Gate Bipolar Transistor) and inductor of each power module of the liquid-cooled energy storage converter are coated with thermal conductive glue and attached to the liquid cooling plate to cool the IGBT and inductor. However, due to production process and other issues, there are local temperature differences between different IGBTs or inductors. When the temperature difference is greater than a certain threshold, the consistency of the energy storage converter module deteriorates, which will lead to large differences in the life of the energy storage converter and inconsistent power derating curves.
[0003] In order to solve the temperature difference problem between different energy storage converter modules, the common solution is to add a proportional valve to the liquid cooling inlet of each module, and adaptively adjust the opening of the proportional valve according to the temperature difference of different modules, thereby changing the coolant flow in the liquid cooling pipeline of each module to reduce the temperature difference of each power module. Figure 1 As shown, the specific temperature difference adjustment process is as follows:
[0004] (1) The PCS main control board collects the shell temperature Tij of each IGBT in different modules (i represents the module number, j represents the number of IGBTs in module i), calculates the maximum value Tmax, minimum value Tmin, and average value Tav of all IGBT shell temperatures in all modules, and starts the flow regulation function if the maximum value Tmax-minimum value Tmin>preset threshold T1.
[0005] (2) Select the PCS module with a large temperature difference from the average value Tav. The selection method is: select the PCS module with |Tij-Tav|>preset threshold T2.
[0006] (3) By combining thermal simulation and testing, the relationship curve between the IGBT temperature difference ΔT and the flow change ΔL is determined.
[0007] (4) According to the temperature and flow relationship curve, for modules with IGBT temperature higher than the average value Tav, increase the flow, and for modules with IGBT temperature lower than the average value Tav, reduce the flow. The specific method of flow regulation is: according to the proportional valve flow and drive current curve, determine the proportional valve drive current, and according to the relationship between the drive current and the PWM duty cycle, determine the PWM wave requirements of the PCS main control board.
[0008] In the above method, thermal simulation needs to be combined with actual testing to determine the relationship curve between the temperature difference ΔT of the IGBT and the flow change ΔL. Given that the IGBT temperature and flow relationship curves of the PCS module at different operating powers and ambient temperatures are significantly different, a large number of simulations and tests are required for various power and ambient temperature combinations to determine the relationship curve between the IGBT temperature difference and flow. This workload is heavy and complex, which brings certain challenges to the implementation of the solution. Summary of the invention
[0009] The technical problem to be solved by the present invention is: in view of the above-mentioned defects of the prior art, a method for adaptive temperature difference adjustment of a liquid-cooled converter is provided.
[0010] To achieve the above object, the present invention provides a method for adaptively adjusting temperature difference of a liquid-cooled converter, the method comprising the following steps:
[0011] Step S1, presetting different IGBT target temperature values Tov according to different power values of the converter module;
[0012] Step S2, collecting the IGBT real-time temperature value T, and calculating the temperature deviation between the IGBT real-time temperature value T and the IGBT target temperature value Tov according to the power value of the converter module and the IGBT target temperature value Tov;
[0013] Step S3, calculating the optimal control parameters of the temperature feedback control algorithm by using a particle swarm optimization algorithm according to the temperature deviation;
[0014] Step S4, outputting a PWM optimal duty cycle according to the temperature deviation and the optimal control parameter, and driving a proportional valve corresponding to the converter module to adjust the opening;
[0015] In step S3, during the iteration process of the particle swarm optimization algorithm, a curve function of the liquid cooling pipe flow rate L and the IGBT real-time temperature T is fitted according to the values of the liquid cooling pipe flow rate L and the IGBT real-time temperature T, and the liquid cooling pipe flow rate L corresponding to the IGBT target temperature value Tov is obtained according to the curve function, and the PWM optimal duty cycle D is inferred, and the inertia weight of the particle swarm optimization algorithm is adjusted according to the PWM optimal duty cycle D.
[0016] Preferably, the temperature feedback control algorithm is a PI control algorithm, and the control parameters are a proportional control parameter kp and an integral control parameter ki.
