# US 2025/0165664 A1 · Analytical method for dynamic analysis of natural gas network

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0165664 A1
- Document ID: US20250165664A1
- Status: Published application
- Kind: A1
- Application number: 19/026,349
- Filed: 2025-01-16
- Published: 2025-05-22
- Earliest claimed: 2023-11-03
- Assignees: Southeast University
- Applicants: Southeast University
- Inventors: Suhan Zhang; Wei Gu; Shuai Lu
- Web page: https://patentgrep.com/patent/US20250165664A1

## CPC

- G06F 30/18: Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling (circuit design at the physical level G06F30/39; network planning tools for wireless communication networks H04W16/18)
- G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30)
- G06F 30/28: using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
- G06F 2111/10: Numerical modelling
- G06F 2113/08: Fluids
- G06F 2113/14: Pipes
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Disclosed is an analytical method for dynamic analysis of a natural gas network in the field of energy system modeling and operational analysis, which includes establishing adynamic model of natural gas transmission according to the conservation equations, and reconstructing the dynamic model into the equations in a heat conduction equation form. The present disclosure directly constructs an analytical method for dynamic analysis of a natural gas network, avoiding approximation errors, numerical dispersion, and dissipation compared with the traditional numerical methods. The discretization process is avoided during the solution, greatly improving the computational efficiency and solution accuracy of dynamic analysis of the natural gas network.

## Claims

### Claim 1

An analytical method for dynamic analysis of a natural gas network, comprising:
establishing a dynamic model of natural gas transmission according to the conservation equations, and reconstructing the dynamic model into the equations in a heat conduction equation form;
transforming the heat conduction equation into a partial differential equation set with homogeneous boundary conditions using a method of variation of parameters, and constructing a general analytical expression for the partial differential equation set under constant pressure boundaries;
constructing, based on superposition principle, practical analytical expressions for the dynamic model of natural gas transmission;
constructing, according to system topologies and time-invariant, and causal properties of a natural gas system, a natural gas network equivalent model based on the practical analytical expressions; and
determining pressure at load nodes and mass flows at source nodes based on the natural gas network equivalent model.

### Claim 2 (depends on 1)

The analytical method for dynamic analysis of a natural gas network according to claim 1, wherein the establishing a dynamic model of natural gas transmission according to the conservation equations comprises the following steps:
establishing a dynamic model of natural gas transmission in a pipeline, comprising a mass conservation equation and a momentum conservation equation as below:
∂
p
x
,
t
∂
t
+
c
2
S
⁢
∂
q
x
,
t
∂
x
=
0
,
∂
p
x
,
t
∂
x
+
λ
⁢
wq
x
,
t
2
⁢
SD
=
0
,
(
1
)
where x and t represent space and time, respectively, and c represents a speed of sound, taken as 340 m/s; px,t represents a pressure of natural gas, Pa; qx,t represents mass flow of natural gas, kg/s; w represents an average velocity of natural gas, m/s; D represents an inner diameter of the pipeline, m; S represents a cross-sectional area of the pipeline, m2; and λ represents a resistance coefficient of the pipeline; and
setting initial conditions and boundary conditions in the dynamic analysis of the natural gas system as follows:
p
x
,
0
=
φ
p
,
x
,
q
x
,
0
=
φ
q
,
x
,
p
0
,
t
,
q
L
,
t
,
(
2
)
where px,0 and qx,0 represent distributions of pressure and mass flow in the pipeline at moment 0 respectively, namely, the initial conditions; φp,x and φq,x represent initial condition functions of pressure and mass flow, respectively; L represents a length of the pipeline, m; p0,t represents a time-series of pressure at an inlet of the pipeline, namely, pressure boundary; and qL,t represents a time-series function of mass flow at an outlet of the pipeline, namely, flow boundary.

