Intelligent Battery Management System and Method for Optimizing Battery Set to the Best Performance

Abstract

This innovation is related to intelligent battery management system and method for optimizing the battery set to the best performance. The management system is supported by a cutting-edge method and corresponding embodiment, which can automatically re-combine batteries in a battery set in parallel connection or in series connection or in mixed series-parallel connection, or mixed parallel-series connection, and makes each individual cell in a battery being accessible, which means each individual cell can be monitored (i.e. its parameters can be measured) and charged or discharged. Therefore, with this system, batteries can be charged or discharged to their best performance.

Description

FIELD OF INVENTION

An intelligent battery management system and method, cutting-in at a new angle, optimizes rechargeable battery set to the best performance, which can overcome the backwards of all current management systems and methods, and reduce the manufacturing cost at better performances. The management system is supported by a cutting-edge method and corresponding embodiment, which can automatically re-combine batteries in a battery set in parallel connection or in series connection or in mixed series-parallel connection, or mixed parallel-series connection, each individual cell in a battery set can accessed, which means each individual cell can be monitored (i.e. its parameters can be measured) and charged or discharged. With this system, batteries can be charged or discharged to their best performance.

BACKGROUND OF INVENTION

Battery management system (BMS) is an important part of the Electric Vehicle (EV). It protects the battery system from damage, predicts and increases battery life, and maintains the battery system in an accurate and reliable operational condition. The BMS performs several tasks such as measuring the system voltage, current and temperature, the cells state of charge (SoC), state of health (SoH), and remaining useful life (RUL) determination, protecting the cells, thermal management, controlling the charge/discharge procedure, data acquisition, communication with on-board and off-board modules, monitoring and storing historical data and the most important task is the cell balancing. Imbalance of cells in battery systems is very important matter in the battery system life. Because without the balancing system, the individual cell voltages will drift apart over time. The capacity of the total pack will also decrease more quickly during operation then fail the battery system. As mentioned in cited papers [1-2], currently, the most common methods of battery management are passive and active cell balance or equalization, which have been reviewed in [3-7] and described the related US patents listed below. However, none of these methods can access individual battery subset or single battery for measuring, monitoring and balancing, and compensation. The intelligent battery management system and methods of this intervention can solve the problem.

RELATED PUBLISHES

• [1] Predictive Intelligent Battery Management System to Enhance the Performance of Electric Vehicle, Mohamad Abdul-Hak, Nizar Al-Holou and Utayba Mohammad, Electrical & Computer Engineering Department, University Of Detroit Mercy, Detroit
• [2] Passive and Active Battery Balancing comparison based on MATLAB Simulation, Mohamed Daowd1, Noshin Omar1,2, Peter Van Den Bossche2, Joeri Van Mierlo1, 2011 IEEE
• [3] Jian Cao, Nigel Schofield and All Emadi, “Battery balancing methods: A comprehensive review,” IEEE Vehicle Power and Propulsion Conference, VPPC 08. pp. 1-6, 2008.
• [4] K. Zhi-Guo, Z. Chun-Bo, L. Ren-Gui and C. Shu-Kang, “Comparison and Evaluation of Charge Equalization Technique for Series Connected Batteries,” 37th IEEE Power Electronics Specialists Conference, pp. 1-6, 2006.
• [5] S. Moore and P. Schneider, “A Review of Cell Equalization Methods for Lithium Ion and Lithium Polymer Battery Systems,” in Proceedings of the SAE 2001 World Congress, 2001.
• [6] M. J. Isaacson, R. P. Hollandsworth, P. J. Giampaoli, F. A. Linkowsky, A. Salim and V. L. Teofilo, “Advanced lithium ion battery charger,” 15th Annual Battery Conference on Applications and Advances, pp. 193-198, January 2000.
• [7] N. H. Kutkut, and D. M. Divan, “Dynamic Equalization Techniques for Series Battery Stacks,” IEEE Telecommunications Energy Conference, INTELEC '96, pp. 514-521, 1996.