[0017] Preferably, the temperature feedback control algorithm is a PID control algorithm, and the control parameters are a proportional control parameter kp, an integral control parameter ki, and a differential control parameter kd.
[0018] Preferably, in step S3, the specific method of calculating the optimal control parameters of the temperature feedback control algorithm by using the particle swarm optimization algorithm is as follows:
[0019] Step S31, setting the particle position vector, the particle position vector is composed of a proportional adjustment parameter kp and an integral adjustment parameter ki; determining the value range of the proportional adjustment parameter kp and the integral adjustment parameter ki, and randomly generating a matrix as a particle swarm position matrix according to the value range; setting the particle growth rate, learning factor, inertia weight, and maximum number of iterations; initializing the individual historical best position pbest and the global historical best position gbest of each particle;
[0020] Step S32, assign the particles in the particle swarm position matrix to the proportional adjustment parameter kp and the integral adjustment parameter ki in the PI controller in turn, obtain the corresponding PWM optimal duty cycle D through PI adjustment, drive the proportional valve to adjust the opening, collect the IGBT real-time temperature value T as the initial fitness value of each particle; compare the individual historical best position pbest and the global historical best position gbest with the individual and global current fitness values respectively, and update the individual historical position pbest and the global historical best position gbest according to the optimal values;
[0021] Step S33, update the position and speed of the particle to obtain a new position value of the particle, assign the new position value of the particle to the proportional adjustment parameter kp and the integral adjustment parameter ki in the PI controller, obtain the corresponding PWM optimal duty cycle D through PI adjustment, drive the proportional valve to adjust the opening, collect the IGBT real-time temperature value T as the new fitness value of each particle; compare it with the individual historical best position pbest and the global historical best position gbest, and update the individual historical position pbest and the global historical best position gbest according to the optimal value; obtain the liquid cooling pipeline flow L according to the PWM optimal duty cycle D, record the liquid cooling pipeline flow L and the IGBT real-time temperature T in each iteration, fit the curve function of the liquid cooling pipeline flow L and the IGBT real-time temperature T according to the recorded value, and after obtaining a stable curve function, obtain the liquid cooling pipeline flow L corresponding to the IGBT target temperature value Tov according to the curve function and the IGBT target temperature value Tov, and then inversely deduce the PWM optimal duty cycle D corresponding to the liquid cooling pipeline flow L, and adjust the inertia weight according to the PWM optimal duty cycle D;
[0022] Step S34, determine whether the termination condition is met, if so, jump to step S35, otherwise jump to step S33;
[0023] Step S35, outputting the global best historical position gbest, and adjusting the opening of the proportional valve according to the global best historical position gbest.
[0024] Preferably, in step S33, adjusting the inertia weight means adjusting the inertia weight value to:
[0025]
[0026] Where: k is the inertia weight of the kth iteration, D k is the optimal PWM duty cycle of the PI regulation output in the kth iteration, D k ' is the optimal PWM duty cycle of the kth iteration obtained by inverse deduction based on the curve function, n is the number of particles, and α is a constant parameter.
[0027] Preferably, in step S33, the method for obtaining the liquid cooling pipe flow L according to the PWM optimal duty cycle D is: obtaining the driving current I according to the PWM optimal duty cycle D and the characteristics of the proportional valve driving circuit, and obtaining the liquid cooling pipe flow L according to the driving current I and the relationship between the driving current and the flow of the proportional valve; the method for inverting the PWM optimal duty cycle D corresponding to the liquid cooling pipe flow L is: obtaining the driving current I according to the relationship between the liquid cooling pipe flow L, the driving current and the flow of the proportional valve, and obtaining the PWM optimal duty cycle D according to the driving current I and the characteristics of the proportional valve driving circuit.
[0028] Preferably, in step S33, the criterion for obtaining a stable curve function is: the number of iterations of the particle swarm optimization algorithm reaches an iteration number threshold.
[0029] Preferably, in step S33, the criterion for obtaining a stable curve function is: using the curve function fitted in the previous iteration to predict the data of the current iteration, when the difference between the predicted value and the actual value is less than a preset residual threshold, it is considered that the fitted curve function is stable.
