1. A leakage detection system of a coke dry quenching primary dedusting cooling system is characterized by comprising a data acquisition module, a leakage intelligent diagnosis module, a diagnosis information output module, a system leakage verification module and an expert self-learning module;

the data acquisition module comprises a water inlet pressure sensor PT01, a water inlet flow sensor F01, a water inlet flow regulating valve FV01, a backwater pressure sensor PT02, a backwater cut-off valve XZV01 and a backwater flow sensor F02;

a water inlet pressure sensor PT01, a water inlet flow sensor F01 and a water inlet flow regulating valve FV01 are arranged on a cooling water inlet pipeline of the primary coke dry quenching dust remover, and cooling inlet water sequentially passes through the water inlet pressure sensor PT01, the water inlet flow sensor F01 and the water inlet flow regulating valve FV01 and then enters the dust remover;

the return water pressure sensor PT02, the return water shut-off valve XZV01 and the return water flow sensor F02 are arranged on a cooling water outlet pipeline of the primary coke dry quenching dust remover, and cooling outlet water sequentially passes through the return water pressure sensor PT02, the return water shut-off valve XZV01 and the return water flow sensor F02 and then enters a return water pipeline;

the water inlet pressure sensor PT01, the water inlet flow sensor F01, the water inlet flow regulating valve FV01, the water return pressure sensor PT02, the water return cut-off valve XZV01 and the water return flow sensor F02 are all connected to the dry quenching control system;

the data acquisition module also comprises a dust collector bottom temperature sensor T01, and the dust collector bottom temperature sensor T01 is also connected to the dry quenching control system;

the data acquisition module further comprises a circulating gas H2 content sensor A01 and a circulating gas CO content sensor A02, the circulating gas H2 content sensor A01 and the circulating gas CO content sensor A02 are arranged on a rear end pipeline of a heat exchanger of the dry quenching circulating system, and the circulating gas H2 content sensor A01 and the circulating gas CO content sensor A02 are also connected into the dry quenching control system.

2. The method for detecting the leakage of the coke dry quenching primary dedusting cooling system as recited in claim 1, wherein the intelligent diagnosis method in the intelligent leakage diagnosis module comprises the following steps:

the leakage intelligent diagnosis module judgment condition comprises the following four inputs:

1) the difference F between the inlet flow and the outlet flow is F01-F02 overrun and the duration thereof is overrun;

2) the inlet-outlet pressure difference P is P01-P02 overrun and the duration thereof is overrun;

3) the hydrogen H2 content a01 of the recycle gas increased while the CO content a02 of the recycle gas increased;

4) the bottom temperature T01 of the dust remover is suddenly changed in the normal heat exchange state.

3. The method for detecting the leakage of the coke dry quenching primary dedusting cooling system as claimed in claim 2, wherein the judgment conditions of the inlet-outlet flow difference F ═ F01-F02 overrun and the duration overrun are as follows:

F＝F01-F02

i F I > delta F and t > TF

Wherein: δ F — reasonable deviation of flow difference;

t is the condition satisfying the holding time;

TF-minimum value of hold time;

f01 — inlet flow rate;

f02 — outlet flow;

the delta F and TF value initial values are set by an operator according to the measurement precision and experience of the instrument, and are acquired from an expert knowledge base in the later period.

4. The method for detecting the leakage of the coke dry quenching primary dedusting cooling system as claimed in claim 2, wherein the conditions for judging that the inlet-outlet pressure difference P is P01-P02 overrun and the duration time is overrun are as follows:

P＝P01-P02

p is delta P and t is TP

Wherein: δ P — reasonable deviation of differential pressure;

t is the condition satisfying the holding time;

TP-minimum value of retention time;

p01-inlet pressure;

p02 — outlet pressure;

the initial values of the delta P and TP values are set by the experience of operators, and are acquired from an expert knowledge base in the later period.