Related U.S. Patent Documents
	Application Number	Filing Date	Pat. No.
	U.S. Pat. No. 12/514,058	Nov. 12, 2007	8,350,529
	U.S. Pat. No. 13/445,423	Apr. 12, 2012	8,344,694
	U.S. Pat. No. 12/423,545	Apr. 14, 2009	8,310,177
	U.S. Pat. No. 12/116,800	May 7, 2008	8,273,474
	U.S. Pat. No. 12/880,801	Sep. 13, 2010	8,258,754
	U.S. Pat. No. 12/946,438	Nov. 15, 2010	8,271,054
	U.S. Pat. No. 11/921,442	Feb. 1, 2006	8,173,285
	U.S. Pat. No. 12/851,508	Aug. 5, 2010	8,134,340
	U.S. Pat. No. 12/941,365	Nov. 8, 2010	8,089,249

SUMMARY OF THE INVENTION

An intelligent battery management system and method, cutting-in at a new angle, optimizes rechargeable battery set to the best performance, which can overcome the backwards of all current management systems and methods, and reduce the manufacturing cost at better performances. The management system is supported by a cutting-edge method and corresponding embodiments, which can automatically re-combine batteries in a battery set in parallel connection or in series connection or in mixed series-parallel connection, or mixed parallel-series connection, each individual cell in a battery set can accessed, which means each individual cell can be monitored (i.e. its parameters can be measured) and charged or discharged. With this system, batteries can be charged or discharged to their best performance. Besides battery set, the system comprises connection controller [the combined connection controller, or parallel connection controller and series connection controller], micro-processor/controller, meters for battery measurement and monitoring, charger, load serve pack (electrical and mechanical), and thermal management apparatus.

DETAIL DESCRIPTION OF THE INVENTION

<System Configuration>

The system comprises battery set 100, connection controller [i.e. the combined connection controller 200, or parallel connection controller 300 and series connection controller 400, as show in FIGS. 1(a) and 2(a), respectively], micro-processor or controller 500, meters for battery measurement and monitoring 700, charger 800, load serve pack (electrical and mechanical) 900, and thermal management apparatus 600. In battery set 100, each unit can be a single battery, or can be a battery sub-set [like 100′ or 100″ as shown in FIGS. 1(b) and 2(b)], which is consist of multiple single batteries in series connection 100′, or in parallel connection 100″.

<Apparatus Embodiments and Methods for Arbitrarily Controlling Battery Connection>

Battery connection controller is the apparatus embodiments for arbitrarily controlling battery connection. In FIGS. 1(b) and 2(b) are shown two examples of battery connection controllers, with 2×12 power lines for 12 batteries or battery sub-sets (here 12 just for example, actually, lines can be 2, 3, 4, . . . and up to 10,000), the lines connected to negative pole are denoted by N1, N2, N3, . . . and the lines connected to positive pole are denoted by P1, P2, P3, . . . .

I. Combined Connection Controller

Connection controller is one of the key elements of this invention. In FIGS. 3 to 8 show details of combined connection controllers and separated connection controllers. The solid lines show the power lines [P1, P2, P3, . . . N1, N2, N3, . . . , PN, PP, C1, C2, C3, . . . ], and the dash lines show switch control lines [p1, p2, p3, . . . n1, n2, n3, . . . pn, pp, c1, c2, c3, c4, . . . ] which conduct the switch signal from micro-processer or controller 500. The cross of any two solid lines does not mean a connect of the two solid lines unless corresponding cross of dash lines [the zoom-in is shown in FIG. 4] satisfy certain conditions which is shown in FIG. 5. At the cross of two solid lines, there is a corresponding cross of two dash lines and a switch. The status [open or close] of switch determines the connection [not connected or connected, respectively] of two crossed solid lines, while the voltage difference of two crossed dash lines determines the status of switch. Control lines [pp, pn] are for connection control of power assembly poles (most outside layer, in parallel), and control lines [p1, p2, p3, . . . ] and lines [n1, n2, n3, . . . ] are for series connection control of inside layers, and control lines [c1, c2, c3, . . . ] (associated with [n1, n2, n3, . . . ]) are for parallel connection control of inside layers. Power lines [P1, P2, P3, . . . ] are connected to positive pole of battery or battery sub-sets, power lines [N1, N2, N3, . . . ] are connected to negative pole of battery or battery sub-sets. Power lines [PP, PN] are the assembly positive and negative power poles of the whole battery set, respectively. Bridge power lines [C1, C2, C3, . . . ] (and their control lines [c1, c2, c3, . . . ] also) are for the internally parallel connection, and they can be either crossed with negative power lines [N1, N2, N3, . . . ] (and with control lines [n1, n2, n3, . . . ] correspondingly) as shown in FIGS. 3, 5, 6, 8, or crossed with positive power lines [P1, P2, P3, . . . ] (and with control lines [p1, p2, p3, . . . ] correspondingly) which is not shown in Figures. The circles at end of solid and dash lines indicate the sockets for the connection (when be plugged) to positive and negative poles of batteries/subsets or switch controller, i.e. microprocessor.