[0030] Preferably, in step S33, the specific method of fitting the curve function of the liquid cooling pipe flow L and the IGBT real-time temperature T according to the recorded values is: using a multi-order polynomial and a least squares method for fitting.
[0031] Preferably, in step S34, the termination condition is: the number of iterations exceeds the maximum number of iterations or the difference between the IGBT real-time temperature T and the IGBT target temperature Tov is less than a preset IGBT temperature difference threshold.
[0032] The present invention has the following beneficial effects: this solution does not need to know the relationship curve between the liquid cooling pipeline flow and the IGBT temperature difference in advance, but sets the corresponding IGBT target temperature value according to the different power values of the PCS module, uses the particle swarm optimization algorithm to find the optimal temperature control parameter value, outputs the PWM optimal duty cycle to drive the proportional valve to adjust the opening, changes the liquid cooling pipeline flow, and reduces the temperature difference. In the iterative process of the particle swarm optimization algorithm, the liquid cooling pipeline flow is obtained according to the PWM optimal duty cycle output by the temperature adjustment module, and the liquid cooling pipeline flow and the IGBT real-time temperature are curve fitted. The corresponding liquid cooling pipeline flow is deduced according to the fitted curve function and the set IGBT target temperature value, and the PWM optimal duty cycle is obtained. The inertia weight is adjusted according to the PWM optimal duty cycle to speed up the iteration speed and accuracy of the particle swarm optimization algorithm. This solution reduces a lot of simulation and actual measurement workload, has strong feasibility, and has higher temperature difference adjustment accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0034] Figure 1 This is a flow chart of the traditional converter module temperature difference adjustment.
[0035] Figure 2 A schematic diagram of a temperature difference control loop provided in an embodiment of the present invention.
[0036] Figure 3 A schematic diagram of the steps of a method for adaptively adjusting temperature difference of a liquid-cooled converter provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0037] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0038] The embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings of the specification. It should be understood that the embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.
[0039] The present invention is suitable for regulating the temperature difference of each module of a liquid-cooled converter. There are multiple modules in the liquid-cooled converter, and the liquid inlet of each module is connected to the liquid outlet of a liquid-cooled host through a three-stage pipeline, and the liquid outlet of each module is connected to the liquid inlet of the liquid-cooled host through a three-stage pipeline. A proportional valve is provided in the three-stage pipeline of the liquid inlet of each module, and a proportional valve driving circuit is provided on the PCS main control board. The PCS main control board outputs a PWM wave to the proportional valve driving circuit. By controlling the duty cycle of the PWM wave, the driving current output by the driving circuit can be changed, thereby adjusting the opening of the proportional valve and the pipeline flow of the liquid cooling pipeline, and adjusting the temperature of the IGBT. A temperature sensor is provided on the IGBT for transmitting the real-time temperature of the IGBT to the PCS main control board.
[0040] like Figure 2-Figure 3 As shown, an embodiment of the present invention provides a method for adaptively adjusting temperature difference of a liquid-cooled converter, the method comprising the following steps:
[0041] Step S1, presetting different IGBT target temperature values Tov according to different power values of the converter module;
[0042] Step S2, collecting the IGBT real-time temperature value T, and calculating the temperature deviation between the IGBT real-time temperature value T and the IGBT target temperature value Tov according to the power value of the converter module and the IGBT target temperature value Tov;
[0043] Step S3, calculating the optimal control parameters of the temperature feedback control algorithm by using a particle swarm optimization algorithm according to the temperature deviation;
[0044] Step S4: outputting a PWM optimal duty cycle according to the temperature deviation and the optimal control parameter, and driving a proportional valve corresponding to the converter module to adjust the opening.
[0045] In step S3, during the iteration process of the particle swarm optimization algorithm, a curve function of the liquid cooling pipe flow rate L and the IGBT real-time temperature T is fitted according to the values of the liquid cooling pipe flow rate L and the IGBT real-time temperature T, and the liquid cooling pipe flow rate L corresponding to the IGBT target temperature value Tov is obtained according to the curve function, and the PWM optimal duty cycle D is inferred, and the inertia weight of the particle swarm optimization algorithm is adjusted according to the PWM optimal duty cycle D.
