US 2020/0370408 A1 · METHOD AND SYSTEM FOR GUARANTEEING SAFETY OF OFFSHORE OIL WELL CONTROL EQUIPMENT

Abstract

The present disclosure belongs to the field of marine engineering, and in particular relates to a method and system for guaranteeing the safety of offshore oil well control equipment. The method comprises: identifying the state of a main structure of the offshore oil well control equipment, identifying the state of a hydraulic control unit of offshore oil well control equipment, identifying the state of an electronic control unit of offshore oil well control equipment, predicting the state of the offshore oil well control equipment and making a real-time decision based on existing information; and the system for guaranteeing the safety of the offshore oil well control equipment comprises a state identification subsystem of a main structure, a state identification subsystem of the hydraulic control unit, a state identification subsystem of the electronic control unit and a state prediction and real-time decision subsystem.

A method for guaranteeing the safety of an offshore oil well control equipment, comprising five major steps:

identifying the state of a main structure of the offshore oil well control equipment, identifying the state of a hydraulic control unit of the offshore oil well control equipment, identifying the state of a electronic control unit of the offshore oil well control equipment, predicting the state of the offshore oil well control equipment and making a real-time decision based on existing information;
wherein the identifying the state of the main structure of the offshore oil well control equipment specifically comprises:
S101:
obtaining states of nodes V11 V12 . . . V1a A11 A12 . . . A1b C11 C12 . . . C1c at a characteristic value acquisition layer of the main structure of the offshore oil well control equipment by signal processing, the corresponding states being V11(m11) V12(m12) . . . V1a(m1a) A11(s11) A12(s12) . . . A1b(s1b) C11(w11) C12(w12) . . . C1c(w1c);
S102:
obtaining failure probabilities of corresponding components based on an established Bayesian network, wherein for a component F1x(x=1,2, . . . m), a method for solving the fault probabilities is as follows:
P

(
F
1

x
)
=
∑
i
=
1
n

∑
j
=
1
y
1

i

P

(
F
1

x
|
I
1

i

(
j
)
)

P

(
I
1

i

(
j
)
|
V
11

(
m
11
)
,
…


V
1

a

(
m
1

a
)
,
A
11

(
s
11
)


…


A
1


b

(
s
1

b
)
,
C
11

(
w
11
)


…


C
1

c

(
w
1

c
)
)
wherein, P(F1x|I1i(j)) represents the component indicated by the x-th node in a state identification layer of the main structure of the offshore oil well control equipment being in a fault state when the i-th node in an intermediate inference layer of the main structure of the offshore oil well control equipment is in a state j;
P(I1i(j)|V11(m11) . . . V1a(m1a),A11(s11) . . . A1b(s1b), C11(w11) . . . C1c(w1c)) represents the probability at which the i-th node in the intermediate inference layer of the main structure of the offshore oil well control equipment is in the j-th state when the nodes V11 V12 . . . V1a A11 A12 . . . A1b C11 C12 . . . C1c in the characteristic value acquisition layer of the main structure of the offshore oil well control equipment are in the states of V11(m11) V12(m12) . . . V1a(m1a) A11(s11) A12(s12) . . . A1b(s1b) C11(w11) C12(w12) . . . C1c(w1c);
y1i represents the number of states possessed by the i-th node in the intermediate inference layer of the main structure of the offshore oil well control equipment;
n represents the total number of nodes in the intermediate inference layer of the main structure of the offshore oil well control equipment;
and a probability is obtained by existing database through statistics;
S103:
judging whether the respective component is in a fault state according to the following rule:
(1) the component is considered to be in a fault state when the fault probability is greater than 70%;
(2) the component is considered to have a potential fault when the fault probability is between 50% and 70%;
(3) the component is considered to be normal when the fault probability is less than 50%;
the identifying the state of the hydraulic control unit of the offshore oil well control equipment specifically comprises:
S201:
obtaining corresponding states of nodes A21 A22 . . . A2d PT21 P22 . . . PT2e at a characteristic value acquisition layer of the hydraulic control unit of the offshore oil well control equipment by signal processing, the corresponding states being A21(s21) A22(s22) . . . A2d(s2d) P21(w21) PT22(w22) . . . PT2e(w2e);
S202:
obtaining fault probabilities of corresponding components based on an established Bayesian network, wherein for a component F2x(x=1,2, . . . q), a method for solving the fault probabilities by changing p to q is as follows:
P