Let's assume there are total Nb battery units, Nset battery subsets, and there are Nsb battery units in each subset, so Nset×Nsb=Nb, we need further to group these Nset battery subset into Ns rows in series connection, Np battery units inside each row in parallel connection, or to group these Nset battery subset into Np columns in parallel connection, Ns battery units inside each column in series connection, so Ns×Np=Nset, and Nsb×Ns×Np=Nb. Each battery unit is single battery along, or a single battery with balance/compensation device together. For example, Tesla RoadStar has 6831 battery, 69 cells are wired in parallel to create a brick, 9 (i.e. 3×3) bricks are connected in series to create sheets, and 11 sheets are inserted in series into the pack casing, in total, this creates a pack made up of 6,831 cells, which means Nsb=1, Np=69, and Ns=99 (3×3×11).

In FIG. 3 is shown an example of combined connection controller for series connection application using Nset=12 battery units or subsets. The connection controller divides 12 units into 3 groups [7 3 1 10], [2 11 8 4] and [12 6 9 5], and connect 4 units in each group in series, then connect the 3 groups in parallel. The white triangle in all relevant figures mean the switch is off (disconnected), and the black triangle means the switch is on (connected).

In FIG. 6 is shown example of combined connection controller for parallel connection application using Nset=12 battery units or subsets. The combined connection controller divides 12 units into 4 groups [7 2 12], [3 11 6], [1 8 9], and [10 4 5], and connect 3 units in each group in parallel, then connect the 4 groups in series.

Let's give more description on FIGS. 4 and 5, so as to have better understand how switch control works, and further to understand how connection controller shown in FIGS. 3 and 6 works. For the combined connection controller, two kinds of switch are needed, with ΔV1 and ΔV2 (ΔV2>=ΔV1, including the case ΔV2=ΔV1) as their switch voltages respectively. For better operation, difference (ΔV2−ΔV1) should not too large comparing with ΔV1, say within ΔV1, i.e. (ΔV2−ΔV1)<ΔV1. The switch can be a mechanic type, or an electrical type, but for the connection in outside layer, switch with higher power is needed. Therefore, 4 control voltage levels V1, V2, V3 and V4 (as shown in FIG. 5(a)) or 5 level voltage levels V0, V1, V2, V3 and V4 (as shown in FIG. 5(b)) are needed to perform the switching control. In these figures, all p-lines (i.e. p1, p2, p3, . . . ) are denoted as pL; all c-lines (i.e. c1, c2, c3, . . . ) are denoted as cL, all n-lines (i.e. n1, n2, n3, . . . ) are denoted as nL, and ppL means pp line, pnL means pn line.

