double VPaiNextState(ChessState state, StrategyState strategy)
{
int Tot = 0; bool isKing = state.State[strategy.SlcR][strategy.SlcC] == ChessType.BKing;
string mstate = ChessState.StateToStr(state);
string astrategy = StrategyState.StaToStr(strategy);
//插入决策
DataOperation.InsertStrategy(astrategy);
VSTimes[] NsT = new VSTimes[10000];
NsT = DataOperation.SelectProbTimes(mstate, astrategy);
//计算V(s+1)
double NS = 0.0;
for (int i = 0; i < NsT.Count(); i++)
{
NS += NsT[i].Vs * NsT[i].times;
Tot += NsT[i].times;
}
if (Tot > 0) NS /= Tot;
//从数据库里读出v(s)
double vs = DataOperation.SelectVState(mstate);
//从数据库读出瞬时回报值
double r = DataOperation.SelectReward(mstate, astrategy, isKing);
//计算现在的v(s)
vs = vs + a * (r + u * NS - vs);
return vs;
}
public static VSTimes[] SelectProbTimes(string state, string strategy)
{
int i = 0;
VSTimes[] tmps = new VSTimes[10000];
BlackTD0DataContext blackTD0 = new BlackTD0DataContext();
var tmp = from c in blackTD0.STATEs
from o in blackTD0.ASTRATEGies
from e in blackTD0.PROBs
from f in blackTD0.VSTATEs
where c.SNO == e.SNO && f.SNO == c.SNO && o.ANO == e.ANO && c.MSTATE == state && o.STRATEGY == strategy
select new
{
f.VALUE,
e.TIMES
};
if (tmp.Count() > 0)
{
foreach (var c in tmp)
tmps[i++] = new VSTimes(c.VALUE, c.TIMES);
}
return tmps;
}