(
F
2

x
)
=
∑
i
=
1
q

∑
j
=
1
y
2

i

P

(
F
2

x
|
I
2

i

(
j
)
)

P

(
I
2

i

(
j
)
|
A
21

(
s
21
)


…


A
2

d

(
s
2

d
)
,
PT
21

(
w
21
)


…


PT
2

e

(
w
2

e
)
)
wherein, P(F2x|I2i(j)) represents the component indicated by the x-th node in a state identification layer of the hydraulic control unit of the offshore oil well control equipment being in a fault state when the i-th node in an intermediate inference layer of the hydraulic control unit of the offshore oil well control equipment is in a state j;
P(I2i(j)|A21(s21) . . . A2d (s2d), PT21(w21) . . . PT2e (w2e) represents the probability at which the i-th node in the characteristic value acquisition layer of the hydraulic control unit of the offshore oil well control equipment is in the j-th state when the nodes A21 . . . A2d PT21 . . . PT2e in the characteristic value acquisition layer of the hydraulic control unit of the offshore oil well control equipment are in the states of A21(s21) . . . A2d(s2d) PT21(w21) . . . PT2e(w2e);
y2i represents the number of states possessed by the i-th node in the intermediate inference layer of the hydraulic control unit of the offshore oil well control equipment;
q represents the total number of nodes in the intermediate inference layer of the hydraulic control unit of the offshore oil well control equipment;
and the conditional probability is obtained by existing database through statistics;
S203:
judging whether the respective component is in a fault state according to the following rule:
(1) the component is considered to be in a fault state when the fault probability is greater than 70%;
(2) the component is considered to have a potential fault when the fault probability is between 50% and 70%;
(3) the component is considered to be normal when the fault probability is less than 50%;
the identifying the state of the electronic control unit of offshore oil well control equipment specifically comprises:
S301:
obtaining corresponding states of control information layers C31 C32 . . . C3f of the electronic control unit of offshore oil well control equipment by signal processing, the corresponding states being C31(w31) C32 (w32) . . . C3f(w3f);
S302:
obtaining fault probabilities of corresponding components based on an established Bayesian network, wherein for a component F3x(x=1,2, . . . r), a method for solving the fault probabilities is as follows:
P

(
F
3

x
)
=
∑
i
=
1
s

∑
j
=
1
y
3

i

P

(
F
3

x
|
I
3

i

(
j
)
)

P

(
I
3

i

(
j
)
|
C
3

1

(
w
3

1
)
,
C
3

2

(
w
3

2
)


…


C
3

f

(
w
3

f
)
)
wherein, P(F3x|I3i(j)) represents the component indicated by the x-th node in a state identification layer of the electronic control unit of offshore oil well control equipment being in a fault state when the i-th node in an intermediate inference layer of the electronic control unit of offshore oil well control equipment is in a state j;
P(I3i(j)|C31(w31),C32(w32) . . . C3f(w3f)) represents the probability at which the i-th node in the intermediate inference layer of the electronic control unit of offshore oil well control equipment is in the j-th state when the nodes in the control information layer C31 C32 . . . C3f of the electronic control unit of offshore oil well control equipment are in the states of C31(w31) C32(w32) . . . C3f(w3f);
y3i represents the number of states possessed by the i-th node in the intermediate inference layer of the electronic control unit of offshore oil well control equipment;
s represents the total number of nodes in the intermediate inference layer of the electronic control unit of offshore oil well control equipment;
and the conditional probability is obtained by the existing database through statistics;
S303:
judging whether the respective component is in a fault state according to the following rule:
(1) the component is considered to be in a fault state when the fault probability is greater than 70%;
(2) the component is considered to have a potential fault when the fault probability is between 50% and 70%;
(3) the component is considered to be normal when the fault probability is less than 50%;
the predicting the state of the offshore oil well control equipment specifically comprises:
S401:
obtaining corresponding states of current state layers F11 F12 . . . F1m F21 F22 . . . F2p F41 F32 . . . F3r of the offshore oil well control equipment by collecting state identification signals, the corresponding states being F11(z41) F12(z42) . . . F1m(z4m) F21(u41) F22(u42) . . . F2p(u4p) F31(v41) F32(v42) . . . F3r(v4r);
S402:
obtaining predicted fault probabilities of corresponding components based on an established Bayesian network, wherein for a component P4x(x=1,2, . . . t), a method for solving the predicted fault probabilities is as follows:
P