For 4 level control voltages (in FIG. 5(a)), all nL have three level voltages V1, V2 and V3; all pL have three level voltages V1, V2 and V4; all cL have two level V1 and V2, ppL always stays at V2; pnL always stays at V1, where V1=−ΔV1/2, V2=+ΔV1/2, V3=V1−ΔV2, V4=V2+ΔV2, here is logical table for switch control:

• • pL or cL=V1 & nL=V2 conduct (dV=+ΔV1)
  • pL or cL=V1 & nL=V1 not conduct (dV=0)
  • pL or cL=V2 & nL=V2 not conduct (dV=0)
  • pL or cL=V2 & nL=V1 not conduct (dV=−ΔV1)
  • pL or cL=V4 & nL=V1 not conduct (dV=−ΔV2−ΔV1)
  • pL or cL=V2 & nL=V3 not conduct (dV −ΔV2−ΔV1)
  • ppL=V2 & pL=V1 not conduct (dV=−ΔV1)
  • ppL=V2 & pL=V2 not conduct (dV=0)
  • pnL=V1 & nL=V1 not conduct (dV=0)
  • pnL=V1 & nL=V2 not conduct (dV=+ΔV1<+ΔV2)
  • ppL=V2 & pL=V4 conduct (dV=+ΔV2)
  • pnL=V1 & nL=V3 conduct (dV=+ΔV2)

For 5 level control voltages (in FIG. 5(b)), all nL have two level voltages V0 and V1; all pL and cL have two level V0 and V2, ppL has two level V0 and V4; pnL has two levels V0 and V3. The default control voltage of all switch control lines are set at V0, voltages V1 and V2 are used for inside layer (series and parallel) connection control; V3 and V4 (associate with V1 and V2) are used for connection control of power assembly poles (most outside layer, in parallel), where V1=−ΔV1/2, V2=+ΔV1/2, V3=V0−ΔV2, V4=V0+ΔV2, here is logical table for switch control:

• • pL or cL=V2 & nL=V1 conduct (dV=ΔV1)
  • pL or cL=V2 & nL=V0 not conduct (dV=ΔV1/2)
  • pL or cL=V0 & nL=V1 not conduct (dV=−ΔV1/2)
  • pL or cL=V0 & nL=V0 not conduct (dV=0)
  • pnL=V3 & nL=V0 conduct (dV=ΔV2)
  • pnL=V3 & nL=V1 conduct (dV=ΔV2−ΔV1/2)
  • pnL=V0 & nL=V0 conduct (dV=0)
  • pnL=V0 & nL=V1 conduct (dV=−ΔV1/2)
  • ppL=V4 & pL=V0 conduct (dV=ΔV2)
  • ppL=V4 & pL=V2 not conduct (dV=ΔV2−ΔV1/2)
  • ppL=V0 & pL=V0 not conduct (dV=0)
  • ppL=V0 & pL=V2 not conduct (dV=ΔV1/2)

II. Separate Connection Controllers

Unlike the combined connection controller 200, parallel connection controller 300 and series connection controller 400 can be built separately, as shown in FIG. 2(a), so called separate connection controllers, their detail layouts are shown in FIGS. 8 and 7, respectively. In this case, we only need two control voltage levels for each of them. In FIG. 7 is shown an example of using series connection controller (rather than combined connection controller in FIG. 3) for the part of series connection in same application as shown in FIG. 3 and FIGS. 8 (a) and 8(b) are shown two examples of using parallel connection controller (rather than combined connection controller in FIG. 6) for part of parallel connection in the same application as shown in FIG. 6.

The function in FIG. 7 only does all series connections within each of the 3 groups, but not do parallel connection for the three groups yet. The switches on pp and pn control lines in FIG. 8 (a, b) will take care of the parallel connection, i.e. connecting the positive poles of units 7, 2 and 12 to power output line PP and connecting negative poles of units 10, 4 and 5 to power output line PN. All the switches at crosses of nL (i.e. n1, n2, n3, . . . ) and pL (i.e. p1, p2, p3, . . . ) are never used. So the circuit is redundant if only for series connection application in FIG. 3.

