Hybrid energy storage system capacity optimization configuration method based on improved NSGA-II algorithm
Through the improved NSGA-II algorithm, the capacity configuration of the hybrid energy storage system is optimized, which solves the problem of insufficient power supply of the hybrid energy storage system, reduces the entire life cycle cost and improves the economic and stability of the system.
Patent Information
- Application Number
- CN202410949945.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-08
AI Technical Summary
The capacity configuration problem of hybrid energy storage systems leads to insufficient power supply, which increases system costs, and it is difficult for the existing technology to effectively optimize capacity configuration, affecting the normal operation of the equipment.
Using the improved NSGA-II algorithm, a multi-objective capacity optimization model aimed at the minimum annual investment cost and minimum load power failure rate is established. Through cross-section, variation and local search operations, the rated capacity of vanadium batteries and supercapacitors will be optimized and Power ensures the economic and stability of the system.
It reduces the full life cycle cost of energy storage configuration, reduces the load power shortage rate, improves the power supply reliability and economy of the system, and avoids the problem of insufficient power supply.
Smart Images

Figure CN118920540A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to an energy storage system optimization configuration method, in particular to a hybrid energy storage system capacity optimization configuration method. Background Art
[0002] Photovoltaic power generation occupies an important position in renewable energy with its simple power generation process and zero emission characteristics. However, photovoltaic energy is dependent on the weather and its output power fluctuates greatly. The energy storage system provides an effective solution to this problem. It can store excess electricity, release it in time, and buffer photovoltaic power fluctuations. At present, due to the diversified application requirements of microgrids for energy storage systems, a single energy storage technology cannot fully meet the requirements. Therefore, hybrid energy storage systems have become a research field that has attracted much attention. The combination of all-vanadium liquid flow batteries and supercapacitors provides hybrid energy storage systems with high capacity, long life and high power output, improves power supply reliability, and meets different load requirements. However, the high price of energy storage equipment may increase system costs, and its improper configuration may lead to insufficient power supply and affect the normal operation of equipment. Therefore, the capacity configuration problem of hybrid energy storage systems has become the focus of current academic attention. Summary of the invention
[0003] The present invention provides a hybrid energy storage system capacity optimization configuration method to solve the capacity configuration problem of a hybrid energy storage system composed of an all-vanadium liquid flow battery and a supercapacitor.
[0004] In order to achieve the above object, the technical solution adopted by the present invention is:
[0005] A capacity optimization configuration method for a hybrid energy storage system based on an improved NSGA-Ⅱ algorithm, wherein the hybrid energy storage system is composed of a vanadium battery and a supercapacitor, comprising the following steps:
[0006] Step 1: Capacity optimization model construction: Taking the annual average investment cost of the hybrid energy storage system and the load power shortage rate as the minimum as the goal, establish the objective function and establish constraints, thereby obtaining a multi-objective capacity optimization model for the hybrid energy storage system;
[0007] Step 2: Use the improved NSGA-II algorithm to solve the multi-objective capacity optimization model of the hybrid energy storage system established in step 1 to obtain the optimal solution set of the rated capacity and rated power of the vanadium battery and supercapacitor. The improved NSGA-II algorithm process is as follows:
[0008] (2.1), initialize the population and set the maximum number of cyclic generations;
[0009] (2.2) Perform crossover and mutation on the initial population, thereby generating new individuals in the population;
[0010] (2.3) Use non-dominated sorting and crowding distance sorting to distinguish the individuals in the population after crossover and mutation in step 2, and determine the superiority and inferiority relationship between individuals;
[0011] (2.4) Introduce local search operation for individuals with dominance priority 1, compare the individuals after search with the previous individuals, retain individuals with better performance, and thus find a better solution;
[0012] (2.5) A new fitness function is introduced to prioritize solutions that are closer to the Pareto frontier, improving the algorithm's optimization ability in the later stages. Solutions that have not been confirmed to be retained are still distinguished based on non-dominated sorting and crowding distance sorting;
[0013] The currently best individuals retained constitute the best solution set, thus forming a new population;
[0014] (2.6) Determine whether the loop algebra has reached the maximum loop algebra. If not, repeat steps (2.2)-(2.5); if it has reached, output the current optimal solution set.
