A smart charging optimization method based on electric vehicles

Abstract

The present invention discloses an intelligent charging optimization method based on electric vehicles, including: step 1, dividing the full life interval of lithium-ion batteries, and establishing a lithium-ion battery capacity and health status model; step 2, taking the battery charging time and aging loss as optimization targets, and optimizing charging based on the estimated battery health status through a particle swarm algorithm; step 3, obtaining an optimized charging current distribution under the full life cycle based on optimal current distribution. The intelligent charging control method based on electric vehicles provided by the present invention divides the full life interval of lithium-ion batteries, and optimizes charging based on a particle swarm optimization algorithm, a multi-stage constant current charging process that balances battery aging loss and charging time, shortens charging time, reduces charging temperature, and improves battery life.

Description

Technical Field

[0001] The invention relates to an intelligent charging optimization method based on an electric vehicle, and belongs to the field of electric vehicle charging. Background Art

[0002] Power batteries are an important system for storing energy in electric vehicles and are very important for the development of automobiles. Compared with other batteries, lithium-ion batteries have the advantages of long service life, low self-discharge rate, high specific energy, no memory effect, and environmental protection. They are the main development trend in the field of energy storage in the future. In order to improve the practicality of battery-powered equipment and user satisfaction, it is crucial to choose a reasonable, fast and safe charging strategy.

[0003] At present, the most widely used charging strategy is constant current and constant voltage charging, but this charging method consumes too much time in the constant voltage charging stage, and will cause adverse effects such as energy loss, large temperature rise, and short life of lithium-ion batteries. Pulse charging can significantly reduce the planned reaction inside the battery during the charging process and can effectively shorten the charging time, but this charging method will accelerate the aging of the battery and the control strategy is too complicated. The multi-level constant current charging strategy is considered to be an improved engineering charging method that can effectively shorten the charging time, improve the charging efficiency, and extend the cycle life of the battery. At present, there is no strict and accurate basis for the definition of the number of constant current segments in the multi-level constant current charging strategy, which is generally 3 to 5 orders, and the optimization objectives are different. Some people have proposed to use a group of algorithms, genetic algorithms, etc. to achieve multi-objective optimization of battery charging capacity and charging time, so as to calculate the optimized charging current at each stage of multi-level constant current charging.

[0004] However, these methods only study the external characteristics of lithium-ion batteries, without fully considering the impact of the physical and chemical reaction mechanisms inside the battery on the charging process, and have certain limitations. Summary of the invention

[0005] The present invention designs and develops an intelligent charging optimization method based on electric vehicles, performs charging optimization based on a particle swarm optimization algorithm, and takes into account a multi-stage constant current charging process with balanced consideration of battery aging loss and charging time, thereby shortening the charging time, reducing the charging temperature, and increasing the battery life.

[0006] The technical solution provided by the present invention is:

[0007] An intelligent charging optimization method based on an electric vehicle, comprising:

[0008] Step 1: Divide the full life cycle of lithium-ion batteries and establish a lithium-ion battery capacity and health status model;

[0009] Step 2: Taking the battery charging time and aging loss as optimization targets, the particle swarm algorithm is used to optimize charging based on the estimated battery health status;

[0010] Step 3: Obtain the optimized charging current distribution over the entire life cycle based on the optimal current distribution.

[0011] Preferably, in step 1, the health state of lithium ions is set to SOH, and the calculation formula is:

[0012]

[0013] In the formula, C N is the factory rated capacity of the battery, C R is the residual capacitance, C loss To consume the capacitor.

[0014] Preferably, in step 2, the optimization objective function is:

[0015]

[0016] In the formula, α and β are weight coefficients, and the set value is 0.5, t now is the charging end time, t max is the maximum charging time, t min is the initial charging time, L now is the aging loss of the battery at the end of charging, L max is the maximum aging loss of the battery, L min is the minimum aging loss of the battery.

[0017] Preferably, the step 2 comprises:

[0018] Step 1: Initialize the population, define the current combination of each level as a particle, define the order of the charging current as the dimension of the particle, randomly generate 20 groups of 5-dimensional current combinations, and randomly give the initial speed of the particle;

[0019] Step 2: Calculate the model parameters of the lithium-ion battery based on the generated current combination. When the battery terminal voltage is greater than the upper cut-off voltage of 4.2V, the charging current is transferred to the next level. When the charging capacitance reaches the set capacitance, the charging is completed and the output parameter value t now and L now ;

[0020] Step 3: According to the optimization objective function, calculate the fitness value of each particle and compare the fitness value of the particle with the local optimal value P best For comparison:

[0021] When the fitness value of a particle is less than the local optimal value, the fitness value of the particle is the local optimal value;

[0022] Compare the local optimal values ​​of all particles to obtain the global optimal value g best ;

[0023] Step 4: Update and adjust the speed and position of the particles;

[0024] Step 5: Repeat steps 2 to 4. When the global optimal fitness value is no longer updated, the global optimal solution is obtained and the optimal charging current combination is output.

