Model prediction control strategy method, system and device based on improved current inner loop and medium
By adopting an improved current inner-loop model predictive control strategy, combined with droop control and multi-step model predictive control, the voltage fluctuation problem caused by frequent load switching in DC microgrids was solved, achieving rapid current tracking and voltage stability, thereby improving power quality and system stability.
Patent Information
- Application Number
- CN202511423696.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-30
- Publication Date
- 2026-01-20
AI Technical Summary
Traditional PI control strategies are difficult to cope with the sharp fluctuations in DC bus voltage and power quality problems caused by frequent load switching in DC microgrids. They have slow response speed, current overshoot, and long settling time.
An improved current inner-loop model predictive control strategy is adopted, which combines droop control and multi-step model predictive control. Through real-time data acquisition and evaluation function optimization, the optimal switching signal is output to achieve rapid current tracking and voltage stability.
It significantly improves current response speed, reduces bus voltage fluctuation, improves DC voltage power quality, and enhances system stability and robustness.
Smart Images
Figure CN121367307A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of power control, and in particular to a model predictive control strategy method based on an improved current inner loop, a system, a device and a medium. BACKGROUND
[0002] Global energy shortage, environmental pollution and climate change, and other pressing problems are accelerating the application and development of renewable energy such as wind power and photovoltaic. Direct current microgrid has the advantages of optimal configuration and flexible control of energy storage devices, which effectively reduces the impact of large-scale distributed energy grid connection on the stable operation of the power system. The power electronic converter used in large quantities in the direct current microgrid has extremely fast response speed, which can easily lead to sharp rise and fall of direct current bus voltage and severe fluctuation of output power of intermittent distributed power sources such as wind power and photovoltaic, and thus seriously affect the power quality of direct current voltage.
[0003] The traditional PI control strategy often has the drawbacks of output oscillation and slow response when facing such problems, and it is difficult to maintain the stability of the direct current bus voltage when the load is frequently switched. Therefore, scholars have proposed a model predictive control (MPC) strategy. MPC has the characteristics of fast tracking speed and good stability, and when applied to the inner loop of the control of the battery bidirectional DC-DC converter, it can significantly improve the response speed and provide a new solution for large-scale access of distributed power to the grid.
[0004] In contrast, the traditional PI current inner loop control strategy has a lag and can only act when the current changes, so it has the problems of current overshoot and long stabilization time, and its fast regulation ability is relatively weak. In the case of frequent load switching, this control strategy can easily lead to voltage drop and slow response, which seriously affects the power quality of direct current voltage. SUMMARY
[0005] In view of the above existing problems, the present application is proposed. Therefore, the present application provides a model predictive control strategy method based on an improved current inner loop to solve the problem of sharp fluctuation of the bus voltage of the direct current microgrid based on droop control in the case of frequent load switching, which seriously affects the power quality of direct current voltage.
[0006] To solve the above technical problems, the present application provides the following technical solutions:
[0007] In a first aspect, the present application provides a model predictive control strategy method based on an improved current inner loop, comprising:
[0008] Collecting real-time operation data in the control cycle of the battery;
[0009] Based on the real-time operation data, the outer ring control of the battery adopts a droop control method to dynamically generate a current reference value;
[0010] Based on the real-time operation data and the current reference value, the outer ring control of the battery adopts an improved model predictive control to predict the current value and the voltage value in a control period, and to optimize the predicted values through a set evaluation function, and to output an optimal switching signal.
[0011] As a preferred scheme of the improved current inner ring-based model predictive control strategy method, the real-time operation data collected in the control period of the battery includes a DC bus voltage, a battery current, and a battery terminal voltage.
[0012] As a preferred scheme of the improved current inner ring-based model predictive control strategy method, the outer ring control of the battery includes:
[0013] The droop control obtains a reference value of the output voltage through a droop coefficient, combines the collected real-time battery current, and calculates a reference value of the battery current through a pi loop.
[0014] As a preferred scheme of the improved current inner ring-based model predictive control strategy method, the improved model predictive control includes:
[0015] The model predictive control is performed in multiple steps to track the reference current of the battery without difference, the prediction time domain is set to multiple steps, and only the first step switching state with the optimal control effect is selected as the final output.
[0016] The first evaluation index of the evaluation function of the model predictive control is a load measurement voltage, and the bus voltage is taken as the second evaluation index.
