Oct 09, 2019 Pageview：11
Refers to the accumulation of surplus capacity, battery uses energy storage device such as after a certain period of time after the rest of the size. General battery modeling method can be divided into two categories: one is physical modeling method; Another kinds are the system identification and parameter estimation modeling method.
Battery in specified conditions (including the intensity of discharge, the discharge current and discharge termination voltage) sent out how much electricity or discharge duration called battery capacity, unit or a. a. h min.Car battery is a reversible dc power supply, which can convert chemical energy to electricity, also can convert electrical energy into chemical energy, it with the generator in parallel.
Battery don't have to for a long time, it will slowly discharge by oneself, to scrap.At regular intervals, therefore, should start a car, for battery charging.Another way is pulled down the two electrodes on the battery, unplug from the electrode column that should pay attention to both positive and negative electrode thread, to remove the negative line first, or remove negative and chassis connections.Then pull with the positive sign (+), on the other side of the battery service life to a certain extent, to a certain period has to be replaced.When replacement should follow the same order, but in the up electrode wire, order, by contrast, take the positive first, and then connect the cathode.
The residual capacity refers to the battery energy storage device such as after a certain time of use of the rest of the size.
Battery because of its small volume, explosion-proof, voltage stability, no pollution, light weight, high discharge performance, less maintenance, and low price, is widely used in post and telecommunications, electric power, transportation, aerospace, emergency lighting, military communication, and many other fields.Battery has already become one of the key components of the system, it is directly related to the safe and reliable operation of the reliable operation of the whole equipment.But in use process, because of the residual capacity can predict, light caused by accident, heavy tragic outcome.Therefore, it is necessary to establish an efficient battery management system, accurate and reliable to predict the residual capacity of battery becomes battery management system is the most basic and most to task.
Widely used at home and abroad charged state of SOC (stateofcharge) to indicate the residual capacity of battery.Is directly reflect the sustainable supply of battery capacity and an important parameter in the health.Because the battery has a different type, purpose, and the external environment, the influence of many factors, so the prediction methods, various battery model used are also different.General battery modeling method can be divided into two categories: one is physical modeling method;Another kind is the system identification and parameter estimation modeling method.
Physical modeling method to predict
(1) discharge experiment method.Discharge test is generally accepted that the most reliable estimation method.According to a certain discharge ratio of current continuous discharge the battery to zero, the product of the discharge current and time is the surplus capacity.Discharge test method is mainly used for laboratory calculate the estimation accuracy of the battery charging efficiency, inspection or used for battery maintenance, applicable to all cells.However, this method has two obvious shortcomings need a lot of time and human cells had to interrupt the ongoing work, unable to real-time online prediction.For static backup battery can use, but for the important occasion, with this method is a certain risk, because during the discharge, system running under no battery backup, once the main power supply problems or mains interruption, the whole system will be paralyzed, causing immeasurable loss.
(2) Ann method.Ann method when the essence is the battery as a black box, that flow into the battery power have certain proportion relationship with out of battery power, without considering the battery electrical characteristics of the structure of the internal and external, so this kind of method applies to all kinds of batteries.At the same time as you can see, in the application of Ann method in the initial value problems require calibration need to precise calculation of charge and discharge efficiency need to accurately measure the current, current measurement, will cause calculation error, will exist for a long time integral accumulated error in high temperature condition and current error volatile situation.Therefore, in actual applications using Ann method, generally according to the using environment and conditions to consider to charge and discharge rate, temperature and battery aging and self-discharge rate and other factors to compensate.
(3) the density method.Density method is mainly used in lead-acid battery.Due to get higher in the process of charging the electrolyte density, gradually reduce the discharge process, and the density of battery capacity and a certain linear relationship, therefore, by measuring the density of the electrolyte can predict the size of the.Due to the need for measuring the electrolyte density method, mainly used in open type lead-acid battery, if can develop higher accuracy of the sensor, the density of a capacity for extremely important occasions, it can be implanted in the production of the sealed battery.
(4) the open circuit voltage method.Open circuit voltage is refers to the voltage of the battery in open circuit state, close to the battery electromotive force on the value.Open circuit voltage method is on the basis of the residual capacity of battery have certain linear proportional relationship with open circuit voltage and set up, by measuring the open circuit voltage can be directly by the size of the residual capacity.Its advantage is not dependent on battery size, size and speed of discharge, only to open circuit voltage as the test parameters, is relatively simple.
System identification and parameter estimation model method to predict
(1) neural network.Because the battery is a complex nonlinear system, the charging and discharging process is difficult to establish accurate mathematical model.And neural networks with distributed parallel processing, nonlinear mapping and adaptive learning features, can better reflect the basic characteristic of nonlinear, can give the corresponding output under external excitation, so can to a certain extent dynamic characteristics simulation battery, estimate the SOC.
To estimate the battery USES mostly typical three layer artificial neural network.General direct collection and discharge current, voltage and temperature of the battery or the variable current combination measuring method, determine the electromotive force and internal resistance as the input of the neural network model, and SOC as output.The input and output layer neurons as the linear function commonly number of hidden layer nodes depends on the complexity of the problem and analysis accuracy, according to the network convergence speed in the training and the training is completed to determine the error.Artificial neural network is suitable for all kinds of battery, but the error of this method is highly affected by the training data and the training methods, and the noise exist in the actual use affect the study and application of the network.
(2) kalman filter method.The core idea of kalman filtering theory, it is for the state of the power system to make optimal in the sense of minimum variance estimation, it applies to a linear system can also be applied to nonlinear system.
When using kalman filter method to estimate, the first thing to establish suitable for kalman filter to estimate the battery model, and the model must have two aspects: 1) can well reflect dynamic characteristics of battery, order number can't too high at the same time, in order to reduce the computational complexity, the processor is advantageous for the project implementation;2) the model must be able to accurately reflect the relationship between the voltage of the battery electromotive force and, so that the closed loop are estimated to have higher precision.
Faraday constant = C / 96485 F molLiCoO2 molar mass M = 97.8698 g/mol, if lithium ion all embedded, its theoretical capacity = F/M = 985.8535 C/g = 273.848 mAh/g after conversion unit this is just a theory of capacity, in order to ensure its reversible structure, actual lithium ion embedded coefficient of only 0.5, the production process of LiCoO2 as cathode lithium ion batteries theoretical capacity of 274 mah/g, how to calculate?Please advise about is around 140.
The page contains the contents of the machine translation.
Leave a message
We’ll get back to you soon