In FIG. 8, one may note that, for parallel connection application in FIG. 6, configuration in FIG. 8 (a or b) and configuration in FIG. 7 should be used together. If connecting the 3 positive poles of units 3, 11 and 6 (or negative poles of units 7, 2 and 12), and connecting the 3 positive poles of units 1, 8 and 9 (or negative poles of units 3, 11 and 6), and connecting the 3 positive poles of units 10, 4 and 5 (or negative poles of units 1, 8 and 9), then the series connection table shown on top left corner of FIG. 7 is converted into parallel connection table shown on top left corner of FIG. 8(a, or b). Let consider row (i.e. units 7, 2, and 12), they are already connected to units 3, 11, and 6 through series connection part, respectively. Therefore, as shown in FIG. 8(a), connecting negative pole of unit 7 only with positive pole of units 11 and 6 only can achieve the parallel connection for negative pole the 1^st row and for positive pole the 2^nd rows, all these connections 7-3, 2-3, 2-11, 2-6, 12-3, 12-11, and 12-6 are not necessary, same applied for (2^nd and 3^rd) rows, and so on. One may note that the circuit is still redundant (some switch will not be used). If circuit as shown in FIG. 8(b) is built, the redundancy can be removed.

III. Monitoring and Measurement Accessories

Using this invented apparatus and method, either each battery subset or each single battery (if subset has only single battery, which means without using battery subset) can be accessed by monitoring instrument and its electrical and electrochemical parameters (such as voltages [open & load], current, resistances [discharge, ohmic←electrolyte & electronic types, non-ohmic or nonlinear, etc.], impedances and impedance spectrum, electric capacity and discharge capacity, SOC[state of capacity], SOH[state of healthy], etc.) can be measured and be monitored as long as the micro-controller select this subset. The monitoring and measuring process can be done every day, or even moment when battery is not at loading or recharging, such as when waiting at red light. Measurement or monitoring is rotated subset by subset or is continued from the measurement for last subset, and result for each subset is stored and updated timely.

As mentioned above, either each battery subset or each single battery (means without using “battery subset”) can be accessed by monitoring instrument. However, if selecting the latter, i.e. applying connection controller described above to access EVERY SINGLE battery unit without using “buttery subset”, it is very cost and reduces the product competition power (of course, we can build a system in this way if cost is not an issue); while on the other hand, if selecting the former, i.e. applying connection controller described above to access every battery subset, without special treatment, each single battery units inside each battery subset cannot be directly accessed by monitoring instrument. The special treatment described below will solve this problem.

FIGS. 9 and 10 show a method (special treatment) to access each single battery and do the measurement for each single battery in a battery subset, using same (i.e. no extra needed) combined connection controller or separated controllers for measurement, loading and recharging, and using a clicker device called selection clicker (either stripe by sliding as show in FIGS. 9 and 10, or short cylinder or circle by rotating [not shown in charts, it can be constructed by connecting the two ends of stripe clickers in FIGS. 9 and 10]) to switch between “measurement” and “application” (including loading and recharging). When switching to “measurement”, all terminates of connection controller, i.e. power lines [P1, P2, P3, . . . N1, N2, N3, . . . ] are disconnected from the terminates of all battery subsets, and connected to the all terminates of each single battery in a selected battery subset, meanwhile, the connection (parallel or series) inside the battery subset is decomposed as shown in FIGS. 9 and 10 [from (b) to (a)]; When switching to “application”, all terminates of connection controller, i.e. power lines [P1, P2, P3, . . . N1, N2, N3, . . . ] are connected to the terminates of battery subsets, after disconnected from the all terminates of each single battery in a selected battery subset, meanwhile, the connection (parallel or series) inside the battery subset is re-composed as shown in FIGS. 9 and 10 [from (a) to (b)].

In FIGS. 9(a) and (b) are shown an example (but not limited to this configuration) of clicker device for the case of series connection inside a battery subset. In FIGS. 10(a) and (b) are shown an example (but not limited to this configuration) of clicker device for the case of parallel connection inside a battery subset. In both FIGS. 9 and 10, charts (a) indicate switching to “measurement”; charts (b) indicate switching to “application” (i.e. loading and recharging, for either series or parallel connections). The “measurement” can access each subset in random, or in sequence. If number of power lines in connection controller is greater than the number of batteries in a subset, measurement can be done in one action for all batteries in one battery subset; otherwise, measurement needs multiple actions for one subset. For example, if number of power lines is 12 (as shown in FIG. 3) and number of batteries in a subset is 33 (as shown in FIGS. 9 and 10), then the measurement need 3 actions: first action is to connect and measure battery No. 1 to 12 in the subset, second action is to connect and measure battery No. 13 to 26 in the subset, and the third action is to connect and measure No. 27 to 33 in the subset.