[0046] In some embodiments of the present invention, the temperature feedback control algorithm is a PI adjustment algorithm, and the control parameters are a proportional adjustment parameter kp and an integral adjustment parameter ki.
[0047] In some other embodiments of the present invention, the temperature feedback control algorithm is a PID control algorithm, and the control parameters are a proportional control parameter kp, an integral control parameter ki, and a differential control parameter kd.
[0048] PID regulation adds a differential term on the basis of PI regulation, which can respond to system changes more quickly and improve the dynamic performance of the system, but its control algorithm is more complex. In practical applications, it is necessary to select the corresponding temperature regulation algorithm according to the actual situation. The particle swarm optimization algorithm based on PID regulation and the particle swarm optimization algorithm based on PI regulation have similar implementation principles. The following only takes PI regulation as an example for explanation.
[0049] In some embodiments of the present invention, the temperature feedback control algorithm is a PI adjustment algorithm. In step S3, the specific method for calculating the optimal control parameters of the temperature feedback control algorithm by using a particle swarm optimization algorithm is as follows:
[0050] Step S31, setting the particle position vector, the particle position vector is composed of a proportional adjustment parameter kp and an integral adjustment parameter ki; determining the value range of the proportional adjustment parameter kp and the integral adjustment parameter ki, and randomly generating a matrix as the particle swarm position matrix according to the value range; setting the particle growth rate, learning factor, inertia weight, and maximum number of iterations; initializing the individual historical best position pbest and the global historical best position gbest of each particle.
[0051] In the embodiment of the present invention, the number of particles is set according to specific circumstances.
[0052]
[0053] Particle speed growth formula:
[0054]
[0055] in: represents the kth iteration speed, Represents the k+1th iteration speed. ω is the inertia weight, which represents the influence of individual past performance on the present, and is generally taken as 0.5~1. When the value of the inertia weight coefficient is large, the global search ability of the particle is stronger, and when it is small, the local search ability is stronger. In the embodiment of the present invention, the setting value of the inertia weight will be adjusted later. ω can be set to a fixed value or change according to a linear decreasing rule in the early stage. λ1 is the individual learning factor and λ2 is the group factor, which are generally taken as 0-2. R1 and R2 are random numbers between 0 and 1. k is the number of iterations, which is generally taken as 100~2000. If the number of iterations is too small, the solution is unstable, and if it is too large, it wastes time.
[0056] The speed limit formula is as follows:
[0057]
[0058] The position update formula is as follows:
[0059]
[0060] Step S32, assign the particles in the particle swarm position matrix to the proportional adjustment parameter kp and the integral adjustment parameter ki in the PI controller in turn, obtain the corresponding PWM optimal duty cycle D through PI adjustment, drive the proportional valve to adjust the opening, and collect the IGBT real-time temperature value T as the initial fitness value of each particle; compare the individual historical best position pbest and the global historical best position gbest with the individual and global current fitness values respectively, and update the individual historical position pbest and the global historical best position gbest according to the optimal values.
[0061] Step S33, update the position and speed of the particle to obtain a new position value of the particle, assign the new position value of the particle to the proportional adjustment parameter kp and the integral adjustment parameter ki in the PI controller, obtain the corresponding PWM optimal duty cycle D through PI adjustment, drive the proportional valve to adjust the opening, collect the IGBT real-time temperature value T as the new fitness value of each particle; compare it with the individual historical best position pbest and the global historical best position gbest, and update the individual historical position pbest and the global historical best position gbest according to the optimal value; obtain the liquid cooling pipeline flow L according to the PWM optimal duty cycle D, record the liquid cooling pipeline flow L and the IGBT real-time temperature T in each iteration, fit the curve function of the liquid cooling pipeline flow L and the IGBT real-time temperature T according to the recorded values, and after obtaining a stable curve function, obtain the liquid cooling pipeline flow L corresponding to the IGBT target temperature value Tov according to the curve function and the IGBT target temperature value Tov, and then inversely deduce the PWM optimal duty cycle D corresponding to the liquid cooling pipeline flow L, and adjust the inertia weight according to the PWM optimal duty cycle D.