### Claim 3 (depends on 2)

The analytical method for dynamic analysis of a natural gas network according to claim 2, wherein the reconstructing the dynamic gas flow model into the equations in a heat conduction equation form comprises the following steps:
deriving a partial derivative for x from a second equation in Eq. before substituting into a first equation in Eq. (1) to obtain a heat conduction form equation as follows:
∂
q
x
,
t
∂
x
=
-
2
⁢
DS
λ
⁢
w
⁢
∂
2
p
x
,
t
∂
x
2
⇒
⁢
{
∂
p
x
,
t
∂
t
=
α
⁢
∂
2
p
x
,
t
∂
x
2
α
=
2
⁢
Dc
2
λ
⁢
w
,
(
3
)
where α represents a coefficient term in the heat conduction form equation; accordingly, the partial derivative of pipeline outlet pressure for x can be obtained through the following equation and is defined as mt:
∂
p
∂
x
|
x
=
L
=
-
λ
⁢
w
2
⁢
DS
⁢
q
L
,
t
=
m
t
;
(
4
)
and
Eq. (1) being converted to:
{
∂
p
x
,
t
∂
f
=
α
⁢
∂
2
p
x
,
t
∂
x
2
,
∂
p
x
,
t
∂
x
|
x
=
L
=
m
t
p
⁢
(
0
,
t
)
=
p
0
,
t
,
p
⁢
(
x
,
0
)
=
φ
p
,
x
.
(
5
)

### Claim 4 (depends on 3)

The analytical method for dynamic analysis of a natural gas network according to claim 3,
wherein the transforming the heat conduction equation into a partial differential equation set with homogeneous boundary conditions using a method of variation of parameters comprises the following steps:
taking, by p0,t, a constant value ψp,1 in response to the pressure boundary being constant; introducing an intermediate variable ux,t and reconstructing px,t as follows:
p
x
,
t
=
u
x
,
t
+
x
⁢
(
∂
p
∂
x
|
x
=
L
)
+
p
o
,
t
;
(
6
)
substituting x=0 and t=0 into Eq. (6) to obtain initial conditions and a set of boundary conditions for ux,t, respectively:
{
u
x
,
0
=
p
x
,
0
-
x
⁢
(
∂
p
∂
x
|
x
=
L
)
|
t
=
0
-
p
0
,
0
p
0
,
t
=
u
0
,
t
+
p
0
,
t
⇒
u
0
,
t
=
0
;
(
7
)
deriving a partial derivative for x on both sides of Eq. (6) at x=L to obtain a further set of boundary conditions for ux,t, expressed as follows:
∂
p
x
,
t
∂
x
=
∂
u
x
,
t
∂
x
+
∂
p
x
,
t
∂
x
❘
"\[RightBracketingBar]"
x
=
L
⇒
∂
u
x
,
t
∂
x
|
x
=
L
=
0
;
(
8
)
deriving a second-order partial derivative for x and a first-order partial derivative for t on both sides of Eq. (6), respectively:
∂
2
p
x
,
t
∂
x
2
=
∂
u
x
,
t
2
∂
x
2
,
∂
p
x
,
t
∂
t
=
∂
u
x
,
t
∂
t
+
x
⁢
∂
∂
t
(
∂
p
x
,
t
∂
t
❘
"\[LeftBracketingBar]"
x
=
L
)
;
(
9
)
and
substituting Eq. (9) into Eq. (5), with Eq. (5) re-expressing a partial differential equation set with homogeneous boundary conditions as follows:
{
∂
u
x
,
t
∂
t
=
α
⁢
∂
2
u
x
,
t
∂
x
2
+
f
x
,
t
u
x
,
0
=
P
x
,
0
-
x
(
∂
p
∂
x
❘
"\[RightBracketingBar]"
x
=
L
)
❘
"\[RightBracketingBar]"
t
=
0
-
p
0
,
0
,
∂
u
x
,
t
∂
x
❘
"\[RightBracketingBar]"
x
=
L
=
0
u
0
,
t
=
0
,
f
x
,
t
=
-
x
⁢
∂
∂
t
(
∂
p
∂
x
❘
"\[RightBracketingBar]"
x
=
L
)
=
-
x
⁢
∂
m
t
∂
t
.
(
10
)

