EMPF PROJECT TO BRING BATTERY PROGNOSTICS TECHNOLOGY TO THE WARFIGHTER

The ability to quickly evaluate the remaining charge in a fielded battery and to reliably identify batteries that have reduced capacity with the potential for failure are the hallmarks of battery prognostics applications. Modern war fighters depend heavily on the availability of reliable battery power for their communications, night vision, fire control, and perimeter security systems. Any uncertainty concerning the present energy capacity or the state of charge in a battery adds to the risk of any mission. The uncertainty of a battery's state-of-charge (SOC), its state-of-health (SOH), or its state-of-life (SOL) can lead to premature disposal by the war fighter causing a cost and a supply logistics nightmare. This same uncertainty prevents the transition from primary batteries to the less robust but less costly rechargeable batteries due to the inability to assess future power source degradation. The Navy MANTECH challenge is to greatly reduce this uncertainty in the field by transforming proven battery prognostics technology from the laboratory into an affordable, reliable, and manufacturable package acceptable for wide-scale deployment in DOD applications.

During Operation Iraqi Freedom, the usage rate of the main primary battery (BA-5590) reached 6000 batteries per day, according to the U.S. Central Command during actions in April of 2003 [1]. Demand for primary batteries exceeded the supply and taxed the logistics system to the point that a joint staff battery power desk was established to deal solely with this critical item. The battery shortage was in fact among the top briefing points made to Gen. Tommy Franks at his daily briefings and was even elevated as a briefing item to the Commander-in-Chief [1]. At an acquisition and disposal cost of over $109 for each battery, both the availability and affordability of this critical item demand an alternative solution. Battery prognostics can reduce acquisition costs and increase critical supply levels by making the alternate rechargeable battery more attractive and prevent the unnecessary disposal of primary batteries already in use.

Making the transition from primary batteries to rechargeable batteries offers the opportunity to realize significant financial savings in addition to relieving a huge burden on the logistics system of supplying a huge number of primary batteries to the field. War fighters in the field need confidence that the rechargeable batteries are as reliable as their disposable equivalents [2]. Despite the financial attractiveness of using rechargeable batteries, primary batteries are preferred because they are a known entity and their respective energy density is higher than a rechargeable. The uncertainty in battery life coupled with the extra combat load, carrying a charger, and the logistics support needed to support the use of rechargeable batteries are factors that have made rechargeables unattractive in the past [2]. Technology advances are needed to reduce the maintenance levels associated with using rechargeable batteries. The ability to know the battery SOC and the SOH at any given time is a cornerstone issue. The benefits of rechargeable batteries will not be realized until these obstacles are addressed.

PROJECT OBJECTIVES:
The EMPF staff will utilize their resident electronics manufacturing expertise at the Center to develop the electronics packaging and manufacturing processes needed to manufacture a high density, configurable, and affordable battery prognostics tester using commercial off-the-shelf components that will:

(1) Estimate the remaining charge in a battery that will enable the secondary battery user to know how much energy is available;

(2) Identify batteries that have reduced capacity, impending failures, or poor health;

(3) Meet the harsh environmental military requirements; and

(4) Mitigate component obsolescence.

To reach these goals, a high degree of accuracy must be attained in the measurement of the applicable electrical parameters, which need to be determined. This accuracy will increase the user's confidence in a battery's health and capacity and thus prolong the usage of each battery. This technology has wide-scale application across all of DOD and may also have commercial usage in police departments, fire departments, and wireless communications applications.

TECHNICAL APPROACH:
The measurement methodology has been developed [3] to determine the SOC, SOH, and the SOL of primary and rechargeable batteries ranging from small primary single cells to large banks of rechargeable batteries [4]. Unfortunately, these parameters are not directly measurable. To determine theses parameters, several techniques must be utilized. Penn State's Applied Research Laboratory (PSU/ARL) in conjunction with the EMPF has developed the electronics algorithms and an impedance measuring technique and has related the measurements to the specific battery chemistry. The algorithms are based on electric circuit models of the battery that include Randles circuit behavior with a Warburg (diffusion) element inserted into the model (see reference 6 for a more detailed explanation of impedance measurements and Warburg impedance).

The flow chart used to make the predictions is shown in Figure 1-1. The impedance is measured at several frequencies and is compared against the knowledge base which has been established. Theses parameters are then placed in the ARMA (aggressive moving average) algorithms which are used to determine the SOC of the battery.
Fuzzy logic is also used to find the SOC and SOH of the battery. Fuzzy Logic is a departure from classical two-valued sets and logic that uses "soft" linguistic (e.g. large, hot, tall) system variables and a continuous range of truth values in the interval [0,1], rather than strict binary (true or false) decisions and assignments. The SOC and the SOH of the battery can be determined by using the fuzzy logic combined with the use of an expert database system.

An artificial neural network (ANN) is an information-processing paradigm inspired by the way the densely interconnected, parallel structure of the mammalian brain processes information. This application uses the ANN after it has been trained to respond in the proper manner. This response is determined after an expert database is installed.

Finally, all three estimates are input into a decision fusion algorithm to increase the confidence in the assessment of the SOC, SOH, and SOL of the battery. This decision fusion also has access to the raw data which was used to make the previous estimates and the results are displayed to the user.

PRELIMINARY RESULTS:
The technology has been tested successfully on several types of batteries. The investigations on a wide array of battery chemistries and capacities on both primary and rechargeable batteries yielded very impressive results. The accuracy of the prognostics has been verified to be within 5.5% of the actual value in all cases. The hardware is currently in prototype status and is actively being tested for reliability and effectiveness in simulated field conditions.

DEFINITIONS:
1) State of Charge (SOC) is the ratio of remaining battery capacity to rated battery capacity. SOC is used to estimate Battery Power Time (BPT), which is the remaining electrical power supply time based upon the power demands of the equipment in use. SOC can also be used to estimate the time required to fully charge a battery using a particular charging station.

2) State of Health (SOH) is the physical condition of a battery. SOH is used to estimate losses in rated capacity, as well as predicting impending failures such as shorts or severe corrosion.

3) State of Life (SOL) is the remaining life of the battery. SOL is used to estimate the number of remaining charge/discharge cycles for a battery.

REFERENCES:
[1] Geoff S. Fein, "Battery Supplies Ran Dangerously Low in Iraq,” National Defense Magazine, September, 2003.

[2] Penn State ARL staff, "Using Battery Prognostics maximize the efficient use of rechargeable batteries by special operations forces.”

[3] J. D. Kozlowski, U.S. Patent No. 6,307,378 B1, "Method and Apparatus for Measurement of Electrochemical Cell and Battery Impedances," granted October, 2001.

[4] J. D. Kozlowski, T. Cawley, and C. S. Byington, "Model-Based Predictive Diagnostics for Primary and Secondary Batteries: Phase II Report," Technical Memorandum, File No. 01-164, Applied Research Laboratory, The Pennsylvania State University, January, 2001.

[5] J. D. Kozlowski, "Electrochemical Cell Prognostics using Online Impedance Measurements and Model-Based Data Fusion Techniques," Proceedings of the 2003 IEEE Aerospace Conference, March, 2003.

[6] Bard & Faulkner, "Electrochemical Methods: Fundamentals and Applications, Second Edition.”