(
P
4

x
)
=
∑
i
=
1
u

∑
j
=
1
y
4

i

P

(
P
4

x
|
I
4

i

(
j
)
)

P

(
I
4

i

(
j
)
|
F
1

1

(
z
4

1
)
,
…


F
1

m

(
z
4

m
)
,
F
2

1

(
u
4

1
)
,
…


F
2

p

(
u
4

p
)
,
F
3

1

(
v
4

1
)
,
…


F
3

r

(
v
4

r
)
)
wherein, P(P4x|I4i(j)) represents the probability at which the component indicated by the x-th node in a state prediction layer of the offshore oil well control equipment will have a fault within a future certain time when the i-th node of a state prediction intermediate inference layer of the offshore oil well control equipment is in a state j;
P(I4i(j)|F11(z41), . . . F1m(z4m), F21(u41), . . . F2p(u4p), F31(v41), . . . F3r(v4r) represents the probability at which the i-th node of the state prediction intermediate inference layer of the offshore oil well control equipment is in the j-th state when the nodes F11 F12 . . . F1m F21 F22 . . . F2p F31 F32 . . . F3r in the current state layers of the offshore oil well control equipment are in states of F11(z41) F12(z42) . . . F1m(z4m) F21(u41) F22(u42) . . . F2p(u4p) F31(v41) F32(v42) . . . F3r(v4r);
y4i represents the number of states possessed by the i-th node in the state prediction intermediate inference layer of the offshore oil well control equipment;
u represents the total number of nodes in the state prediction intermediate inference layer of the electronic control unit of offshore oil well control equipment;
and the conditional probability is obtained by expert decision;
S403:
judging whether the respective component will have a fault within 24 hours according to the following rule;
(1) the component is considered to have high probability of the fault occurrence within 24 hours when the predicted fault probability is greater than 70%;
and
(2) the component is considered to have small probability of the fault occurrence within 24 hours when the predicted fault probability is less than 70%;
the making a real-time decision based on existing information specifically comprises:
S501:
judging whether a fault occurs currently, making panel display and performing shutdown for maintenance if there is the fault, and proceeding to the next step if not;
S502:
judging whether there is a predicted fault, allowing the equipment continue to run if there is no predicted fault, and making the panel display and judging whether the predicted fault is a major fault if yes, wherein a method for judging whether the predicted fault is a major fault is as follows:
(1) calculating the predicted fault weights by referring to the following formula:
εx=P(P4x)×γx
wherein, εx is a corresponding fault weight;
P (P4x) is the predicted fault probability of the x-th component in S402;
γx is corresponding weight of the x-th component in the system and obtained by expert scoring;
(2) considering the predicted fault to be a major fault if the fault weight εx is greater than 0.4;
S503:
performing the shutdown for maintenance if the predicted fault is the major fault, and proceeding to the next step if not;
and
S504:
repeating S501 to S504.

531 more elements

Method and system for guaranteeing safety of offshore oil well control equipment

Life of this patent

Method and system for guaranteeing safety of offshore oil well control equipment

Life of this patent

Abstract

Claims8