[0015] In a further step 1, the charging and discharging power constraint, SOC constraint and power constraint of the hybrid energy storage unit are used as constraint conditions.
[0016] In the further step 2, a local search operation is introduced to perform a local search around the point with a non-dominated sorting level of 1. The individuals after the search are compared with the previous individuals, and the individuals with better performance are retained, so as to find a better solution to ensure the uniform distribution and extensiveness of the solution.
[0017] Compared with the prior art, the present invention has the following advantages:
[0018] The present invention aims at an independent photovoltaic hybrid energy storage system, establishes a multi-objective capacity configuration model with the goal of minimizing the average annual investment cost of the energy storage system and minimizing the load power shortage rate, and optimizes the capacity of the hybrid energy storage device in the system. Compared with the traditional NSGA-II algorithm, the improved NSGA-II algorithm of the present invention has higher convergence accuracy and is easier to jump out of the local optimal solution. At the same time, the full life cycle cost of the energy storage configuration is also reduced, and the volatility is reduced, which further verifies the superiority of the improved NSGA-II algorithm of the present invention in the capacity optimization configuration of the hybrid energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 It is a structural diagram of an independent photovoltaic microgrid hybrid energy storage system in an embodiment of the present invention.
[0020] Figure 2 Schematic diagram of comparing advantages and disadvantages in the embodiments of the present invention.
[0021] Figure 3 It is an optimization flow chart of the improved NSGA-Ⅱ algorithm in an embodiment of the present invention.
[0022] Figure 4 It is a typical day photovoltaic and load power variation curve in the case analysis.
[0023] Figure 5 It is the power variation curve of VRB and SC in the case analysis.
[0024] Figure 6 It is the result of multi-objective capacity optimization configuration of hybrid energy storage in the case analysis. DETAILED DESCRIPTION
[0025] The present invention is further described below in conjunction with the accompanying drawings and embodiments.
[0026] This embodiment discloses a method for optimizing the capacity configuration of a hybrid energy storage system, wherein the hybrid energy storage system is composed of a vanadium battery and a supercapacitor, and comprises the following steps:
[0027] Step 1: Build a capacity optimization model.
[0028] With the goal of minimizing the average annual investment cost of the hybrid energy storage system and minimizing the load power shortage rate, the objective function is established, and the charging and discharging power constraints, SOC constraints, and power constraints of the hybrid energy storage unit are used as constraints, thereby obtaining a multi-objective capacity optimization model for the hybrid energy storage system. The specific description is as follows:
[0029] The randomness and volatility of light energy lead to large fluctuations in photovoltaic power output, which makes it difficult to directly supply the load. Introducing energy storage systems to the photovoltaic side can effectively alleviate the problem of photovoltaic power fluctuations. Figure 1 As shown, Figure 1 The topological structure of the energy storage photovoltaic system considered in this embodiment includes a photovoltaic micro-source, a hybrid energy storage system and a load unit. The hybrid energy storage system is composed of a vanadium battery and a supercapacitor, and a P PV (t i ) is the total real-time power of the photovoltaic unit, P VRB (t i ) is the real-time charging and discharging power of the vanadium battery, P SC (t i ) is the real-time charging and discharging power of the supercapacitor, P Load (t i ) is the real-time power of the load.
[0030] When the system is running, the photovoltaic unit and the energy storage system jointly power the load in the microgrid. Without considering energy loss, the unbalanced power in an independent microgrid can be defined as the difference between the load demand and the photovoltaic power generation. The power balance relationship between the units is shown in the following formula:
[0031] P Load (t i )=P PV (t i )+P VRB (t i )+P SC (t i ) (1)
[0032] The hybrid energy storage system described in this embodiment is composed of a vanadium battery VRB and a supercapacitor SC. Therefore, the power of the hybrid energy storage system P hess As shown below:
[0033] P hess (t i )=P VRB (t i )+P SC (t i ) (2)
[0034] In an independently operated photovoltaic hybrid energy storage system, the hybrid energy storage system is mainly used to supply power to the load. Among them, the all-vanadium liquid flow battery and supercapacitor are used from the beginning to t i The calculation formula for the accumulated charge and discharge power at all times is:
[0035]
[0036] In formula (3): E VRB (0), E SC (0) are the initial capacities of the vanadium liquid flow battery and supercapacitor respectively; E VRB (t i )、E SC (t i ) are respectively energy storage from 0 to t i The accumulated charge and discharge power in a certain period of time. The rated capacity of the all-vanadium liquid flow battery and supercapacitor can be obtained as shown in the following formula:
[0037]
[0038] In formula (4): E VRB 、E SC are the rated capacities of all-vanadium liquid flow batteries and supercapacitors; SOC VRB,max , SOC VRB,min , SOC SC,max , SOC SC,minThey are the maximum and minimum states of charge of all-vanadium liquid flow battery and supercapacitor, respectively.