Unlike the monitoring and measuring process for battery subsets, as mentioned before, which can be done every day, or even every moment when battery is not at loading or recharging. For single battery inside a subset, measuring, monitoring, adjustment, and re-combination process can be done every month, or on very weekend. According to measured results, battery in one subset may be made an adjustment or/and re-combination, i.e. exchange location with battery in another subset or with battery within same subset, to get an optimization configuration.

<Recharging and Loading Assignment Optimization>

Using invented embodiments, batteries can be arbitrarily connected, either in series, or in parallel, and the number battery subsets and number of single batteries in a subset can be arbitrarily assigned. Regarding how to connect, i.e. how many batteries are assigned in a subset and these batteries in a subset are connected in parallel or in series, or in mixed, and how many battery subsets are used in series connection in a branch and many parallel branches are assigned, or how many battery subsets are used in parallel connection in a layer and many series layers are assigned, . . . and so on, it is completely determined by modeling analysis and optimization of recharging and loading (including the balance elements and motor inductance together), which is based on the measurement or monitoring data of batteries. The detail of modeling analysis and optimization algorithms for recharging and loading varies from case to case of application. The optimization is based on impedance and other parameters, which describe battery's performance, also including motor inductance and balance elements. In the below, only a simple and strait forward example is given to illustrate the basic idea of management, and what optimization of configuration for battery connection means, but actually detail algorithms are much complex than what described in this example from here below.

Assume we have 8000 Li-ion batteries with average open voltage 3.2 Volts, and their intern resistances have a distribution (from larger, median, to smaller). If design motor working at 320 Volts, then we need a connection matrix of battery to achieve this goal and there are many connection ways to do that, here we show three direct or straight forward ways: the 1^st way, parallel connection before series connection, is dividing 8000 batteries into 100 groups, and connect 80 batteries in parallel (80 columns) in each group, and then connect all 100 groups in series (100 rows); the 2nd way, series connection before parallel connection, is diving 8000 batteries into 80 groups, and connect 100 batteries in series in each group, then connect all 80 groups in parallel; the 3^rd way is diving 8000 batteries into 10 (rows)×10 (columns) subset battery matrix, and each battery subset is consist of 10 (rows)×8 (columns) single battery matrix. Actually, the 3^rd way includes 4 different compositions (ways) as shown in the table below:

Compositions
10 × 10 subsets            Compositions 10 × 8 batteries in each subset
parallel connection before parallel connection before series connection
series connection
parallel connection before series connection before parallel connection
series connection
series connection before   parallel connection before series connection
parallel connection
series connection before   series connection before parallel connection
parallel connection

Actually, the 3^rd way can be extended to many ways, i.e. diving 8000 batteries into n1 (rows)×m1 (columns) subset battery matrix, and each battery subset is consist of n2 (rows)×m2 (columns) single battery matrix, with constrains n1×n2=100 and m1×m2=80. The integral solutions of n1×n2=100 are 1×100, 2×50, 4×25, 5×20, 10×10 (total 5 solutions); The integral solutions of m1×m2=80 are 1×80, 2×40, 4×20, 5×16, 8×10 (total 5 solutions), so there are 5×5×4=100 ways extended from the 3^rd way. Further more, the 3^rd way only include one level battery subset. It can be extended to even more ways by including more levels of battery subset, such as, sub-subset in a subset, and sub-sub-subset in a sub-subset, . . . and so on . . . and there are total a few hundreds ways to connect these 8000 batteries.

As we already seen above, extended from the 3^rd ways above, there are total a few hundreds ways to connect these 8000 batteries, the optimization means finding out the best way for the application from these hundreds composition ways (circuits), and within the best way (circuit), finding out best combination (for cell matching) for all individual battery cells. Therefore optimization needs help from programming or modeling and it is complex to describe through language. However, optimization idea can described through a simple example applied on the way and the 2^nd way.