[0062] Step S34, determine whether the termination condition is met, if so, jump to step S35, otherwise jump to step S33. In the embodiment of the present invention, the termination condition is: the number of iterations exceeds the maximum number of iterations or the difference between the IGBT real-time temperature T and the IGBT target temperature Tov is less than a preset IGBT temperature difference threshold. The IGBT temperature difference threshold is set according to actual conditions.
[0063] Step S35, outputting the global best historical position gbest, and adjusting the opening of the proportional valve according to the global best historical position gbest.
[0064] In the embodiment of the present invention, in the step S33, adjusting the inertia weight means adjusting the inertia weight value to:
[0065]
[0066] Where: kis the inertia weight of the kth iteration, D k is the optimal PWM duty cycle of the PI regulation output in the kth iteration, D k ' is the optimal PWM duty cycle of the kth iteration obtained by inverse deduction based on the curve function, n is the number of particles, and α is a constant parameter.
[0067] The particle swarm optimization algorithm is an existing technology, and its principle will not be described here. This solution derives the liquid cooling pipe flow L corresponding to the IGBT target temperature Tov through the fitted function, and then reversely derives the PWM optimal duty cycle of the PI regulation output, and adjusts the inertia weight according to the PWM optimal duty cycle, thereby adjusting the speed and position of the next iteration and speeding up the iteration speed.
[0068] In an embodiment of the present invention, in step S33, the method for obtaining the liquid cooling pipe flow L according to the PWM optimal duty cycle D is: obtaining the driving current I according to the PWM optimal duty cycle D and the characteristics of the proportional valve driving circuit, and obtaining the liquid cooling pipe flow L according to the driving current I and the relationship between the driving current and the flow of the proportional valve; the method for inverting the PWM optimal duty cycle D corresponding to the liquid cooling pipe flow L is: obtaining the driving current I according to the relationship between the liquid cooling pipe flow L, the driving current and the flow of the proportional valve, and obtaining the PWM optimal duty cycle D according to the driving current I and the characteristics of the proportional valve driving circuit.
[0069] In some embodiments of the present invention, in step S33, the criterion for obtaining a stable curve function is: the number of iterations of the particle swarm optimization algorithm reaches an iteration number threshold. The iteration number threshold is set according to actual conditions.
[0070] In some other embodiments of the present invention, in step S33, the criterion for obtaining a stable curve function is: using the curve function fitted in the previous iteration to predict the data of the current iteration, when the difference between the predicted value and the actual value is less than a preset residual threshold, it is considered that the fitted curve function is stable. The residual threshold is set according to actual conditions. In practical applications, whether the curve is stable can also be judged based on the standard deviation, mean square error, root mean square error, etc. of the residual.
[0071] In some embodiments of the present invention, in step S33, the specific method of fitting the curve function of the liquid cooling pipe flow L and the IGBT real-time temperature T according to the recorded values is: using a multi-order polynomial and a least squares method for fitting. Specifically, assuming that the relationship model between the IGBT real-time temperature T and the liquid cooling pipe L is as follows:
[0072] T=aL 2 +bL+c
[0073] The least squares method requires that each T i The difference between the above models is Tim The sum of squares is minimal:
[0074]
[0075] T im is the model value, T i is the actual value.