[0039] Based on the photovoltaic hybrid energy storage system established above, this embodiment uses two indicators to establish a multi-objective capacity optimization model for the hybrid energy storage system, namely the average annual investment cost of the energy storage system and the load power shortage rate, which characterize the economy and stability of the system, and establishes an objective function with the goal of minimizing the average annual investment cost of the hybrid energy storage system and minimizing the load power shortage rate. The objective function is as follows:
[0040]
[0041] In formula (5): C TC is the total investment cost function of energy storage; f LPSP is the load power failure rate function. Where:
[0042] (a) Objective function 1: Minimize the average annual investment cost of energy storage
[0043] The average annual investment cost of energy storage is calculated by the average annual purchase cost of energy storage batteries C I , average annual maintenance cost C OM , Annual replacement cost C R and the average annual salvage value C RES The composition is shown in the following formula:
[0044] C TC =C I +C OM +C R -C RES (6)
[0045] The average annual purchase cost of energy storage system C I The calculation formula is as follows:
[0046]
[0047] In formula (7): C BAT is the unit cost of energy storage device; C BOP is the cost of auxiliary equipment; C PCS is the cost of DC / DC and DC / AC; C SC,E and C VRB,E are the unit capacity costs of all-vanadium liquid flow batteries and supercapacitors respectively; C SC,P and C VRB,P are the unit power costs of all-vanadium liquid flow batteries and supercapacitors respectively; C SC,B and C VRB,B are the unit capacity costs of all-vanadium liquid flow battery and supercapacitor auxiliary device respectively; C SC,PS and C VRB,PS are the unit power costs of DC / DC and DC / AC respectively; PVRB is the charging and discharging power of the vanadium battery, P SC is the supercapacitor charging and discharging power; i is the discount rate; y is the project period, years.
[0048] The average annual maintenance cost of the energy storage system C OM The calculation formula is as follows:
[0049]
[0050] In formula (8), γ is the coefficient of capacity operation and maintenance cost.
[0051] When the service life of the energy storage equipment is not enough to cover the entire cycle of the actual project, it is necessary to consider replacing the energy storage equipment. After N years, the average annual replacement cost of the energy storage system is C R The calculation formula is as follows:
[0052]
[0053] In formula (9), z is the number of times the battery is replaced in the project; m is the life cycle of the battery cell.
[0054] The average annual residual cost of the energy storage system is C RES The calculation formula is as follows:
[0055]
[0056] In formula (10): RES is the residual value rate of the energy storage device.
[0057] (b) Objective function 2: Minimize the load power failure rate of the energy storage system
[0058] When the photovoltaic independent microgrid is running, the photovoltaic power generation unit and the energy storage system jointly supply power to the load. Therefore, this embodiment focuses on the ability of the photovoltaic storage system to ensure the normal operation of the load, so this embodiment uses the load power shortage rate (loss of power supply probability, LPSP) to characterize the system power supply reliability. Its expression is as follows:
[0059]
[0060] Where: P Loads (t i )-(P PV (t i )+P hess (t i ))>0;P Loads (t i ), P PV (t i ) and P hess(t i ) are t i The power of load, photovoltaic and hybrid energy storage at the moment; N is the total number of sampling points. Obviously, f LPSP The smaller it is, the higher the power supply reliability.
[0061] In this embodiment, for the above two objective functions, the charging and discharging power constraint, SOC constraint, and power constraint of the hybrid energy storage unit are used as constraint conditions, where:
[0062] 1) The charging and discharging power constraints of the energy storage unit are as follows:
[0063]
[0064] In formula (12): P VRB (t i ) and P SC (t i ) are t i Real-time power of energy storage battery at all times.