[0025] Preferably, in step 4, the update equation is:

[0026]

[0027] Where i is the i-th particle, j is the dimension of the particle, k is the number of iterations, ω is the inertia weight, rand1 and rand2 are random numbers between 0 and 1, c1 is the particle individual, and c2 is the learning factor of the population;

[0028] I ij (k+1)=I ij (k)+v ij (k+1);

[0029]

[0030] In the formula, k max is the maximum number of iterations, ω max is the maximum inertia weight, ω min is the minimum inertia weight.

[0031] Preferably, the k max The value is 200, the ω max The value is 1, ω min The value is 0.2.

[0032] The beneficial effects described in the present invention are as follows: the intelligent charging control method based on electric vehicles provided by the present invention divides the full life interval of the lithium-ion battery, and performs charging optimization based on the particle swarm optimization algorithm, a multi-stage constant current charging process that balances the battery aging loss and charging time, shortens the charging time, reduces the charging temperature, and increases the battery life. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is the fitness curve of the particle swarm optimization algorithm described in the present invention.

[0034] Figure 2 The voltage and current curves of the lithium-ion battery optimization charging method described in the present invention.

[0035] Figure 3This is the relationship between the surface temperature and the charging time of the example described in the present invention.

[0036] Figure 4 The relationship between SOH and charging time of lithium-ion batteries. DETAILED DESCRIPTION

[0037] The present invention is further described in detail below in conjunction with the accompanying drawings so that those skilled in the art can implement the invention with reference to the description.

[0038] like Figure 1 As shown, the present invention provides an intelligent charging optimization method based on electric vehicles, which performs charging optimization based on a particle swarm optimization algorithm, a multi-stage constant current charging process that balances battery aging loss and charging time, shortens charging time, reduces charging temperature, and improves battery life, including:

[0039] Step 1: Divide the full life cycle of lithium-ion batteries and establish a lithium-ion battery capacity and health status model;

[0040] Step 2: Taking the battery charging time and aging loss as optimization targets, the particle swarm algorithm is used to optimize charging based on the estimated battery health status;

[0041] Step 3: Obtain the optimized charging current distribution over the entire life cycle based on the optimal current distribution.

[0042] Define the state of health (SOH) of lithium-ion batteries and establish the relationship between battery capacity and state of health:

[0043]

[0044] In the formula, C N is the factory rated capacity of the battery, C R is the residual capacitance, C loss To consume the capacitor.

[0045] The lithium-ion battery charging optimization process includes:

[0046] Step 1: Initialize the population, define the current combination of each level as a particle, define the order of the charging current as the dimension of the particle, and randomly generate 20 groups of 5-dimensional current combinations that satisfy the following formula:

[0047] 0≤I m+1 ≤I m ≤4,(m=1,2,3,4,5)

[0048] At the same time, the initial velocity of the particles is given randomly, and each particle is evenly distributed within the definition domain and takes random values.

[0049] Step 2: Simulate the charging mode of lithium-ion batteries. Calculate the model parameters of lithium-ion batteries according to the generated current combination. When the battery terminal voltage is greater than the upper cut-off voltage of 4.2V, the charging current is transferred to the next level. When the charging capacitance reaches the set capacitance, the charging time is recorded. Charging end t now , output parameter value t now and L now ;

[0050] Taking the battery charging time and aging loss as the optimization target, the optimization objective function is:

[0051]

[0052] In the formula, α and β are weight coefficients, and the set value is 0.5, t now is the charging end time, t max is the maximum charging time, t min is the initial charging time, L now is the aging loss of the battery at the end of charging, L max is the maximum aging loss of the battery, L min is the minimum aging loss of the battery.

[0053] The optimization objectives are limited as shown in Table 1:

[0054] Table 1 Setting range of limiting conditions during charging

[0055]

[0056] Step 3: Calculate the fitness of each particle. According to the objective function, calculate the fitness value of each particle and compare the fitness value of the particle with the local optimal value P. best For comparison:

[0057] When the fitness value of a particle is less than the local optimal value, the fitness value of the particle is the local optimal value;

[0058] Compare the local optimal values ​​of all particles to obtain the global optimal value g best ;

[0059] Step 4: Update and adjust the particle's velocity and position. The update equation is:

[0060]

[0061] Where i is the i-th particle, j is the dimension of the particle, k is the number of iterations, ω is the inertia weight, rand1 and rand2 are random numbers between 0 and 1, c1 is the particle individual, c2 is the learning factor of the population, and the value range of c1 and c2 is 1 to 4;

[0062] Iij (k+1)=I ij (k)+v ij (k+1);

[0063]

[0064] In the formula, k max is the maximum number of iterations, the value is 200, ω max is the maximum inertia weight, which is 1, ω min is the minimum inertia weight, and its value is 0.2.