5. The method for detecting the leakage of the coke dry quenching primary dedusting cooling system as claimed in claim 2, wherein the judgment conditions for the increase of the hydrogen H2 content A01 of the circulating gas and the increase of the CO content A02 of the circulating gas are as follows:

if a leak condition occurs in the cooling system, water enters the circulating gas as water vapor, and the following reactions occur when the water vapor encounters hot coke in the coke dry quenching furnace:

C+H2O＝CO+H2----(1)

as shown in the formula (1), water gas is generated by the reaction of water vapor and coke, and the components are mainly CO and H2, so that the condition of water leakage can be indirectly known by observing the real-time values and the change rates of CO and H2 in the circulating gas;

the judgment conditions based on the H2 measurement are:

A_H>A_HS

wherein:

A_H-a comprehensive judgment value calculated from the H analysis data;

α₁-H2 analyzer data deviation weighting coefficients;

f_H(t) - - -H2 analyzer real-time data a 01;

A_Hh-an upper limit of the normal value of H2;

α₂-a rate of change weighting for H2 analyzer data;

-the rate of change of the H2 analyzer over a period of n;

f_H(t-n) -H2 analyzer data n time periods prior to the current time;

A_HS-H2 analyzing the data to calculate a synthetic judgment value set-point;

where α is₁、A_Hh、n、α₂And A_HSSetting a value initial value by experience of an operator, and acquiring the value initial value from an expert knowledge base in a later period;

the judgment conditions based on the CO measurement are:

A_CO>A_COS

wherein:

A_CO-a comprehensive judgment value calculated from the CO analysis data;

α₃-CO analyzer data deviation weighting coefficients;

f_co(t) - - -CO analyzer real time data a 02;

A_coh-upper limit of normal value of CO;

α₄-a rate of change weighting for CO analyzer data;

-rate of change of CO analyzer over time period n;

f_co(t-n) -data of the CO analyzer n time periods prior to the current time;

A_COS-a set value of a comprehensive judgment value calculated from the CO analysis data;

where α is₃、A_coh、n、α₄And A_COSThe initial value of the value is set by the experience of operators, and is acquired from the expert knowledge base in the later period.

6. The method for detecting the leakage of the coke dry quenching primary dedusting cooling system as claimed in claim 2, wherein the judgment condition that the bottom temperature T01 of the dust catcher is suddenly changed in the normal heat exchange state is as follows:

T₀₁<T_01S

wherein:

T₀₁-a comprehensive judgment value calculated from the analysis data of T01;

α - -T01 temperature drop coefficient;

f_T(T) - - -T01 real-time data;

-rate of change of T01 over a period of n;

f_T(T-n) - - -T01 data n time periods before the current time;

T_01S-a comprehensive judgment value set value calculated by T01;

here T_01SAnd the initial value of the value n is set by experience of operators, and is acquired from an expert knowledge base in the later period.

7. The system for detecting the leakage of the coke dry quenching primary dedusting cooling system as claimed in claim 1, wherein the diagnostic information output module is specifically:

after the expert system diagnoses that the system is leaked, the diagnosis information is sent to a picture of the control system to be displayed, the expert system participates in judgment manually, and if the diagnosis result of the expert system diagnosis is not verified, the expert system directly enters into the interlocking control operation; the operation reaction is timely, and the processing speed is improved; if the diagnosis result is required to be verified, the program enters a leakage system verification module;

interlock control operation: and the water inlet flow regulating valve FV01 is closed in an interlocking manner, the water supply control of the cooling system is stopped, and the backwater cut-off valve XZV01 is closed in an interlocking manner, so that the backwater backflow generated by a negative pressure system at the cooling sleeve is prevented from occurring, the corresponding discharging operation of the dust removal bin is not allowed by the interlocking, and the alarm output is given.

8. The system for detecting the leakage of the coke dry quenching primary dedusting cooling system as claimed in claim 1, wherein the system leakage verification module is used for rapidly cutting off a water inlet valve by closing a return valve of a cooling water pipe, observing the change condition of the pressure in the cooling water pipe, judging that the cooling system has the leakage condition if the pressure loss condition occurs in unit time, and quantitatively judging the leakage rate according to the pressure loss time; in order to improve the verification accuracy, the above operation needs to be repeated for many times.

