Conduct a domestic and international research status analysis on the energy composition and optimization control strategies of hybrid energy storage power supply systems.
1.Current research status of energy composition in hybrid energy storage power supply systems
In foreign countries, research on hybrid energy storage power systems for new energy vehicles started earlier. FIAT has proposed a hybrid energy storage power system consisting of lead-acid batteries and supercapacitors to drive vehicles. This structure can reduce the workload of lead-acid batteries, quickly restore the braking energy of vehicles, and improve fuel economy. However, due to the disadvantages of heavy weight, low energy density, and inability to support high-power charging and discharging, lead-acid batteries will lead to increased internal consumption, shortened service life, shortened driving range, and decreased climbing and acceleration performance. To address the aforementioned issues with lead-acid batteries, Alireza Khaligh et al. proposed a hybrid energy storage power supply system consisting of a supercapacitor lithium battery composite power supply and two brushless DC motors. The test results have shown that due to the high-power charging and discharging characteristics of lithium batteries, the working load can be slowed down after the addition of supercapacitors, and the braking energy of the entire vehicle can be quickly recovered, thereby improving the overall economy of the vehicle. However, due to the limited energy storage capacity of lithium batteries and supercapacitors, the vehicle’s endurance will be weak. To address the issue of weak endurance, Amin and Trilaksono Bambang proposed a hybrid energy storage power system consisting of fuel cells, lithium batteries, and supercapacitors. During car operation, fuel cells provide the average power required by the car, lithium batteries provide insufficient power or absorb excess power, and supercapacitors are used to recover and output instantaneous high currents. This topology can reduce the power requirements of lithium batteries, thereby extending their service life. However, due to the complex energy composition in hybrid energy storage power supply systems, it is difficult to couple energy between different power sources.
In China, Zhang Meng proposed a hybrid energy storage power system consisting of fuel cells and lithium batteries. The hybrid energy storage power system adds a lithium battery on the basis of fuel cells, which can recover the energy generated by the vehicle’s regenerative braking and improve the efficiency of the hybrid energy storage power system. But as the number of uses of lithium batteries increases, their capacity will gradually decline, and when the lithium battery is in an overcharged state, it will lead to a shortened lifespan. Zhang Junkai proposed a hybrid energy storage power system consisting of fuel cells, lithium batteries, and supercapacitors. Due to the addition of supercapacitors, high-power charging and discharging of lithium batteries are avoided and the usage frequency is reduced, achieving the idea of using supercapacitors to “cut peaks and fill valleys” for lithium batteries and extending their lifespan. This topology also verifies that the three energy source configuration has better fuel economy compared to the two energy source configuration. However, in a multi energy source power system, the energy composition of a hybrid energy storage power system needs to integrate the working mode and energy coupling of the hybrid energy storage power system.
In summary, the structure of the hybrid energy storage power system composed of fuel cells, lithium batteries, and supercapacitors is relatively complex, and the energy coupling between each power source is difficult. Moreover, there is relatively little exploration of it both domestically and internationally, and a systematic research theory for hybrid energy storage power has not yet been formed. Therefore, the research on the energy composition of hybrid energy storage power systems is still worth further exploration.
2. Current status of research on optimization control strategies
There is relatively more research on the optimization control of hybrid energy storage power supply systems abroad. The MPC algorithm is based on different prediction models and adopts the rolling optimization principle, which has the advantages of strong robustness and accurate control effect. Therefore, most of the optimization control strategies for hybrid energy storage power systems currently use the MPC algorithm. Amin et al. adopted an MPC control strategy by inputting the voltage of each energy source to output its reference current, and corrected the results with bus voltage and load current. Finally, the duty cycle was obtained through hysteresis control and input to each energy source. This strategy can achieve optimal economic efficiency for hybrid energy storage power systems. However, this strategy did not consider the driver model, resulting in poor control performance. Josevski et al. proposed a pipeline MPC that overcomes the lack of a driver model established in the control strategy. Limiting the constraint conditions within a tubular range and solving the optimization problem to obtain the minimum fuel consumption has high economy and stability. However, the application of this strategy requires prior knowledge of the vehicle’s historical driving information, which has limitations for optimizing control of hybrid energy storage power systems. Josevski et al. proposed a control strategy based on Stochastic Model Predictive Control (SMPC), which does not require prior knowledge of the driving cycle, but is based on the vehicle’s latent stochastic model for online real-time prediction, greatly reducing fuel consumption. But this strategy cannot be updated in real time, thereby weakening the adaptability of the working conditions. Cairano et al. proposed a Stochastic Model Predictive Control Learning (SMPCL) control strategy based on stochastic learning (MPC). This strategy updates the Markov chain model in real-time through online learning on the basis of SMPC, optimizes vehicle energy control using MPC, and achieves better fuel economy. However, this strategy can only optimize control for hybrid energy storage power systems with single or dual control objectives, and cannot be applied to hybrid energy storage power systems with multiple control objectives. Homchaudhuri et al. proposed a hierarchical energy management method by combining the Internet of Vehicles with MPC. This control strategy divides the controller into upper and lower layers to achieve multi-objective optimization control.