### Claim 5 (depends on 4)

The analytical method for dynamic analysis of a natural gas network according to claim 4, wherein the constructing a general analytical expression for the partial differential equation set under constant pressure boundaries comprises the following steps:
defining ux,t* as a solution of a homogeneous problem of Eq. (10), and solving the homogeneous problem as follows:
u
x
,
t
*
=
T
0
(
t
)
⁢
∑
n
=
1
∞
⁢
k
2
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
,
(
11
)
where k2 represents a coefficient term; T0(t) represents a function component about time in ux,t*; n represents the number of Fourier components; referring to a form of ux,t* a form of ux,t in Eq. (10) is assumed as follows:
u
x
,
t
=
∑
n
=
1
∞
⁢
u
n
,
t
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
,
(
12
)
where un,t represents a function component about time in ux,t, fx,t in Eq. (10) is expanded according to a base in Eq. (11):
f
x
,
t
=
∑
n
=
I
∞
⁢
f
n
,
t
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
,
(
13
)
f
n
,
t
=
2
L
⁢
∫
0
L
f
x
,
t
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
⁢
dx
,
(
14
)
where fn,t represents a function component about time in fx,t, obtained by inverse Fourier transformation shown in Eq. (14); Eq. (12) to (14) are substituted into a first equation in Eq. (10):
{
∂
u
n
,
t
∂
t
+
g
n
⁢
u
n
,
t
=
f
n
,
t
g
n
=
(
2
⁢
n
-
1
)
2
⁢
π
2
⁢
a
4
⁢
L
2
u
n
,
0
=
2
L
⁢
∫
0
L
u
x
,
0
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
⁢
d
⁢
x
,
(
15
)
where gn represents a coefficient term in an n-th nonhomogeneous partial differential, un,0 represents a value of un,t when t=0; Eq. (15) is solved to obtain analytical expressions for un,t and ux,t, respectively as follows:
u
n
,
t
=
e
-
g
n
⁢
t
⁢
{
u
n
,
0
+
∫
0
t
f
n
,
τ
⁢
e
g
n
⁢
τ
⁢
dτ
}
,
(
16
)
u
x
,
t
=
∑
n
=
I
∞
⁢
e
-
g
n
⁢
t
⁢
{
u
n
,
0
+
∫
0
t
f
n
,
τ
⁢
e
g
n
⁢
τ
⁢
d
⁢
τ
}
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
;
(
17
)
substituting Eq. (17) into Eq. (6) to obtain an analytical expression for px,t at constant pressure:
p
x
,
t
=
∑
n
=
1
∞
⁢
e
-
g
n
t
⁢
{
u
n
,
0
+
∫
0
t
f
n
,
τ
⁢
e
g
n
τ
⁢
d
⁢
τ
}
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
-
λ
⁢
w
⁢
x
2
⁢
D
⁢
s
⁢
q
L
,
t
+
p
0
,
t
;
(
18
)
and
substituting Eq. (18) into a second equation in Eq. (1) to obtain an analytical expression for qx,t at constant pressure:
q
x
,
t
=
-
2
⁢
DS
λ
⁢
w
⁢
∂
p
x
,
t
∂
x
=
q
L
,
t
+
{
∑
n
=
1
∞
⁢
(
1
-
2
⁢
n
)
⁢
π
⁢
D
⁢
S
λ
⁢
w
⁢
L
⁢
e
g
n
t
⁢
{
u
n
,
0
+
∫
0
t
f
n
,
τ
⁢
e
g
n
τ
⁢
d
⁢
τ
}
⁢
cos
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
}
,
(
19
)
where p0,t represents a constant-voltage boundary, taking a constant value; Eq. (18) and Eq. (19) are general analytical expressions for the partial differential equation set of the dynamic model for natural gas transmission under constant pressure boundaries.