Let's talk about the 2^ad way first in the below, because it is straightforward. As we know that, open circuit voltage of battery is a nonlinear, but monotonic function of SOC. So if open-circuit voltages are matched, the subsets or batteries will have equal SOC. Under loading, terminal voltage will vary significantly from open-circuit voltage due to internal impedance of battery or subset, which is again a nonlinear function of SOC. If all of the batteries in a column are properly managed through re-combination by connection controller of this invention, their internal impedances should be nearly equal. We can simply group batteries by impedance, so that, within each column, every batteries have almost same internal impedance, but different column has different total internal impedance, then when re-charging voltage is applied, the voltages divided onto each battery in a column are almost same, but different column is assigned different current according to different total impedance—the column with smaller impedance (corresponding higher SOC) passes more current and so gets more charged energy—which is consistent with higher capacity even if without help from balance elements (or circuits) and equalization, so all columns can reach their maximum charging at almost same time. Without this procedure, once any one bad battery in the matrix reach it's maximum charging, the re-charging process must stop for all battery, even if most of the batteries are far away from completely charged, in this case, balance elements and equalization is a must, but cost is much higher. When discharging with load, the batteries need to be re-connected, so that, within each column, the every batteries have closest SOC (not necessary completely consistence with impedance due to battery intrinsic complexity), but different column has different total SOC, the SOC's of some of them are larger, but some are smaller. Considering battery prefers working at smaller current (longer life time), the invented management system will use all batteries, i.e. all columns, when automobile is speeding up [need more power], or climbing up steep hills, so that each battery works at smallest current. However, when automobile is driving on flat road, or at low speed, management system will not use all battery columns, again, considering battery prefers working at smaller current, the management system only excludes only one, or two, or a few (but “one” may be the best) battery column with lowest SOC from loading, i.e. the management system always picks out the column with lowest SOC, and exclude it from the loading at low power consuming state. May be 5 minute later, this one is longer the one with lowest SOC, the system will pick out a new one with lowest SOC for instead. In this way, the stored energy of all batteries can be almost completely used before next re-charging. In the example of 8000 battery cells mentioned above, there are 100 rows and 80 columns. After exclude the 1 column with lowest SOC, there are 79 columns left for loading, so we could use either the 1^st way (parallel connection before series connection) or the 2^nd way (series connection before parallel connection) to composite the 100 rows×79 columns battery matrix.

The 1^st way functions almost same as the 2^nd way except it has local self-compensation within each row through parallel connections of every internal layer. While, the 2^nd way does not have because the parallel connections are only at the two outside terminals (no internal parallel connections). As mentioned above, we can group batteries with closest internal impedance or closest SOC in same column, bases on this assignation, we could further assign the batteries for each rows so that all the rows have closest total impedance 1/[1/Zk1+1/Zk2+1/Zk3+ . . . +1/Zk79] (still using the 8000 cells example above, where k=1, 2, 3, . . . , 100). Whatever in re-charging or discharging, within a row, a cell with smaller impedance (larger SOC) will pass more current, while a cell with larger impedance (smaller SOC) will pass less current, which is equivalent to a local current cycle (around average) from larger SOC cell to small SCO cell, canceling the latter current (re-charging) and enhancing the former current—this is so called local self-compensation.

<Thermal System and its Management>

“Hole-net thermal bed” is a preferred embodiment for thermal management. The individual cells within a battery set differ due to manufacturing variations, temperature gradients, and aging effects (also temperature dependent). Therefore, battery or battery subset need to be installed in a thermal bed, which is made up of good thermal conduct material, such as copper, Aluminum, etc. Inside the walls of battery thermal bed, there is hole-net (i.e. many air flow channels) which is connected to thermal controlling system, the air with almost same temperature is conducted to thermal beds of all batteries through the hole-net, so that all of the battery have almost same temperature as the battery bed, which eliminates the temperature gradients (cell to cell and inside every cell) at the best situation. When starting use of battery at cold weather, the heating system

Claims

1-10. (canceled)

11. battery management system, with corresponding embodiments and methods, comprising battery set, connection controller, switch controller, battery monitoring meters, charger, and load, with thermal control system as option, for optimizing battery set to the best performance, wherein any single battery, or any battery in battery subset or battery set can arbitrarily be accessed and be connected, to satisfy the requirements for measurement, monitoring, compensation and balance, recharging, loading and thermal control;