[0076]
[0077] According to the multivariate function, find the extreme value and find the partial derivative of V:
[0078]
[0079]
[0080] The values of the three coefficients a, b, and c are obtained as follows:
[0081]
[0082] The present invention has the following beneficial effects: this solution does not need to know the relationship curve between the liquid cooling pipeline flow and the IGBT temperature difference in advance, but sets the corresponding IGBT target temperature value according to the different power values of the PCS module, uses the particle swarm optimization algorithm to find the optimal temperature control parameter value, outputs the PWM optimal duty cycle to drive the proportional valve to adjust the opening, changes the liquid cooling pipeline flow, and reduces the temperature difference. In the iterative process of the particle swarm optimization algorithm, the liquid cooling pipeline flow is obtained according to the PWM optimal duty cycle output by the temperature adjustment module, and the liquid cooling pipeline flow and the IGBT real-time temperature are curve fitted. The corresponding liquid cooling pipeline flow is deduced according to the fitted curve function and the set IGBT target temperature value, and the PWM optimal duty cycle is obtained. The inertia weight is adjusted according to the PWM optimal duty cycle to speed up the iteration speed and accuracy of the particle swarm optimization algorithm. This solution reduces a lot of simulation and actual measurement workload, has strong feasibility, and has higher temperature difference adjustment accuracy and efficiency.
[0083] The above are only specific implementations of the present invention, which cannot be used to limit the scope of the present invention. Equivalent changes made by ordinary technicians in this technical field based on this creation, as well as changes known to technicians in this field, should still fall within the scope of the present invention.
Claims
1. A method for adaptive temperature difference regulation of a liquid-cooled converter. It is characterized in that The method comprises the following steps: Step S1, presetting different IGBT target temperature values Tov according to different power values of the converter module; Step S2, collecting the IGBT real-time temperature value T, and calculating the temperature deviation between the IGBT real-time temperature value T and the IGBT target temperature value Tov according to the power value of the converter module and the IGBT target temperature value Tov; Step S3, calculating the optimal control parameters of the temperature feedback control algorithm by using a particle swarm optimization algorithm according to the temperature deviation; Step S4, outputting a PWM optimal duty cycle according to the temperature deviation and the optimal control parameter, and driving a proportional valve corresponding to the converter module to adjust the opening; In step S3, during the iteration process of the particle swarm optimization algorithm, a curve function of the liquid cooling pipe flow rate L and the IGBT real-time temperature T is fitted according to the values of the liquid cooling pipe flow rate L and the IGBT real-time temperature T, and the liquid cooling pipe flow rate L corresponding to the IGBT target temperature value Tov is obtained according to the curve function, and the PWM optimal duty cycle D is inferred, and the inertia weight of the particle swarm optimization algorithm is adjusted according to the PWM optimal duty cycle D.
2. The method for adaptively adjusting temperature difference of a liquid-cooled converter according to claim 1, It is characterized in that The temperature feedback control algorithm is a PI adjustment algorithm, and the control parameters are a proportional adjustment parameter kp and an integral adjustment parameter ki.
3. The method for adaptive temperature difference adjustment of a liquid-cooled converter according to claim 1, It is characterized in that The temperature feedback control algorithm is a PID control algorithm, and the control parameters are a proportional control parameter kp, an integral control parameter ki, and a differential control parameter kd.
4. The method for adjusting the temperature difference of a liquid-cooled converter according to claim 2, It is characterized in that In step S3, the specific method of calculating the optimal control parameters of the temperature feedback control algorithm by using the particle swarm optimization algorithm is as follows: Step S31, setting the particle position vector, the particle position vector is composed of a proportional adjustment parameter kp and an integral adjustment parameter ki; determining the value range of the proportional adjustment parameter kp and the integral adjustment parameter ki, and randomly generating a matrix as a particle swarm position matrix according to the value range; setting the particle growth rate, learning factor, inertia weight, and maximum number of iterations; initializing the individual historical best position pbest and the global historical best position gbest of each particle; Step S32, assign the particles in the particle swarm position matrix to the proportional adjustment parameter kp and the integral adjustment parameter ki in the PI controller in turn, obtain the corresponding PWM optimal duty cycle D through PI adjustment, drive the proportional valve to adjust the opening, collect the IGBT real-time temperature value T as the initial fitness value of each particle; compare the