代码拉取完成,页面将自动刷新
import numpy as np
import pandas as pd
# import matplotlib.pyplot as plt
def get_flow(coefficient):
index = 0
# 读取 input.csv 文件
input = pd.read_csv(f"./data/input/input{index+1}.csv", header=0)
row = index
info = pd.read_csv("./data/origin/source.csv", header=0)
area = info.iloc[row, 5]
true_peak = info.iloc[row, 8]
Sr = info.iloc[row, 6]
Ks = info.iloc[row, 7]
print("Sr:", Sr)
print("Ks:", Ks)
print("area:", area)
print("true peak:", true_peak)
TIME = input.iloc[:, 0] # 时间
Rain = input.iloc[:, 1] # 时段降雨量
EM = input.iloc[:, 2] # 时段蒸发量
QS = input.iloc[:, 3] # 时段预报流量
QU = input.iloc[:, 4] # 支流汇入,取QU=0
# 输入预报流域面积
Area = area # 水库控制流域面积
B, alpha0, JL, KKS, KKSS, CS = coefficient[:6]
z = coefficient[6:9]
ni = coefficient[9:12]
H01 = coefficient[12:15]
Hc1 = coefficient[15:18]
# Hr1 = coefficient[18:21]
delta = coefficient[21:24]
LS, LSS, L = map(round, coefficient[24:27])
Gam = coefficient[27:30]
EC, KKG, LG = coefficient[30], coefficient[31], round(coefficient[32])
Hs = ni * z
H0 = ni * z * H01
Hc = ni * z * Hc1
# Hr = ni * z * Hr1
H = H0.copy()
h0 = H[0]
nz = ni[0] * z[0]
B0 = h0 / nz
dt = TIME[1] - TIME[0]
Nt = len(Rain)
M = len(z)
PE0 = np.zeros(Nt)
# G = np.zeros(M)
# Gr = np.zeros(M)
E = np.zeros(M)
RSS = np.zeros(Nt)
RS = np.zeros(Nt)
RG = np.zeros(Nt)
RRR = np.zeros(Nt)
balance = np.zeros(Nt)
TRS = np.zeros(Nt)
TRSS = np.zeros(Nt)
TRG = np.zeros(Nt)
Fm, T0, E0 = 0, 0, 0
# 循环计算
for i in range(Nt):
P = Rain[i]
# 计算净雨量(扣除蒸发和截留)
if P >= JL:
P -= JL
JL = 0
else:
JL -= P
P = 0
if P <= EM[i]:
DEM = EM[i] - P
PE = 0
PE0[i] = PE
else:
DEM = 0
PE = P - EM[i]
PE0[i] = PE
# 计算地表径流
B1 = 1 - B0
Fm1 = Sr * B1 * (T0**0.5) + Ks * T0
Fm2 = Sr * B1 * ((T0 + dt) ** 0.5) + Ks * (T0 + dt)
Dfm = Fm2 - Fm1
X = PE / Dfm
B1 = B / (1 + B0)
eta1 = alpha0 + (1 - alpha0) / (B1 + 1)
if X <= alpha0:
eta = X
elif X < 1:
eta = eta1 - (1 - alpha0) / (B1 + 1) * ((1 - X) / (1 - alpha0)) ** (B1 + 1)
else:
eta = eta1
Df0 = eta * Dfm
Drb = PE - Df0
if Drb < 0:
Drb = 0
if X <= 1:
Fm += PE
else:
Fm += Dfm
S0 = Sr * (1 - B0)
A0 = Ks
SA = (S0**2 + 4 * A0 * Fm) ** 0.5
T0 = ((SA - S0) / (2 * A0)) ** 2
# 壤中流计算
Drs = 0
for k in range(M):
if DEM > 0:
if H[0] + Df0 >= DEM:
E[0], E[1], E[2] = DEM, 0, 0
else:
E[0] = H[0] + Df0
if H[1] > EC * Hs[1]:
E[1] = (DEM - E[0]) * H[1] / Hs[1]
E[2] = 0
else:
if H[1] > EC * (DEM - E[0]):
E[1] = EC * (DEM - E[0])
E[2] = 0
else:
E[1] = H[1]
E[2] = EC * (DEM - E[0]) - E[1]
else:
E[0], E[1], E[2] = 0, 0, 0
Hw = Df0 + H[k] - E[k]
beta0 = Hc[k] / Hs[k] / (Gam[k] + 1)
omega1 = beta0 + (1 - beta0) / (Gam[k] + 1)
beta = Hw / Hs[k]
omega = beta
if beta <= beta0:
omega = beta
elif beta >= beta0 and beta < 1:
omega = omega1 - ((1 - beta0) / (Gam[k] + 1)) * (
(1 - beta) / (1 - beta0)
) ** (Gam[k] + 1)
elif beta >= 1:
omega = omega1
Ds = Hw - Hs[k] * omega
Dr = Ds * delta[k]
Drs += Dr
DF = (1 - delta[k]) * Ds
H[k] = H[k] + Df0 - Ds - E[k]
Df0 = DF
if H[k] < 0:
H[k] = 0
if H[k] > Hs[k]:
H[k] = Hs[k]
RSS[i] = Drs
RS[i] = Drb
RG[i] = Df0
RRR[i] = Drs + Drb + Df0
E0 += np.sum(E)
balance[i] = np.sum(H - H0) + np.sum(RRR) - np.sum(PE0) + E0
# 汇流计算
U = Area / dt / 3.6
TRS[:LS] = 0
TRS[LS] = RS[0] * U
for T in range(1 + LS, Nt):
TRS[T] = TRS[T - 1] * KKS + RS[T - LS] * (1 - KKS) * U
for T in range(1 + LSS, Nt):
TRSS[T] = TRSS[T - 1] * KKSS + RSS[T - LSS] * (1 - KKSS) * U
TRG[:LG] = QS[0] - QU[0]
for T in range(1 + LG, Nt):
TRG[T] = TRG[T - 1] * KKG + RG[T - LG] * (1 - KKG) * U
TR = TRS + TRSS + TRG + QU
QJ = TR.copy()
if L < 0:
L = 0
for T in range(L + 1, Nt):
QJ[T] = CS * QJ[T - 1] + (1 - CS) * TR[T - L]
# # 绘图
# plt.style.use("fivethirtyeight")
# plt.plot(TIME, QJ, "b-", linewidth=2)
# plt.xlabel("time")
# plt.ylabel("flow")
# plt.title("predicted flow")
# plt.legend(["calculated flow"])
# plt.show()
difference = max(QJ) - true_peak
print("peak flow:", max(QJ))
print("difference:", difference)
return QJ, difference
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。