[0065] 2) The SOC constraints of the energy storage unit are as follows:
[0066]
[0067] In formula (13): SOC VRB (t i ) and SOC SC (t i ) are VRB and SC at t i Real-time state of charge at all times.
[0068] 3) The energy storage unit power constraints are as follows:
[0069]
[0070] In formula (14): E VRB (t i )、E SC (ti) are VRB and SC at t i Real-time power status at all times.
[0071] Therefore, this embodiment finally establishes a multi-objective capacity optimization model for the hybrid energy storage system.
[0072] Step 2: Improve the NSGA-II algorithm.
[0073] This embodiment improves the elite retention strategy based on the traditional NSGA-II algorithm. By introducing a new fitness function, it is used to prioritize the retention of solutions that are closer to the Pareto frontier, thereby improving the optimization ability of the algorithm in the later stage. Adaptive crossover and mutation probabilities are also used. In the early stage, the dispersion of the population is high, and a smaller population crossover probability is used; in the later stage, the population continues to converge to the optimal Pareto frontier, and the crossover probability is increased at this time to prevent the population from falling into a local optimum.
[0074] The existing traditional NSGA-II algorithm has the problems of being prone to falling into local optimal solutions and having poor local search capabilities. In order to improve these problems of the traditional NSGA-II algorithm, this embodiment improves the NSGA-II algorithm from the following three aspects:
[0075] S1) Introducing local search operation
[0076] In order to find a better solution, a local search is performed around the point with a non-dominated sorting level of 1. The individuals after the search are compared with the previous individuals, and the individuals with better performance are retained, so as to find a better solution to ensure the uniform distribution and extensiveness of the solution.
[0077] First, sort according to objective function 1 or objective function 2. Assuming that the solution A(x1,x2,x3,x4) with a non-dominated sorting level of 1 in the solution set has an adjacent solution B(y1,y2,y3,y4), then the search range is centered on A and has a radius of x. i -y i (i=1,2,3,4) circle, x' i =x i ±(x i -y i )×rand(), generate a new solution A', perform non-dominated sorting on solutions A and A', if A' can dominate A, then A' replaces A and becomes a better solution. i ,y i and x i ' represent the elements in the solution, i.e. the rated capacity and rated power solved by the algorithm.
[0078] S2) Improved elite retention strategy
[0079] At the same frontier level, crowding distance is needed to further measure the advantages and disadvantages of individuals. Figure 2 For comparison of advantages and disadvantages. Figure 2It can be seen that the congestion of point b is greater than that of point a, so point b is better than point a and is ranked higher. However, from the analysis of the Pareto frontier, it is obvious that point a is better than point b, but because point a is at the end of the congestion ranking, the possibility of being retained in the new population is small, which is not conducive to the local optimization in the later stage of the algorithm. Therefore, a new fitness function based on logic and rules is introduced to improve the optimization ability of the algorithm in the later stage. In the new fitness function, the points are first sorted according to objective function 1 or 2. If the current point is on the left side of the line connecting the previous and next points, it is retained to ensure that it enters the new generation of population. The remaining points are still normally retained as elites according to the dominance relationship and congestion distance.
[0080] S3) Adaptive crossover and mutation probability,
[0081] The crossover probability has an important impact on population diversity and algorithm convergence. In the early stage, the dispersion of the population is high, so a smaller population crossover probability can be used; in the later stage, the population continues to converge to the optimal Pareto frontier. At this time, the crossover probability can be increased to prevent the population from falling into the local optimum. The calculation formula of the adaptive crossover operator is:
[0082]
[0083] In formula (16): P c is the crossover probability, P c ∈[P c,min ,P c,max ], P c,min and P c,max are the minimum and maximum crossover probability respectively. In this paper, P c,min =0.5, P c,max is 1.0; it is the current loop algebra; MaxIt is the maximum loop algebra.
[0084] As the algorithm progresses, the population may over-concentrate in certain areas, causing the search to fall into a local optimal solution. The mutation operation can effectively expand the search range of the population. In the later stage, by expanding the mutation probability, the algorithm's exploratory nature can be increased to avoid premature algorithm maturity. The adaptive mutation formula is:
[0085]
[0086] In formula (17): P m is the mutation probability, P m ∈[P m,min ,P m,max ], P m,min and P m,max are the extreme values of mutation probability, which are 0.1 and P in this embodiment. m,max is 0.7.