[0017] The preferred scheme has the beneficial effect of effectively reducing the sharp fluctuation of the bus voltage, improving the power quality of the DC voltage, and further reducing the volatility of the DC bus voltage.
[0018] As a preferred scheme of the improved current inner ring-based model predictive control strategy method, the output of the optimal switching signal includes:
[0019] Through the improved model predictive control, based on a Boost / Buck prediction model of the bidirectional DC / DC converter and a bus voltage prediction model, the current prediction value and the voltage prediction value in multiple future control periods are optimized through the evaluation function, and the optimal switching signal is finally output to control the action of the converter.
[0020] Real-time acquisition of battery terminal voltage, bus voltage and battery current as feedback data, forming a closed loop control;
[0021] The beneficial effect of the preferred embodiment is to achieve the dual goals of voltage stabilization and current fast tracking.
[0022] As a preferred scheme of the model predictive control strategy method based on the improved current inner loop, wherein the improved model predictive control further comprises:
[0023] Three-step model predictive control is used for prediction, and the reference current of the battery is tracked without difference.
[0024] In the second evaluation index, only the predicted bus voltage is selected to join the evaluation function.
[0025] As a preferred scheme of the model predictive control strategy method based on the improved current inner loop, wherein the evaluation function of the improved model predictive control is represented as:
[0026] J in =ω1|i bat (k+n)-i bat_ref |+ω2|U dc (k+1)-U dc_ref |
[0027] Wherein, i bat_ref represents the current reference value, i bat (k+n) represents the current prediction result after n steps, ω1 represents the weight coefficient of current tracking, ω2 represents the weight coefficient of bus voltage regulation, U dc (k+1) is the predicted output voltage of the load side at k+1 time, U dc _ ref represents the reference value of the output voltage of the load side, J in represents the evaluation function in the model predictive control.
[0028] Secondly, the application provides a model predictive control strategy system based on the improved current inner loop, comprising:
[0029] The acquisition module is used to acquire real-time running data in the battery control period;
[0030] The outer loop control module is used to generate a current reference value dynamically based on the real-time running data and the outer loop control of the battery using the droop control method.
[0031] The inner loop control module is used for adopting improved model predictive control for outer loop control of the battery based on the real-time operation data and a current reference value, predicting the current value and the voltage value in a control period, optimizing the predicted values through a set evaluation function, and outputting optimal switching signals.
[0032] In a third aspect, the present application provides a computer device, comprising:
[0033] a memory and a processor;
[0034] The memory is used for storing computer executable instructions, and the processor is used for executing the computer executable instructions, which realize the steps of the improved current inner loop based model predictive control strategy method.
[0035] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer executable instructions, which realize the steps of the improved current inner loop based model predictive control strategy method when executed by a processor.
[0036] Compared with the prior art, the present application has the following beneficial effects:
[0037] 1. The present application significantly improves the response speed of the current and effectively reduces the voltage fluctuation. Based on the improved current inner loop control method, the active predictive control strategy is adopted, which performs well in rapidity. The rapidity of the current inner loop is no longer disturbed by the load change, thereby effectively reducing the sharp fluctuation of the bus voltage and improving the power quality of the DC voltage.
[0038] 2. The present application further reduces the voltage fluctuation. By adding the DC side voltage as a secondary evaluation index on the basis of the improved MPC control method, the volatility of the DC bus voltage is further reduced.
[0039] 3. The present application enhances the stability of the system. The traditional proportional integral double closed loop control strategy has multiple PI parameters. Even after optimal parameter setting, overshoot may still occur. The current inner loop enables the FCS-MPC controller to reach the steady state faster, i.e. the tracking speed of the battery current is significantly faster than that of the PI controller, thereby significantly enhancing the robustness of the system. BRIEF DESCRIPTION OF DRAWINGS
[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0041] Figure 1 The overall flowchart of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure.
[0042] Figure 2 The battery voltage control block diagram of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure.
[0043] Figure 3 The reference value diagram of the battery voltage output by the droop control of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure.
[0044] Figure 4 The prediction principle diagram of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure.
[0045] Figure 5 The model predictive control flowchart of the improved current inner loop based on the droop control of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure.
[0046] Figure 6 The simulation waveform diagram of the actual value and the reference value of the output voltage of the MPC-bidirectional DCDC of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure.