### Claim 6 (depends on 1)

The analytical method for dynamic analysis of a natural gas network according to claim 1, wherein the constructing, based on superposition principle, practical analytical expressions for the dynamic model of natural gas transmission comprises the following steps:
reconstructing the pressure boundary, expressed as follows:
{
ψ
p
=
[
ψ
p
,
1
,
ψ
p
,
2
,
…
,
ψ
p
,
Nt
]
ψ
p
,
i
=
{
p
0
,
i
i
=
1
p
0
,
i
-
p
0
,
i
-
1
i
>
1
,
t
i
=
i
⁢
Δ
⁢
t
,
(
20
)
where Nt represents a length of the boundary condition, and Δt is a time step; ψp represents a new boundary condition corresponding to a pressure increment, and an element ψp,i represents an increment of a pressure at moment ti for a pressure at moment ti−1;
decomposing, according to superposition principle of natural gas transmission, the pressures px,t as follows:
P
x
,
t
=
∑
k
=
1
i
⁢
χ
k
,
x
,
t
-
t
k
-
1
⁢
1
≤
i
≤
N
⁢
t
,
t
k
-
1
≤
t
≤
t
i
,
(
21
)
where each component χk,x,t satisfies:
{
∂
χ
k
,
t
-
t
k
-
1
∂
t
=
α
⁢
∂
2
χ
k
,
x
,
t
k
-
1
∂
x
2
,
t
>
t
k
-
1
,
0
<
x
<
L
∂
χ
k
,
x
,
t
-
t
k
-
1
∂
t
❘
"\[RightBracketingBar]"
x
=
L
=
{
m
t
k
=
1
0
k
>
1
,
t
>
t
k
-
1
χ
k
,
x
,
t
-
t
k
-
1
❘
"\[RightBracketingBar]"
x
=
0
=
ψ
p
,
k
,
t
>
t
k
-
1
χ
k
,
x
,
t
-
t
k
-
1
❘
"\[RightBracketingBar]"
t
=
t
k
-
1
=
{
φ
p
,
x
k
=
1
0
k
>
1
,
0
≤
x
≤
L
,
(
22
)
when k=1, a form of χk,x,t is consistent with Eq. (18), namely:
χ
k
,
x
,
t
-
t
k
-
1
=
∑
n
=
1
∞
⁢
e
-
g
n
t
⁢
{
u
n
,
0
+
∫
0
t
f
n
,
τ
⁢
e
g
n
τ
⁢
d
⁢
τ
}
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
+
x
⁢
m
t
+
ψ
p
,
1
k
=
1
,
t
>
t
k
-
1
,
(
23
)
when k>1, a form of χk,x,t is as follows:
χ
k
,
x
,
t
-
t
k
-
1
=
ψ
p
,
k
k
>
1
,
t
>
t
k
-
1
;
(
24
)
calculating the initial conditions according to steady-state energy flow in practice, expressed as follows:
φ
p
,
k
=
K
1
⁢
x
+
K
2
,
φ
q
,
x
=
K
3
,
m
0
=
-
λ
⁢
w
⁢
φ
q
,
L
2
⁢
DS
,
(
25
)