The research on control strategies for hybrid energy storage power supply systems in China began in the early 20th century. Shi Xiaofeng combined dynamic programming algorithm with MPC and adopted a hierarchical control optimization control strategy. By predicting future time domain vehicle speed information, he optimized and solved it, achieving good results. However, its computational speed is relatively slow, and it is necessary to improve the solving speed on this basis. Lian Jing et al. proposed an MPC control strategy based on vehicle speed prediction, which not only quickly predicts vehicle speed but also enhances the optimization effect of hybrid energy storage power systems. However, this control strategy is only applicable to linear systems and cannot be applied to nonlinear systems. Zhou et al. proposed the Adaptive Model Predictive Control (AMPC) strategy, which solves the nonlinear and uncertainty problems of hybrid energy storage power systems, thereby improving control accuracy and extending battery life. However, this strategy has limited control objectives for hybrid energy storage power systems and cannot be applied in systems with multiple control objectives. FU et al. proposed a layered MPC control strategy, where the upper controller optimizes the optimal trajectory of battery SOC, while the lower controller controls the performance of the hybrid energy storage power supply system. This strategy not only increases the control objectives of hybrid energy storage power supply systems, but also improves control performance.
The above series of MPC control strategies have good control performance, but due to their algorithm set points being manually set and not involving set point optimization problems, they cannot be optimized and updated in real-time based on the state of the hybrid energy storage power supply system, resulting in a decrease in efficiency and an increase in cost of the hybrid energy storage power supply system. The double-layer structure MPC algorithm includes a steady-state target calculation layer and a dynamic optimization control layer. Its set points are calculated by the steady-state target calculation layer through linear programming (LP). During the operation of the hybrid energy storage power supply system, it can timely update the future output reference values based on changes in the hybrid energy storage power supply system, thereby enhancing the accuracy of the hybrid energy storage power supply system. But now most of them have not adopted the more advanced double-layer structure MPC algorithm in the industrial field.
Overall, the dynamic response characteristics of hybrid energy storage power systems composed of fuel cells, lithium batteries, and supercapacitors are complex and face challenges in multi time scale characteristics. Existing optimization control strategies lack a description of the dynamic response characteristics of hybrid energy storage power systems, making it difficult to simultaneously achieve safety control and performance optimization of hybrid energy storage power systems.
Based on the analysis of the current research status on the energy composition and optimization control strategies of hybrid energy storage power supply systems, it can be concluded that there are three problems with existing hybrid energy storage power supply systems: firstly, hybrid energy storage power supply systems often establish energy balance models from a steady-state perspective, without forming a unified dynamic model for multi energy coupling; Secondly, the existing optimization control strategies have not yet taken into account the multi time scale response characteristics of hybrid energy storage power supply systems, making it difficult to balance fast and slow dynamics to form an integrated optimization control strategy; Thirdly, due to the rapid response of hybrid energy storage power supply systems, existing control strategies can only use rule-based or offline optimization control structures, resulting in a decrease in the performance of hybrid energy storage power supply systems.
In response to the above issues, starting from the core challenges faced by hybrid energy storage power systems such as “multi energy complementarity, multiple time scales, and multi agent collaboration”, research will be conducted on unified dynamic modeling, rapid response, real-time scheduling, and collaborative optimization control of hybrid energy storage power systems. Comprehensive energy management and real-time optimization control strategies will be constructed for hybrid energy storage power systems to ensure the safe operation of hybrid energy storage power systems Provide support for improving energy utilization efficiency.
It mainly includes the following three contents:
(1) Unified dynamic modeling of components in hybrid energy storage power supply systems
Based on the variation law of energy flow in the “source storage load” energy network, a combination of mechanism modeling and data-driven methods is used to study the modeling theory under the three power supply topology structure, and to construct a dynamic model of the components of the hybrid energy storage power supply system;
(2) Study on the multi time scale response characteristics under the interaction mode of three power sources
In response to the multi time scale response characteristics of the three power sources in the hybrid energy storage power supply system, this paper solves the problem of multi time scale response characteristics of the three power sources under external excitation based on event driven mechanism, enabling all three power sources with multi-scale response to respond quickly, and solving the problem of slow state tracking and disturbance suppression response speed in the dynamic/steady-state transition process of the hybrid energy storage power supply system;
(3) Real time scheduling and dynamic control of fast response hybrid energy storage power supply system
In response to the multimodal, highly nonlinear, and complex control methods of the fast response hybrid energy storage power supply system, this paper proposes a dual layer MPC optimization control strategy based on integrated steady-state objective calculation and dynamic optimization control, which enables the fast response hybrid energy storage power supply system to achieve high-precision control under multi energy coupling constraints; And in response to the economic optimization goals and external demands of the hybrid energy storage power supply system under different operating conditions, a coordination mechanism between the external demand response goals and the energy efficiency goals of the hybrid energy storage power supply system is achieved under the hierarchical hybrid energy storage power supply system architecture, thereby conducting real-time scheduling of the three power sources.