[0047] Figure 7 The simulation waveform diagram of the actual value and the reference value of the output current of the battery of the MPC-bidirectional DCDC of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure.
[0048] Figure 8 The simulation waveform diagram of the battery soc of the MPC-bidirectional DCDC of the model predictive control strategy method based on the improved current inner loop according to an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION
[0049] In order to make the above objectives, features and advantages of the present application more apparent, clear and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should belong to the protection scope of the present application.
[0050] REFERENCE Figures 1-8 According to an embodiment of the present application, a model predictive control strategy method based on an improved current inner loop is provided, as shown inFigure 1 As shown, comprising:
[0051] S101, collecting real-time running data in the control cycle of the battery;
[0052] S102, based on the real-time running data, the outer loop control of the battery adopts the droop control method to dynamically generate the current reference value;
[0053] S103, based on the real-time running data and the current reference value, the outer loop control of the battery adopts the improved model predictive control to predict the current value and the voltage value in the control cycle, and to optimize the predicted value through the set evaluation function, and output the optimal switching signal.
[0054] It should be noted that the improved inner loop model predictive control reduces the error of the output current of the battery compared with the traditional pi control method. Through the improved model control algorithm, the output current of the battery can follow the reference value without error and suppress the output voltage fluctuation of the load meter, reduce the fluctuation of voltage and current, and improve the response speed and stability of the load meter voltage.
[0055] The improved model predictive control strategy realizes that the output voltage can quickly follow the voltage reference value output by the droop control and reduce the volatility of the voltage through the multi-step prediction model and the addition of the constraint of the direct current side voltage. It improves the current inner loop control structure and avoids the sharp fluctuation of the bus voltage of the direct current micro-grid when the load is frequently switched.
[0056] In a preferred embodiment, collecting real-time running data in the control cycle of the battery includes direct current bus voltage, battery current and battery terminal voltage.
[0057] In a preferred embodiment, the outer loop control of the battery comprises:
[0058] The droop control obtains the reference value of the output voltage through the droop coefficient, combines the collected real-time battery current, and calculates the reference value of the battery current through the pi loop.
[0059] In a preferred embodiment, the improved model predictive control comprises:
[0060] The multi-step model predictive control is used for prediction, and the reference current of the battery is tracked without error. The prediction time domain is set to multiple steps, and only the first step switching state with the optimal control effect is selected as the final output.
[0061] The first evaluation index of the evaluation function of the model predictive control is the load meter voltage, and the bus voltage is taken as the second evaluation index.
[0062] It should be noted that the scheme adopts a multi-step prediction model to effectively reduce the deviation, which ensures tracking accuracy and avoids excessive increase in controller calculation amount, so three steps are selected as the best scheme. Meanwhile, the control structure is optimized, and the rapid response regulation ability of the current is significantly improved.
[0063] In a preferred embodiment, the output optimal switching signal comprises:
[0064] Through the improved model predictive control, the Boost / Buck prediction model of the bidirectional DC / DC converter and the bus voltage prediction model, the future current prediction value and the voltage prediction value of multiple control periods are optimized through the evaluation function, and finally the optimal switching signal is output to control the action of the converter;
[0065] The battery terminal voltage, bus voltage and battery current are collected in real time as feedback data to form a closed-loop control.
[0066] Specifically, as shown in Figure 2 , it is a battery voltage control block diagram of the application, which shows a double-closed-loop battery management system combining droop control and model predictive control (MPC). Among them, the outer loop control adopts U-I droop control, according to the DC bus voltage reference value (400V), the droop coefficient K d and the real-time battery current i bat , through the formula U dc_ref =400-K d *i bat get U dc_ref , and then generate the current reference value i bat_ref through the outer loop control to realize the global regulation of the bus voltage.
[0067] The inner loop control adopts the improved model predictive control, based on the Boost / Buck prediction model of the bidirectional DC / DC converter and the bus voltage prediction model, the current prediction value i bat (k+3) and the voltage prediction value U dc (k+1) of the future 3 control periods are optimized through the evaluation function, and finally the optimal switching signal g is output to control the action of the converter, the system collects the battery terminal voltage U bat (k), the bus voltage U dc (k) and the battery current i bat (k) at time k as feedback to form a closed-loop control, so as to realize the dual goals of voltage stability and current fast tracking.