12. embodiments called connection controller with corresponding connection control methods for arbitrarily connecting batteries or battery subsets in parallel, or in series, or in any kind of mixed connections, comprising (1) negative power lines with corresponding sockets for connecting to all negative poles of batteries or battery subsets;(2) positive power lines with corresponding sockets for connecting to all positive poles of batteries or battery subsets;(3) one negative power line with corresponding one socket for connecting to assembly negative power poles of the whole battery set;(4) one positive power line with corresponding one socket for connecting to assembly positive power poles of the whole battery set;(5) bridge power lines but without any socket for achieving parallel connections within connection controller, having cross (i.e. have option of connection or disconnection through switch control) with either negative power lines or positive power lines;(6) matrix of switches at the cross of negative power lines and positive power lines,(7) negative and positive switch control lines which have one by one correspondence to each of the power lines mentioned in (1) to (5) of this claim, wherein, the switch control lines are (in pair of negative and positive) connected to each switch in the switch matrix, and is connected to switch controller;wherein, the status [open or close] of switch determines the disconnection or connection of two crossed power lines, respectively, while the voltage difference of two crossed switch control lines determines the status of switch;wherein, the voltage difference of two crossed switch control lines is controlled by switch controller;wherein, bridge power lines are not necessary for series connection controller;

13. embodiments with corresponding method to access any individual battery subset and to access any individual single battery in a battery subset, comprising selection clicker with the use of same (i.e. no extra needed) connection controller for measurement, loading and recharging, wherein, the said selector is either stripe type by sliding, short cylinder or circle type by rotating, which switches between “measurement” and “application”, wherein, (1) when said selector switching to “measurement”, all terminates (positive and negative power lines) of connection controller are disconnected from the terminates of all battery subsets, and connected to the all terminates of each single battery in a selected battery subset, meanwhile, the connection (parallel or series) inside the battery subset is decomposed; (2) when switching to “application”, all terminates of connection controller are connected to the terminates of battery subsets, after disconnected from the all terminates of each single battery in a selected battery subset, meanwhile, the connection (parallel or series) inside the battery subset is re-composed;

14. the management system of claim 1, wherein the management optimization of battery connections, compositions and matches for best recharging and loading performance, comprising (1) measure battery electrical and electrochemical parameters, using invented connection controller and (as an option) selection clicker,(2) optimize connection, i.e. finding out the best connections and battery compositions for recharging and loading, by modeling analysis, from all possible connections and battery compositions based on measured parameters, with balance elements included if any or necessary, and with thermal conditions,(3) use invented connection controller to achieve the best connections;

15. the optimization method of claim 4, wherein the method for loading comprising (1) using all battery when high power needed, such as speed up, or climbing hills,(2) excluding the weakest battery branch or branches from loading when only low power needed, such as on flat road.

16. the management system of claim 1, wherein the connection controller is a single combined connection controller, or is a composition of two separated connection controllers, i.e. a composition of parallel connection controller and series connection controller;

17. the connection controller of claim 2, wherein the switches are either mechanic type, such as relay, but not limited to this, or electronic type, such as, semiconductor chips, to integrated chips, but not limited to these;

18. the connection controller of claim 2, wherein the power lines and control lines are either electrical type, such as metal wires, or conductive wires on printed wires on circuit board; or electronic type, such as, micro circuit wires in integrated chips; but not limited to these;

19. the connection controllers of claim 2, wherein the switch controller is any kind of smart integrated chip, such as, micro-processor, or micro-controller, or CPU, etc, but not limited to these;

20. the management system of claim 1, wherein the embodiment of thermal control system is a battery thermal bed, comprising thermal bed by good thermal conduct material, and hole-net (i.e. many air flow channels) inside the walls of thermal bed, wherein, hole-net is connected to thermal controlling system, the air with almost same temperature is conducted to thermal beds of all batteries through the hole-net, so that all of the battery have almost same temperature as the battery bed, which eliminates the temperature gradients (cell to cell and inside every cell) at the best situation.