where K1, K2, and K3 represent constant coefficients in the initial conditions; m0 represents a value of mt at t=0; and φq,L represents a value of the initial conditions of mass flow at x=L, namely, the mass flow at an outlet of the pipeline at t=0;
expressing mt as a segmented step function as follows:
m
t
=
∑
i
=
1
Nt
m
i
(
δ
t
-
t
i
-
δ
t
-
t
i
-
1
)
⁢
t
>
0
;
(
26
)
substituting fx,t in Eq. (10) and fn,t in Eq. (14) into Eq. (16) as follows:
=
∫
0
t
(
-
2
L
⁢
∫
0
L
x
⁢
∂
m
τ
∂
τ
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
⁢
dx
)
⁢
e
g
n
⁢
τ
⁢
d
⁢
τ
∫
0
t
f
n
,
π
⁢
e
g
n
⁢
τ
⁢
d
⁢
τ
=
-
2
L
⁢
∫
0
L
x
⁢
sin
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
⁢
dx
⁢
∫
0
t
∂
m
τ
∂
τ
⁢
e
g
n
⁢
τ
⁢
d
⁢
τ
8
⁢
L
⁡
(
-
1
)
n
(
2
⁢
n
-
1
)
2
⁢
π
2
⁢
∫
0
t
∂
m
τ
∂
τ
⁢
e
g
n
⁢
τ
⁢
d
⁢
τ
;
(
27
)
modifying an integral formula in Eq. (27) as follows:
∂
m
τ
∂
τ
⁢
e
g
n
⁢
τ
=
∂
(
m
τ
⁢
e
g
n
⁢
τ
)
∂
τ
-
g
n
⁢
m
τ
⁢
e
g
n
⁢
τ
,
(
28
)
g
n
⁢
∫
0
t
k
m
τ
⁢
e
g
n
⁢
τ
⁢
d
⁢
τ
=
∑
i
=
1
k
m
i
(
e
g
n
⁢
t
i
-
e
g
n
⁢
t
i
-
1
)
;
(
29
)
substituting Eq. (28) and Eq. (29) into Eq. (27) as follows:
1
e
g
n
⁢
t
k
⁢
∫
0
t
k
f
n
,
τ
⁢
e
g
n
⁢
τ
⁢
d
⁢
τ
=
-
E
1
⁢
n
(
m
k
-
m
0
e
g
n
⁢
t
k
-
∑
i
=
1
k
m
i
⁢
e
g
n
⁢
Δ
⁢
t
-
1
e
g
n
(
t
k
-
t
i
-
1
)
)
,
(
30
)
E
1
⁢
n
=
8
⁢
L
⁡
(
-
1
)
n
-
1
(
2
⁢
n
-
1
)
2
⁢
π
2
;
(
31
)
and
substituting Eq. (30) into Eq. (18) and Eq. (19), to obtain the practical analytical expressions for the dynamic model of natural gas transmission as follows:
p
x
,
k
=
∑
n
=
1
∞
{
u
n
,
0
e
g
n
⁢
t
k
-
E
1
⁢
n
(
m
k
-
m
0
e
g
n
⁢
t
k
-
∑
i
=
1
k
m
i
⁢
e
g
n
⁢
Δ
⁢
t
-
1
e
g
n
(
t
k
-
t
i
-
1
)
)
}
⁢
sin
⁢
(
2
⁢
n
-
1
)
2
⁢
L
-
λ
⁢
wx
2
⁢
DS
⁢
q
L
,
k
+
p
0
,
k
,
(
32
)
q
x
,
k
=
q
L
,
k
+
{
∑
n
=
1
∞
(
1
-
2
⁢
n
)
⁢
π
⁢
DS
λ
⁢
wL
⁢
{
u
n
,
0
e
g
n
⁢
t
k
-
E
1
⁢
n
(
m
k
-
m
0
e
g
n
⁢
t
k
-
∑
i
=
1
k
m
i
⁢
e
g
n
⁢
Δ
⁢
t
-
1
e
g
n
(
t
k
-
t
i
-
1
)
)
}
⁢
cos
⁢
(
2
⁢
n
-
1
)
⁢
π
⁢
x
2
⁢
L
}
.
(
33
)