[0068] In an alternative embodiment, the improved model predictive control can also be an adaptive model predictive control with online parameter identification, by adding an online parameter identifier, real-time estimation of key parameters of the system, and dynamic updating of the prediction model in the MPC controller, so that it is always highly consistent with the real system. The estimated values of parameters such as inductance and resistance are updated in real time and online, and are updated into the prediction model of the MPC, and the optimal switching signal is output through the evaluation function optimization.
[0069] In another alternative embodiment, the improved model predictive control can also use a continuous control set model predictive control, which first optimizes a continuous and ideal control quantity (such as the average voltage or duty cycle D of the converter), and then delivers the continuous control quantity to a fixed modulator to generate the switching signal.
[0070] As shown in Figure 3 , it is a three-step current inner loop model prediction schematic diagram used in the application. The solid black dot in the figure is the optimal switching state at this moment. As can be seen from the figure, when the prediction step is 3, the optimal state is J5, and the first step switching state output at this time is 0. If the switching state displayed by the one-step prediction result is 1, it will affect the control effect of the system.
[0071] As shown in Figure 4 , it shows the improved current inner loop model predictive control flowchart based on droop control used in the application. In the figure: bat (k) is the output current of the battery k, i bat (k+3) is the predicted output current of the battery k+3, U bat (k) is the output voltage of the battery k, U dc (k) is the output voltage of the load k, U dc (k+1) is the predicted output voltage of the load side k+1, m is the prediction result at the current moment. Since each working mode has two control variables, i.e. the conduction and turn-off of the switching tube, there are 2 n n-step prediction results, i.e. 8 prediction results in this embodiment.
[0072] In a preferred embodiment, the improved model predictive control further comprises:
[0073] Using three-step model predictive control for prediction, the reference current of the battery is tracked without difference;
[0074] In the second evaluation index, only the bus voltage predicted one step is selected into the evaluation function to ensure that the system stability is not affected, and at the same time, the battery current is prevented from out of control.
[0075] In a preferred embodiment, the evaluation function is expressed as:
[0076] J in = ω1|i bat (k+n)-i bat_ref |+ ω2|U dc (k+1)-U dc_ref |
[0077] where i bat_ref represents the current reference value, i bat (k+n) represents the current prediction result after n steps, ω1 represents the weight coefficient of current tracking, ω2 represents the weight coefficient of bus voltage regulation, U dc (k+1) is the predicted output voltage of the load side at time k+1, U dc _ ref represents the reference value of the output voltage of the load side, J in represents the evaluation function in model predictive control.
[0078] It should be noted that the value of ω2 needs to be reasonably set, otherwise it may cause system stability problems, leading to battery current regulation failure, and the values of ω1 and ω2 are 0.9 and 0.1 respectively. In the selection of prediction step, too short prediction step (such as n<3) will reduce the current tracking accuracy, while too long step will significantly increase the calculation burden of the controller. Therefore, this embodiment considers the control accuracy and calculation efficiency comprehensively, and sets the prediction step to n=3, that is, the three-step prediction value of the battery current is included in the evaluation function. For the bus voltage index, since its regulation priority is relatively low, only one-step prediction value U dc (k+1) is used in the evaluation function calculation to balance system performance and calculation complexity.
[0079] The biggest difference between it and the traditional double-loop control strategy is: no difference in following the current reference value; the constraint of the load voltage is introduced into the evaluation function of MPC. The specific description is as follows:
[0080] As Figure 5 shown, generally, when the bidirectional converter is in Boost mode (g1=0): g2=1, the energy storage inductor stores energy, and the bus connecting capacitor releases energy; g2=0, the energy storage inductor releases energy, and the bus connecting capacitor stores energy, which is expressed as:
[0081]
[0082] When the bidirectional converter is in Buck mode (g2=0): g1=1, the energy storage inductor stores energy, and the bus connecting capacitor releases energy; g1=0, the energy storage inductor releases energy, and the bus connecting capacitor stores energy, which is expressed as:
[0083]
[0084] Where g1 and g2 represent the switching states of transistors S1 and S2, respectively (g1 = 1 indicates the transistor is on, g1 = 0 indicates the transistor is off; g2 is similar). Assume i bat i dc The positive direction is during battery discharge, where L represents the battery-side inductance, C represents the load-side capacitance, and i bat Indicates the battery current, i dc U represents the load-side current. bat U represents the battery voltage. dc This indicates the voltage on the load side of the battery.
