def __init__(self, time):
print("path:" + sys.path[0])
#线路静态信息
self.startPoint = trc.SLStartPoint[0]
self.endPoint = trc.SLStartPoint[-1]
self.S = self.endPoint - self.startPoint
self.max_speed = 80 / 3.6
self.T = time #运行总时间
self.avg_speed = self.S / self.T #平均速度
self.dt = 0.2 #运行时间步长
self.low = np.array([0, 0])
self.high = np.array([self.S, self.max_speed])
self.ac = 0.8 #最大加速度
self.de = 1 #最大减速度
self.viewer = None
self.n_actions = 9 #-0.8 :0.2 :0.8
self.n_features = 4 #用于训练的特征，和self.state相对应
self.action_space = [
'-0.8', '-0.6', '-0.4', '-0.2', '0', '0.2', '0.4', '0.6', '0.8'
]
self.action_space = spaces.Discrete(9) #下标从0开始
self.observation_space = spaces.Box(self.low, self.high)
self.seed()
self.done = False
self.filterFactor = 0.8
#线路动态信息
self.trd = trc.TrainAndRoadData()
self.reset()
def __init__(self, start, destination, T, dt):
super().__init__(dt, 0)
#离线信息
self.tripStart = start #旅程起点
self.tripDestination = destination #旅程终点
self.tripTime = T
self.dv = 0.1 #允许速度误差
self.MaxAcc = 1.5 #最大牵引加速度
self.MinAcc = -1.5 #最大制动加速度
self.dadt = 0.5 #加速度变化率，初始为0.5
self.trd = trc.TrainAndRoadData()
self.MinT = self.trd.getMinRestTime(self.tripDestination)
#在线信息（状态）
self.pos = self.tripStart #位置
self.veo = 0 #当前速度
self.time = 0 #仿真时间步
self.vbar = 0 #预测速度
def __init__(self, start, destination, T, dt):
super().__init__(dt, 0)
#离线信息
self.tripStart = start #旅程起点
self.tripDestination = destination #旅程终点
self.tripTime = T
self.dv = 0.1 #允许速度误差
self.MaxAcc = 1.5 #最大牵引加速度
self.MinAcc = -1.5 #最大制动加速度
self.dadt = 0.5 #加速度变化率，初始为0.5
self.trd = trc.TrainAndRoadData()
self.MinT = self.trd.getMinRestTime(self.tripDestination)
#在线信息（状态）
self.pos = self.tripStart #位置
self.veo = 0 #当前速度
self.time = 0 #仿真时间步
self.trueAcc = 0 #当前的加速度acc
self.vbar = 0 #预测速度
self.last_veo = 0
self.last_pos = start
self.last_out = 0
self.step = 0
#xgboost训练好的模型
self.bst = xgb.Booster(model_file='./xgboost_algorithm/xgb.model')
def TanslationBySimulation(switchPoint, index):
#模拟列车运行，将工况控制点转换成（s,v,t,u)组合
dt = 0.2 #时间步长
startPoint = trc.SLStartPoint[0]
endPoint = trc.SLStartPoint[-1]
sl = trc.getRoadspeedLimit(startPoint)
gd = trc.getRoadGradinet(startPoint)
nsl = trc.getNextSpeedLimit(startPoint)
nsld = trc.getSpeedLimitEndPoint(startPoint)
vList = [0] #速度
sList = [startPoint] #位置
tList = [0] #时间
uList = [1] #加速度
gdList = [gd] #坡度 千分
slList = [sl] #路段限速（m/s)
nslList = [nsl] #下一限速的值（m/s）
nsldList = [nsld] #下一限速的距离（m)
train = trm.Train_model(startPoint, 0, 0.6, dt)
PIDC = atoController.PidController(dt, 8, 10, 1)
trd = trc.TrainAndRoadData()
t = 0
accList = [0]
stateList = [2]
state = 2
acc = 0.1
while True:
t = t + dt
laststate = state
state = trc.getRunState(sList[-1], switchPoint)
if state == 1 and laststate != 1:
vbar = vList[-1]
if state == 2:
#牵引
acc = 1
if state == 1:
#巡航
acc = PIDC.Step(vbar, vList[-1])
elif state == 0:
#惰行
acc = 0
elif state == -1:
#制动
acc = -1
if vList[-1] > trd.getEmerencyBrakeSpeed(sList[-1]):
acc = -1
out = train.Step(acc)
trueAcc = out['acc']
stateList.append(state)
accList.append(acc)
sl = trc.getRoadspeedLimit(out['S'])
gd = trc.getRoadGradinet(out['S'])
nsl = trc.getNextSpeedLimit(out['S'])
nsld = trc.getSpeedLimitEndPoint(out['S'])
vList.append(out['v'])
sList.append(out['S'])
tList.append(t)
uList.append(acc)
gdList.append(gd) #坡度 千分
slList.append(sl) #路段限速（m/s)
nslList.append(nsl) #下一限速的值（m/s）
nsldList.append(nsld) #下一限速的距离（m)
if out['S'] > endPoint or out['v'] < 0:
break
#保存数据
plt.plot(sList, accList)
plt.plot(sList, stateList)
plt.show()
trc.plotSpeedLimitRoadGrad('abstract')
plt.plot(sList, vList)
plt.show()
plt.plot(tList, vList)
plt.show()
print('-------------------%d---------------------' % index)
dataList = []
for i in range(0, len(vList)):
s = sList[i]
# t=tList[i]
sr = endPoint - sList[i]
t = tList[i]
tr = tList[-1] - t
vn = vList[i]
un = uList[i]
sr = round(sr, 2)
tr = round(tr, 2)
vn = round(vn, 2)
sl = slList[i]
gd = gdList[i]
nsl = nslList[i]
nsld = nsldList[i]
line = [s, sr, tList[-1], tr, sl, gd, nsl, nsld, un]
#如果list是一维的，则是以列的形式来进行添加，如果list是二维的则是以行的形式进行添加的
dataList.append(line)
tC = np.mat([sList, vList])
targetCurve = pds.DataFrame(data=tC.T, columns=['s', 'v'])
pData = pds.DataFrame(
data=dataList,
columns=['s', 'sr', 't', 'tr', 'sl', 'gd', 'nsl', 'nsld', 'un'])
return pData, targetCurve, tList[-1]