[0065] Step 5: Repeat steps 2 to 4. When the global optimal fitness value is no longer updated, the global optimal solution is obtained and the optimal charging current combination is output.

[0066] Simulation experiment

[0067] In the present invention, as a preferred embodiment, a 3.6V / 2000mAh INR 18650-20R lithium-ion battery is selected as the research object, and Matlab software is used to perform simulation and experimental verification under the Windows 10 operating system.

[0068] like Figure 1 As shown, after 113 iterations, the global optimal fitness value g best No more updates are continued, the particle swarm optimization algorithm iteration reaches convergence, and the optimization is completed. The optimization results show that the particle swarm optimization algorithm can obtain the global optimal solution with fast convergence performance.

[0069] like Figure 2 As shown, through multiple algorithm optimization tests, the best charging current combination of the lithium-ion battery optimized charging mode is [2.954A, 1.686A, 1.014A, 0.6106A, 0.3879A], thereby obtaining the optimized charging voltage and current curve, and the optimized lithium-ion battery charging time is 3937s.

[0070] like Figure 3 As shown in the figure, it is a curve diagram of the relationship between the battery surface temperature and the charging time during the charging process. When the first stage of charging is completed, as the charging current gradually decreases, the heat generated by the battery itself will also decrease; when the heat generated inside the battery is less than the heat dissipated by the external air flow, the temperature curve has an inflection point, and the temperature of the battery surface begins to drop. The results show that during the entire charging process, the surface temperature of the lithium-ion battery is within the specified range.

[0071] During the entire charging process, the cycle life of the lithium-ion battery changes with the charging time as shown below: Figure 4As shown, at the end of charging, the SOH of the battery is 99.150 2%, that is, when charging is completed, the aging loss of the battery is only 0.849 8%.

[0072] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the implementation modes, and they can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.

Claims

1. A smart charging optimization method based on electric vehicles, characterized in that: include: Step 1: Divide the full life cycle of lithium-ion batteries and establish a lithium-ion battery capacity and health status model; Step 2: Taking the battery charging time and aging loss as optimization targets, the particle swarm algorithm is used to optimize charging based on the estimated battery health status; The optimization objective function is: ； In the formula, and is the weight coefficient, set to 0.5, is the charging end time, is the maximum charging time, is the initial charging time, The aging loss of the battery at the end of charging. is the maximum aging loss of the battery, is the minimum aging loss of the battery; Specifically include: Step 1: Initialize the population, define the current combination of each level as a particle, define the order of the charging current as the dimension of the particle, randomly generate 20 groups of 5-dimensional current combinations, and randomly give the initial speed of the particle; Step 2: Calculate the model parameters of the lithium-ion battery based on the generated current combination. When the battery terminal voltage is greater than the upper cut-off voltage of 4.2V, the charging current is transferred to the next level; when the charging capacitance reaches the set capacitance, the charging is completed and the parameter value is output. and ; Step 3: According to the optimization objective function, calculate the fitness value of each particle and compare the fitness value of the particle with the local optimal value. For comparison: When the fitness value of a particle is less than the local optimal value, the fitness value of the particle is the local optimal value; Compare the local optimal values ​​of all particles to obtain the global optimal value ; Step 4: Update and adjust the speed and position of the particles; Step 5: Repeat steps 2 to 4. When the global optimal fitness value is no longer updated, the global optimal solution is obtained and the optimal charging current combination is output; In step 4, the update equation is: ； In the formula, For the Particles, is the dimension of the particle, is the number of iterations, is the inertia weight, and is a random number between 0 and 1. For individual particles, is the learning factor of the population; ； ； In the formula, is the maximum number of iterations, is the maximum inertia weight, is the minimum inertia weight; Step 3: Obtain the optimized charging current distribution over the entire life cycle based on the optimal current distribution.

2. The intelligent charging optimization method based on electric vehicles according to claim 1 is characterized in that: In step 1, the health status of lithium ions is set to , the calculation formula is: ； In the formula, is the factory rated capacity of the battery, is the residual capacitance, To consume the capacitor.

3. The intelligent charging optimization method based on electric vehicles according to claim 2 is characterized in that: Said The value is 200. The value is 1, The value is 0.2.

Images