[0085] The prediction models for battery current in Boost mode and Buck mode can be obtained from the above equations, as shown in the following equation: when the bidirectional converter is in Boost mode (g1 = 0), the predicted current i at time k+1 is... bat (k+1) is represented as:
[0086]
[0087] When the bidirectional converter is in Buck mode (g2 = 0), the predicted current i at time k+1 is... bat (k+1) is represented as:
[0088]
[0089] Among them, T s Indicates the switching cycle.
[0090] The predicted value of the battery current is derived using the above formula, enabling seamless tracking of the current reference value. For example... Figure 6 and Figure 7 As shown, the output value predicted by the current model can always follow the reference value of the droop control output, achieving accurate and error-free tracking. The settling time of the output voltage is basically consistent with the reference value, and the error is within the actual range. Even if the load is switched on or off in the fifth second, the system can still achieve stable tracking. Figure 8 It demonstrates the battery's SOC state and output voltage, and shows that it maintains normal operation even under varying load conditions.
[0091] In an alternative implementation, the evaluation metrics of the evaluation function can also include converter switching losses and current ripple metrics. By adding penalty terms for switching frequency and current ripple magnitude to the evaluation function, the MPC can be guided to select a switching state that is more "friendly" to the converter while ensuring voltage performance.
[0092] In another alternative embodiment, the evaluation index of the evaluation function can also introduce the bus capacitance current or power balance index, and the traditional bus voltage index is a post-compensation, which corrects only after the voltage drop. By introducing the bus capacitance current prediction or power differential as the evaluation index, the pre-inhibition of the disturbance is realized.
[0093] It should be noted that in the case of frequent load switching, the bus voltage of the DC microgrid based on droop control appears sharp fluctuation, which significantly affects the power quality of the DC voltage. The battery current inner loop plays a key role, which ensures that the battery current can closely follow the current reference value set by the droop control ring through accurate regulation of the switching state of the insulated gate bipolar transistor (IGBT) in the bidirectional DC-DC converter. In view of the high requirement of the inner loop for fast response, the traditional PI controller is difficult to meet this demand due to its inherent lagging characteristics. In contrast, model predictive control, as a proactive predictive control method, has a significant advantage in terms of speed. Model predictive control can meet the requirements in terms of speed because it is a proactive predictive control.
[0094] The present application innovates on the basis of traditional model predictive control and proposes an improved model predictive function, aiming to further improve the stability of the system. The core of this method is to use a multi-step prediction model to optimize the control effect by reducing the steady-state deviation. Specifically, when performing multi-step prediction, only the first step switching state with the best control effect is selected as the final output. In addition, in order to achieve fast response and good stability at the same time, the present application also introduces the DC side voltage as a secondary evaluation index into the evaluation function, thereby ensuring the excellent performance of the system under complex working conditions.
[0095] The above is a schematic scheme of the model predictive control strategy method based on the improved current inner loop of the present embodiment. It should be noted that the technical scheme of the model predictive control strategy system based on the improved current inner loop belongs to the same concept as the technical scheme of the model predictive control strategy method based on the improved current inner loop described above. The technical scheme of the model predictive control strategy system based on the improved current inner loop in the present embodiment, which is not described in detail, can be referred to the description of the technical scheme of the model predictive control strategy method based on the improved current inner loop.
[0096] The present embodiment provides a model predictive control strategy system based on an improved current inner loop, comprising:
[0097] The acquisition module is configured to acquire real-time operation data within a battery control period.
[0098] The outer loop control module is configured to generate a current reference value dynamically based on the real-time operation data and the outer loop control of the battery using the droop control method.
[0099] The inner loop control module is used for adopting improved model prediction control for the outer loop control of the battery based on real-time running data and a current reference value, predicting the current value and the voltage value in a control period, optimizing the predicted values through a set evaluation function, and outputting optimal switching signals.
[0100] The embodiment also provides a computer device suitable for the case of implementing the model prediction control strategy based on the improved current inner loop, comprising:
[0101] The memory is used for storing computer executable instructions, and the processor is used for executing the computer executable instructions to implement the method for implementing the model prediction control strategy based on the improved current inner loop.