[0087] This embodiment uses the improved NSGA-Ⅱ algorithm to iteratively optimize, and approaches the real Pareto frontier under the conditions of meeting the load demand and system constraints. The independent variables of the optimization configuration problem are the rated capacity and rated power of the vanadium battery and supercapacitor, that is, x = (E VRB ,P VRB ,E SC ,P SC ). Improve the NSGA-Ⅱ algorithm optimization process as follows Figure 3 As shown, the main steps are as follows:
[0088] (2.1), initialize the population and set the maximum number of cyclic generations;
[0089] (2.2) Perform crossover and mutation on the initial population, thereby generating new individuals in the population;
[0090] (2.3) Use non-dominated sorting and crowding distance sorting to distinguish the individuals in the population after crossover and mutation in step 2, and determine the superiority and inferiority relationship between individuals;
[0091] (2.4) Introduce local search operation for individuals with dominance priority 1, compare the individuals after search with the previous individuals, retain individuals with better performance, and thus find a better solution;
[0092] (2.5) In the later stage of the algorithm, a new fitness function is introduced to prioritize solutions that are closer to the Pareto frontier and improve the optimization ability of the later stage of the algorithm. The solutions that are not confirmed and retained are still distinguished according to the non-dominated sorting and crowding distance sorting;
[0093] The currently best individuals retained constitute the best solution set, thus forming a new population;
[0094] (2.6) Determine whether the loop algebra has reached the maximum loop algebra. If not, repeat steps (2.2)-(2.5); if it has reached, output the current optimal solution set.
[0095] The following is a calculation example analysis, which uses specific data to analyze and illustrate the performance of the hybrid energy storage system capacity optimization configuration method disclosed in the embodiment.
[0096] 1. Basic data and parameter settings
[0097] This example is based on the photovoltaic output power data of an independent microgrid in a remote area for one year and the total load power data of a household. The photovoltaic installed capacity is 8kW and the data sampling interval is 5 minutes. The k-means clustering algorithm is used for clustering analysis to obtain the typical photovoltaic day and load typical day data. This example selects the scene with the highest probability for analysis. The typical day photovoltaic and load power change curves are as follows: Figure 4As shown. The basic parameters of the hybrid energy storage system are shown in Table 1. The hybrid energy storage system is used to regulate energy and balance the supply and demand of isolated microgrids. The wavelet packet decomposition method has the advantage of obtaining detailed data features. The wavelet packet decomposition method is used for power distribution to obtain low-frequency and high-frequency power components. The vanadium battery VRB compensates for the low-frequency part, and the supercapacitor SC compensates for the high-frequency part. Table 1 is shown below:
[0098] Table 1 Basic parameters of hybrid energy storage system
[0099]
[0100] Compared with the limitation that wavelet decomposition can only process low-frequency signals, wavelet packet decomposition can further decompose high-frequency signals with higher resolution, and can more accurately distribute and optimize energy. Figure 5 shown.
[0101] 2. Analysis and explanation
[0102] The experimental platform of this example uses MATLAB / Simulink, and the standard NSGA-Ⅱ and improved NSGA-Ⅱ are used to solve the multi-objective optimization model of the hybrid energy storage system to obtain the optimal Pareto solution set. The population size is set to 50 and the maximum number of iterations is 200. The optimization results of the two algorithms are shown in Table 2. Table 2 is as follows:
[0103] Table 2 Optimization results of two algorithms
[0104]
[0105] As can be seen from Table 2, under the condition of a load power shortage rate of 2%, the improved NSGA-II algorithm in the embodiment obtains an average annual investment cost of 17210.777 yuan for the energy storage system, while the cost obtained by the traditional NSGA-II algorithm is 17314.047 yuan. It can be seen that the improved NSGA-II algorithm in the embodiment reduces the cost and effectively reduces the number of iterations of the algorithm, speeding up the optimization speed. Therefore, the improved NSGA-II algorithm has stronger optimization effect and optimization ability.