individual historical best position pbest and the global historical best position gbest with the individual and global current fitness values respectively, and update the individual historical position pbest and the global historical best position gbest according to the optimal values; Step S33, update the position and speed of the particle, obtain the new position value of the particle, assign the new position value of the particle to the proportional adjustment parameter kp and the integral adjustment parameter ki in the PI controller, obtain the corresponding PWM optimal duty cycle D through PI adjustment, drive the proportional valve to adjust the opening, and collect the IGBT real-time temperature value T as the new fitness value of each particle; Compare it with the individual historical best position pbest and the global historical best position gbest, and update the individual historical position pbest and the global historical best position gbest according to the optimal value; The liquid cooling pipe flow L is obtained according to the PWM optimal duty cycle D, the liquid cooling pipe flow L and the IGBT real-time temperature T in each iteration are recorded, and a curve function of the liquid cooling pipe flow L and the IGBT real-time temperature T is fitted according to the recorded values. After a stable curve function is obtained, the liquid cooling pipe flow L corresponding to the IGBT target temperature value Tov is obtained according to the curve function and the IGBT target temperature value Tov, and then the PWM optimal duty cycle D corresponding to the liquid cooling pipe flow L is inversely deduced, and the inertia weight is adjusted according to the PWM optimal duty cycle D; Step S34, determine whether the termination condition is met, if so, jump to step S35, otherwise jump to step S33; Step S35, outputting the global best historical position gbest, and adjusting the opening of the proportional valve according to the global best historical position gbest.
5. The method for adaptively adjusting temperature difference of a liquid-cooled converter according to claim 4, It is characterized in that In the step S33, adjusting the inertia weight means adjusting the inertia weight value to: Where: k is the inertia weight of the kth iteration, D k is the optimal PWM duty cycle of the PI regulation output in the kth iteration, D k ' is the optimal PWM duty cycle of the kth iteration obtained by inverse deduction based on the curve function, n is the number of particles, and α is a constant parameter.
6. The method for adaptively adjusting temperature difference of a liquid-cooled converter according to claim 4, It is characterized in that In the step S33, the method for obtaining the liquid cooling pipe flow L according to the PWM optimal duty cycle D is: obtaining the driving current I according to the PWM optimal duty cycle D and the characteristics of the proportional valve driving circuit, and obtaining the liquid cooling pipe flow L according to the driving current I and the relationship between the driving current and the flow of the proportional valve; the method for inverting the PWM optimal duty cycle D corresponding to the liquid cooling pipe flow L is: obtaining the driving current I according to the relationship between the liquid cooling pipe flow L, the driving current and the flow of the proportional valve, and obtaining the PWM optimal duty cycle D according to the driving current I and the characteristics of the proportional valve driving circuit.
7. The method for adaptively adjusting temperature difference of a liquid-cooled converter according to claim 4, It is characterized in that In the step S33, the criterion for obtaining a stable curve function is: the number of iterations of the particle swarm optimization algorithm reaches a threshold number of iterations.
8. The method for adaptively adjusting temperature difference of a liquid-cooled converter according to claim 4, It is characterized in that In step S33, the criterion for obtaining a stable curve function is: using the curve function fitted in the previous iteration to predict the data of the current iteration, when the difference between the predicted value and the actual value is less than a preset residual threshold, it is considered that the fitted curve function is stable.
9. The method for adaptively adjusting temperature difference of a liquid-cooled converter according to claim 4, It is characterized in that In the step S33, the specific method of fitting the curve function of the liquid cooling pipe flow L and the IGBT real-time temperature T according to the recorded values is: using a multi-order polynomial and a least squares method for fitting.
10. The method for adaptively adjusting temperature difference of a liquid-cooled converter according to claim 4, It is characterized in that In the step S34, the termination condition is: the number of iterations exceeds the maximum number of iterations or the difference between the IGBT real-time temperature T and the IGBT target temperature Tov is less than a preset IGBT temperature difference threshold.
Citation Information
Cited By
Intelligent power adjusting method based on temperature stability and adjusting device thereof
CN119781576A
Intelligent power regulation method based on temperature stability and regulation device thereof
CN119781576B
Temperature management method and system of liquid cooling string PCS energy storage converter
CN119806257A
Special data communication variable frequency pump for air source heat pump
CN120368596A