[0102] The embodiment also provides a storage medium having a computer program stored thereon, and the program is executed by a processor to implement the method for implementing the model prediction control strategy based on the improved current inner loop.
[0103] The storage medium proposed in the embodiment belongs to the same inventive concept as the method for implementing the model prediction control strategy based on the improved current inner loop proposed in the above embodiment, and the technical details not described in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.
[0104] From the above description about the embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary universal hardware, and of course can be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH memory, a hard disk, or an optical disc, and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.
[0105] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.
Claims
1. A method for model predictive control strategy based on improved current inner loop, characterized in that, The method comprises the following steps: collecting real-time operation data in a battery control period; based on the real-time operation data, the outer loop control of the battery adopts droop control method to dynamically generate current reference value; based on the real-time operation data and the current reference value, the outer loop control of the battery adopts improved model predictive control to predict the current value and voltage value of the control period, and to optimize the predicted value through the set evaluation function, and to output the optimal switching signal.
2. The method of claim 1, wherein the improved current inner loop model predictive control strategy is based on, The real-time operation data collected in the battery control period includes DC bus voltage, battery current and battery terminal voltage.
3. The method of claim 2, wherein the improved current inner loop model predictive control strategy is based on a model predictive control strategy. The outer loop control of the battery comprises: The droop control obtains the reference value of the output voltage through the droop coefficient, and obtains the reference value of the battery current through the pi loop calculation combined with the collected real-time battery current.
4. The method of claim 3, wherein the improved current inner loop model predictive control strategy is based on a model predictive control strategy. The improved model predictive control comprises: The multi-step model predictive control is used for prediction, and the reference current of the battery is tracked without difference, and the prediction time domain is set to multiple steps, and only the first step switching state with the optimal control effect is selected as the final output; The first evaluation index of the evaluation function of the model predictive control is the load measurement voltage, and the bus voltage is taken as the second evaluation index.
5. The method of model predictive control strategy based on improved current inner loop of claim 4, wherein, The optimal switching signal output comprises: Through the improved model predictive control, based on the Boost / Buck prediction model of the bidirectional DC / DC converter and the bus voltage prediction model, the current prediction value and the voltage prediction value of the future multiple control periods are optimized through the evaluation function, and the optimal switching signal is finally output to control the action of the converter; The battery terminal voltage, the bus voltage and the battery current are collected in real time as feedback data to form a closed loop control.
6. The method of claim 5, wherein the improved current inner loop model predictive control strategy is based on, The improved model predictive control further comprises: The three-step model predictive control is used for prediction, and the reference current of the battery is tracked without difference; In the second evaluation index, only the bus voltage of the predicted step is selected to join the evaluation function.
7. The method of claim 6, wherein the improved current inner loop model predictive control strategy is based on a model predictive control strategy. The evaluation function of the improved model predictive control is expressed as: J in = ω1|i bat (k+n)-i bat_ref |+ω2|U dc (k+1)-U dc_ref | wherein, i bat_ref represents the current reference value, i bat (k+n) represents the current prediction result after n steps, ω1 represents the weight coefficient of current tracking, ω2 represents the weight coefficient of bus voltage regulation, U dc (k+1) is the predicted output voltage of the load side at time k+1, U dc _ ref represents the reference value of the output voltage of the load side, J in represents the evaluation function in model predictive control.
8. A system for model predictive control strategy based on improved inner current loop, applying a method for model predictive control strategy based on improved inner current loop as claimed in any one of claims 1 to 7, characterized in that, The method comprises the following steps: a collection module for collecting real-time operation data in a battery control period; an outer loop control module for, based on the real-time operation data, the outer loop control of the battery adopting droop control method to dynamically generate current reference value; an inner loop control module for, based on the real-time operation data and the current reference value, the outer loop control of the battery adopting improved model predictive control to predict the current value and voltage value of the control period, and to optimize the predicted value through the set evaluation function, and to output the optimal switching signal.
9. A computer device, comprising: The method comprises the following steps: a memory and a processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions, which realize the steps of the model predictive control strategy method based on the improved current inner loop according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The memory has stored computer executable instructions, which realize the steps of the model predictive control strategy method based on the improved current inner loop according to any one of claims 1 to 7 when executed by the processor.