### Claim 7 (depends on 1)

The analytical method for dynamic analysis of a natural gas network according to claim 1,
wherein the constructing, according to system topologies, and time-invariant, and causal properties of a natural gas system, a natural gas network equivalent model based on the practical analytical expressions comprises the following steps:
substituting x=L and x=0 into Eq. (32) and Eq. (33), respectively, to obtain pressure at the outlet and flow at the inlet of the pipeline at arbitrary tj as follows:
p
L
,
j
=
∑
n
=
1
∞
e
-
g
n
⁢
t
j
⁢
u
n
,
0
(
-
1
)
n
-
1
-
λ
⁢
wL
2
⁢
DS
⁢
q
L
,
j
+
p
0
,
j
+
∑
n
=
1
∞
λ
⁢
wE
1
⁢
n
2
⁢
DS
⁢
(
-
1
)
n
-
1
⁢
(
q
L
,
j
-
q
L
,
0
e
g
n
⁢
t
j
-
∑
t
=
1
j
q
L
,
i
⁢
e
g
n
⁢
Δ
⁢
t
-
1
e
g
n
(
j
-
i
+
1
)
⁢
Δ
⁢
t
)
,
(
34
)
q
0
,
j
=
q
L
,
j
+
DS
⁢
π
λ
⁢
wL
⁢
∑
n
-
1
∞
(
1
-
2
⁢
n
)
⁢
u
n
,
0
⁢
e
-
g
n
⁢
t
j
+
π
2
⁢
L
⁢
∑
n
-
1
∞
(
1
-
2
⁢
n
)
⁢
E
1
⁢
n
(
q
L
,
j
-
q
L
,
0
e
g
n
⁢
t
j
-
∑
i
=
1
j
q
L
,
i
⁢
e
g
n
⁢
Δ
⁢
t
-
1
e
g
n
(
j
-
i
+
1
)
⁢
Δ
⁢
t
)
;
(
35
)
reorganizing Eq. (34) and Eq. (35) as follows:
[
p
L
q
0
]
=
[
H
1
H
2
H
3
H
4
]
[
p
0
q
L
]
+
[
γ
p
γ
q
]
,
(
36
)
where pL and p0 represent a pressure vector at the outlet and inlet of the pipeline, respectively; qL and q0 represent a flow vector at the outlet and inlet of the pipeline, respectively; H1 to H4 represent constant coefficient matrices with the dimension of Nt×Nt, and γp and γq represent a constant coefficient vector of pressure and mass flow with the dimension of Nt×1, respectively, determined by the initial conditions; the elements may be represented as follows:
{
p
0
=
[
p
0
,
1
,
⋯
,
p
0
,
Nt
]
,
p
L
=
[
p
L
,
1
,
⋯
,
p
L
,
Nt
]
q
L
=
[
q
L
,
1
,
⋯
,
q
L
,
Nt
]
,
q
0
=
[
q
0
,
1
,
⋯
,
q
0
,
Nt
]
,
(
37
)
γ
p
,
j
=
∑
n
=
1
∞
(
u
n
,
0
+
E
1
⁢
n
⁢
m
0
)
⁢
(
-
1
)
n
-
1
⁢
e
-
g
n
⁢
t
j
,
(
38
)
γ
q
,
j
=
DS
⁢
π
λ
⁢
wL
⁢
∑
n
-
1
∞
(
1
-
2
⁢
n
)
⁢
(
u
n
,
0
+
E
1
⁢
n
⁢
m
0
)
⁢
e
-
g
n
⁢
t
j
,
(
39
)
H
1
=
1
,
H
3
=
0
,
H
2.
ji
i
>
j
=
H
4.
ji
i
>
j
=
0
,
(
40
)
H
2.
ji
i
=
j
=
λ
⁢
w
2
⁢
DS
⁢
(
∑
n
=
1
∞
E
1
⁢
n
(
-
1
)
n
-
1
⁢
e
-
g
n
⁢
Δ
⁢
t
-
L
)
,
(
41
)
H
2.
ji
(
i
>
j
)
=
λ
⁢
w
2
⁢
DS
⁢
(
∑
n
=
1
∞
E
1
⁢
n
(
-
1
)
n
-
1
⁢
1
-
e
-
g
n
⁢
Δ
⁢
t
e
g
n
(
j
-
i
+
1
)
⁢
Δ
⁢
t
)
,
(
42
)
H
4.
ji
(
i
=
j
)
=
π
2
⁢
L
⁢
∑
n
=
1
∞
E
1
⁢
n
(
1
-
2
⁢
n
)
⁢
e
-
g
n
⁢
Δ
⁢
t
+
1
,
(
43
)
H
4.
ji
(
i
>
j
)
=
π
2
⁢
L
⁢
∑
n
=
1
∞
E
1
⁢
n
(
1
-
2
⁢
n
)
⁢
1
-
e
-
g
n
⁢
Δ
⁢
t
e
g
n
(
j
-
i
+
1
)
⁢
Δ
⁢
t
;
(
44
)
defining the number of nodes and pipelines in the natural gas system as Ng and Nb, respectively, and defining an identity matrix and a zero matrix as 1 and 0, respectively; introducing incidence matrices Ain, Aout, Anb1, and Anb2 to describe topological properties in the natural gas system, wherein
Ain comprises Ng×Nb sub-matrices for associating flow at end of a branch with flow at a node, and in response to a sub-matrix of an i-th row and j-th column being 1, the flow flows from the end of the branch j into the node i, otherwise, the sub-matrix is 0; Aout comprises Ng×Nb sub-matrices for associating flow at head end of a branch with flow at a node, and in response to a sub-matrix of an i-th row and j-th column being 1, the flow flows from the node i into the head end of the branch