[0106] The distribution of the solutions in the Pareto solution set obtained by the algorithm in the embodiment is as follows: Figure 6 As shown. Figure 6It can be seen that the uniformity and extensiveness of the Pareto solution sets of the two algorithms in space are not much different. However, the solution set of the improved NSGA-II algorithm in the embodiment is obviously below the curve of the conventional NSGA-II algorithm, so the improved algorithm is better than the traditional algorithm in convergence. For the same load power failure rate, the improved NSGA-II algorithm in the embodiment can achieve lower costs than the conventional NSGA-II algorithm.
[0107] As shown in Table 2, the life cycle cost obtained by the traditional NSGA-II algorithm is 346,281 yuan, while the life cycle cost obtained by the improved NSGA-II algorithm in the embodiment is 344,215 yuan. The cost of the improved algorithm in the embodiment is lower than that of the traditional algorithm, and the IGD value is also smaller, which proves that the convergence of the optimal solution set obtained by the improved algorithm is better and the overall optimization result is more reasonable. By analyzing the Pareto frontier and weighing the system economy and volatility, users can choose different configuration schemes according to their needs.
[0108] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. The embodiments described in the present invention are merely descriptions of the preferred embodiments of the present invention, and do not limit the concept and scope of the present invention. The various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction, and such combinations should also be regarded as the contents disclosed in the present disclosure as long as they do not violate the concept of the present invention. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.
[0109] The present invention is not limited to the specific details of the above-mentioned embodiments. Within the technical concept of the present invention and without departing from the design concept of the present invention, various modifications and improvements made to the technical solution of the present invention by technical personnel in this field should fall within the protection scope of the present invention. The technical contents for which protection is sought in the present invention have been fully recorded in the claims.
Claims
1. A hybrid energy storage system capacity optimization configuration method based on an improved NSGA-Ⅱ algorithm, wherein the hybrid energy storage system is composed of a vanadium battery and a supercapacitor, characterized in that: The following steps are involved: Step 1: Capacity optimization model construction: Taking the annual average investment cost of the hybrid energy storage system and the load power shortage rate as the minimum as the goal, establish the objective function and establish constraints, thereby obtaining a multi-objective capacity optimization model for the hybrid energy storage system; Step 2: Use the improved NSGA-II algorithm to solve the multi-objective capacity optimization model of the hybrid energy storage system established in step 1 to obtain the optimal solution set of the rated capacity and rated power of the vanadium battery and supercapacitor. The improved NSGA-II algorithm process is as follows: (2.1), initialize the population and set the maximum number of cyclic generations; (2.2) Perform crossover and mutation on the initial population, thereby generating new individuals in the population; (2.3) Use non-dominated sorting and crowding distance sorting successively to distinguish the individuals in the population after crossover and mutation in step 2, and determine the superiority and inferiority relationship between individuals; (2.4) Introduce local search operation for individuals with dominance priority 1, compare the individuals after search with the previous individuals, retain individuals with better performance, and thus find a better solution; (2.5) A new fitness function is introduced to prioritize solutions that are closer to the Pareto frontier, improving the algorithm's optimization ability in the later stages. Solutions that have not been confirmed to be retained are still ranked based on non-dominated sorting and crowding distance sorting. The currently best individuals retained constitute the best solution set, thus forming a new population; (2.6) Determine whether the loop algebra has reached the maximum loop algebra. If not, repeat steps (2.2)-(2.5); if it has, output the current optimal solution set.
2. The hybrid energy storage system capacity optimization configuration method according to claim 1, characterized in that: In step 1, the charging and discharging power constraint, SOC constraint, and power constraint of the hybrid energy storage unit are used as constraint conditions.
3. The hybrid energy storage system capacity optimization configuration method according to claim 1, characterized in that: In step 2, a local search operation is introduced to perform a local search around the point with a non-dominated sorting level of 1. The individuals after the search are compared with the previous individuals, and the individuals with better performance are retained, so as to find a better solution to ensure the uniform distribution and extensiveness of the solution.
Citation Information
Cited By
All-vanadium redox flow battery system operation optimization method and system
CN119361767A
Calculation method of regulation and control system for hydrogen production through combination of water-wind-light complementary power generation and green power consumption
CN119995038A
Improved NSGA-II-based silicon micro-resonant accelerometer multi-objective optimization method
CN120470642A
Multi-objective optimization method for silicon micro resonant accelerometer based on improved NSGA-II
CN120470642B
