A-Plus Homework Help
For College Students

# State of health in batteries

Document Type:Essay

Subject Area:Technology

Document 1

SOH indicates the measurement because it measures on the time that a battery can hold on charge, showing the time it has been used in what is called “lifetime energy throughput” as well as indicating the time that is left for the battery usage. The battery energy can be compared to the odometer through the use of automotive analogy, and it shows the number of miles that the car has travelled from the time it was new (Chaoui, et al. Due to the irreversible physical and chemical changes that take place in a battery, with its continuous usage, the battery performance decreases as the time goes on up to a point where the battery cannot be used for any longer. At this point, it is said to be dead, to imply that there is no a further usage of the battery because it has already spoiled.

SOH cannot be defined absolutely when it is compared with the State of Charge (SOC). How SOH is determined Cell conductance or impedance or any other parameter that changes with age can be used in indicating the cell SOH. The other changes that might have occurred and which are essential to the user will be shown by the changes in the parameters. Increased temperatures or loss of the rated capacity or internal modifications such as corrosion during the operation are some of the changes that could lead to the changes in the battery performance. The measurement system should hold the set of standard conditions or the initial conditions because SOH is calculated about the state of a new battery. If cell impedance is the monitored parameter in this case, keeping a record of the starting impedance in the new cell is important in this case.

The help of a microprocessor for the complex measurement is required to deliver results. Improving the accuracy of the results can be done by conducting fuzzy logic which combines the experience with the sizes. The method discussed does not require cell modifications because it uses an external measuring device. Application in the industry A hybrid vehicle (HEV) requires a high voltage battery. To maintain the car to be in good condition, the battery’s state of health is an essential factor to look into. Battery modelling Mathematical modelling Mathematical modelling plays a vital role in the use and the design of batteries. This battery modelling field is branched into two parts of study; • Battery output estimation • Battery design a) Estimating the output of the battery Evaluation of the battery performance under specific of the user conditions is a significant problem, especially when dealing with an already constructed battery.

The issue can only be addressed by the use of different models of cells to test the results. The model of statistics to the neural nets and then to the complicated, physics-based models are the range of the approaches that are used to check the results. When the models are based on the test data, this becomes a problem, particularly when the testing becomes impractical. For this to happen, the battery should be ready to deliver or accept some energy (Zou, et al. The vehicle needs to be at some intermediate SOC. Therefore, the disease and the control of the SOC are significant. SOC is estimated by fixing both the current data and the measured voltage into a model, and this ensures that the battery does not lose power when the vehicle is at rest.

The estimate done can as well be averaged with the obtained by averaging the SOC estimate achieved by the integration of current (Zou, et al. The users can know the available runtime of the computer by estimating the SOC. This is achieved by the use of two parameters Ro and Vo. When the battery is at rest, Vo can be determined, and it is the open circuit voltage of a battery and the lithium batteries, it is the accurate indicator of SOC. Ro, on the other hand, is obtained by the real-time measurement of the impedance. Ro is the impedance of the battery. Higher growth of the film implies a smaller proportion. In other words, as the film length increases, the thickness of the film reduces. The comparison follows Arrhenius behavior.

In summary of the mathematical modelling is that this approach is used for the estimation of battery performance. The development of battery is significantly impacted by the mathematics modelling. ­­­­­­The resistance of the cell is very high at a lower temperature, and therefore the driving range drops dramatically. Lower temperatures also have some advantages. For example, it is suitable for parking the vehicle. In this case, both aging and the self-discharge of the car will be slowed down. On the other hand, temperature above the comfort range has its effect in that the battery will be discharged more and the battery life can be shortened because of faster aging and self-discharges of the cell. By this way, one is assured of an immediate replacement of the battery if the one is using develops a problem.

The cells also give a competition to the manufacturing firms, and this result in reduced costs of the batteries, which is an advantage to the owner of the equipment (Gholizadeh & Farzad, 1344). Disadvantages The development of various kinds of batteries results into a competition in the manufacturing fields. It follows that there will be reduced prices for the cells as the firms will be trying to outdo each other. Consequently, there is a likelihood of the production of low standard batteries which will last for a shorter period. Support Vector Machines for Machine Learning This is one of the most popular algorithms for machine learning. The method has a little tuning, and it has been famous from the time they were developed, that is in the year 1990s. The concept of the algorithm is simple, and this makes it exciting to handle.

Maximal-Margin Classifier. This explains how the SVM practically works. To be able to separate the classes of a line, they should be made relaxed. In this case, the separating line is violated. To give the margin a wiggle room in each dimension, a set of coefficients is introduced. In this case, we have many parameters, and this increases the model complexity (Meng, et al. Support Vector Machines (Kernels) The kernel implements the algorithm of the SVM. ANNs has interconnected layers. The input neurons are contained in the first layer, and they send data to the second layer which then transmits the output neurons into the last segment. Choosing the allowed models which have several associated algorithms is involved in the training of an artificial neural network. Where the algorithm best fits The Artificial Neural Network tries to over fit the relationship and this is the reason why the algorithm is rarely used for the predictive modelling.