j, otherwise, the sub-matrix is 0; Anb1 comprises Nb×Ng sub-matrices for associating pressure at end of a branch with pressure at a node, and in response to a sub-matrix of an i-th row and j-th column being 1, pressure at the node j is equal to pressure at the end of the branch i, otherwise, the sub-matrix is 0; Anb2 comprises Nb×Ng sub-matrices for associating pressure at a head end of a branch with pressure at a node, and in response to a sub-matrix of an i-th row and j-th column being 1, pressure at the node j is equal to pressure at the head end of the branch i, otherwise, the sub-matrix is 0; particularly, in response to the node j being a compressor node with a pressure ratio of Kcp and the node j being connected to the head end of the branch i, the sub-matrix of the i-th row and jth column is Kcp×1;
constructing, based on the incidence matrices, equations of mass conservation at nodes and pressure continuity between nodes and pipelines, expressed as follows:
A
in
⁢
q
L
=
A
out
⁢
q
0
=
qn
,
(
45
)
p
L
=
A
nb
⁢
1
⁢
pn
,
p
0
=
A
nb
⁢
2
⁢
pn
,
(
46
)
where pn and qn represent a pressure and a vector at the node, respectively; pL, p0, qL, and q0 are all vectors of Nt×1; the sub-matrices 1 and 0 in coefficient matrices Ain, Aout, Anb1, and Aout2 are all Nt×Nt matrices;
substituting Eq. (46) into a first row of Eq. (36) as follows:
q
L
=
H
2
-
1
(
(
A
nb
⁢
1
-
A
nb
⁢
2
)
⁢
pn
-
γ
p
)
=
H
a
⁢
1
⁢
pn
+
H
a
⁢
2
,
(
47
)
where Ha1 and Ha2 represent a constant coefficient matrix and a vector describing mass flow at an outlet of a pipeline, respectively; Eq. (46) and Eq. (47) are substituted into a second row of Eq. (36) as follows:
q
0
=
H
4
(
H
a
⁢
1
⁢
pn
+
H
a
⁢
2
)
+
γ
q
=
H
a
3
⁢
pn
+
H
a
4
,
(
48
)
where Ha3 and Ha4 represent a constant coefficient matrix and a vector describing mass flow at an inlet of the pipeline, respectively; Eq. (47) and Eq. (48) are substituted into Eq. (45) to associate pressure and mass flow at the node as follows:
qn
=
(
A
in
⁢
H
a
⁢
1
-
A
out
⁢
H
a
⁢
3
)
⁢
pn
+
(
A
in
⁢
H
a
⁢
2
-
A
out
⁢
H
a
⁢
4
)
=
H
a
⁢
5
⁢
pn
+
H
a
⁢
6
,
(
49
)
where Ha5 and Ha6 are a constant coefficient matrix and a vector describing a relationship between pressure and mass flow at a node, respectively; subscripts sr and ns are defined as a variable identification of a gas source node and a gas load node, respectively, and Eq. (49) is expanded to obtain the natural gas network equivalent model based on the practical analytical expressions:
[
qn
sr
qn
ns
]
=
[
H
a
⁢
51
H
a
⁢
52
H
a
⁢
53
H
a
⁢
54
]
[
pn
sr
pn
ns
]
+
[
H
a
⁢
61
H
a
⁢
62
]
,
(
50
)
where Ha51, Ha52, Ha53, and Ha54 are sub-matrices in Ha5, and Ha61 and Ha62 are sub-vectors in Ha6; pnsr and pnns are a pressure vector at a source node and a load node, respectively; qnsr and qnns are a mass flow vector at the source node and the load node, respectively; the mass flow at the source node and the pressure at the load node in the natural gas system are fully expressed as functions of the boundary conditions using matrix transformation, namely, a simplified method of natural gas network equivalent model is as follows:
pn
ns
=
H
a
⁢
54
-
1
(
qn
ns
-
H
a
⁢
53
⁢
pn
sr
-
H
a
⁢
62
)
,
(
51
)
qn
sr
=
H
a
⁢
51
⁢
pn
sr
+
H
a
⁢
52
⁢
pn
ns
+
H
a
⁢
61
.
(
52
)