9. The system for detecting the leakage of the coke dry quenching primary dedusting cooling system as claimed in claim 8, wherein the verification method of the system leakage verification module is as follows:

a) closing a backwater cut-off valve XZV-01;

b) judging that the backwater cut-off valve XZV-01 is completely closed;

c) the following operations were repeated N times:

1) rapidly closing the water inlet flow regulating valve FV-01;

2) judging that the valve inlet flow regulating valve FV-01 is completely closed;

3) observing the pressure change trend of the water outlet pressure sensor PT-02;

4) if the water outlet pressure sensor PT-02 is subjected to rapid pressure loss, continuing the next step 5, if the water outlet pressure sensor PT-02 is not subjected to rapid pressure loss, finishing the verification, and feeding back verification information to the expert self-learning system;

5) recording the pressure loss time;

6) opening a water inlet flow regulating valve FV-01 until a water outlet pressure sensor PT-02 is restored to be close to a pressure value before the water inlet flow regulating valve FV-01 is closed;

7) returning to 1 to restart, carrying out verification operation, and carrying out the next step to d after the system records N times;

d) calculating the average value Tavg of the pressure loss time of N times;

e) determining that a cooling system has a leak, and interlockingly controlling operation of the diagnostic information output module;

f) and (4) quantifying the leakage rate according to the length of the average value Tavg of the pressure loss time, and then sending the calculation result to an expert self-learning module.

10. The system for detecting the leakage of the coke dry quenching primary dedusting cooling system as recited in claim 1, wherein the input of the expert self-learning module is used for carrying out real-time comparison learning on the input information of the system leakage expert diagnostic system and the output information of the leakage verification system, and the learning result is sent to an expert knowledge base for diagnosing the system leakage expert diagnostic system model; the expert self-learning module calculates an expert knowledge result according to the diagnosis conclusion output by the intelligent leakage diagnosis module and the verification conclusion output by the system leakage verification module and the self-learning mechanism of the expert self-learning module, and stores the knowledge into the relational database by the contents of the diagnosis data and the diagnosis rules for future intelligent leakage diagnosis.

Background

With the improvement of the national requirement on environmental protection and the increasing maturity of the dry quenching process technology, the dry quenching production process is widely adopted in the coking field, even the full dry quenching production process is adopted in some advanced large-scale coking enterprises. In the process of a dry quenching process, red coke is loaded from the top of a dry quenching furnace, low-temperature inert gas is blown into a red coke layer of a cooling section of the dry quenching furnace by a circulating fan to absorb sensible heat of the red coke, the cooled coke is discharged from the bottom of the dry quenching furnace, high-temperature inert gas discharged from an annular flue of the dry quenching furnace flows through a primary dust remover to a dry quenching process boiler to carry out heat exchange, steam is generated by the boiler, the cooled inert gas passes through a secondary dust remover and is blown into the dry quenching furnace again by the circulating fan, and the inert gas is recycled in a closed system. The primary dust remover system is positioned between the dry quenching furnace and the boiler and mainly used for removing solid dust in the circulating gas, and the working mode is that large-particle coke powder (more than 6 mm) in the circulating gas is separated by utilizing the gravity dust removal principle to realize the self-purification function of the circulating gas, so that loss scouring of the large-particle coke powder on the boiler tube of the boiler is reduced, and the high-efficiency operation of the dry quenching system is ensured. The heat exchange in the primary dust remover mainly carries out indirect heat exchange between the high-temperature coke powder in the cooling sleeve and the cooling water pipe, the high-temperature coke powder enters the cold area sleeve from the buffer bin at the upper part of the dust remover, the low-temperature coke powder is discharged from the lower part of the cold area sleeve after the heat exchange, and the water in the cooling water pipe continuously enters cold water and discharges hot water in the process. The long run of cooling water pipes causes grinding of the surface and over time there is a potential for leakage. In addition, due to the fluctuation of the material level of the ash discharge system, the sealing effect is poor, air can be sucked in stage, the air and high-temperature coke powder in the cooling sleeve are combusted to cause the oxidation reaction on the surface of the cooling water pipe, the leakage phenomenon of the cooling water pipe is caused, meanwhile, the local stress concentration phenomenon of the cooling water pipe is easily caused by the high temperature caused by combustion, and the local damage of the cooling water pipe is easily caused for a long time, so that the leakage phenomenon of the cooling water pipe is caused. At the present stage, the method for finding the leakage of the cooling water is mainly realized by manual inspection, and the coke powder removal condition and manual judgment are carried out.

When the cooling jacket leaks, the following problems are caused:

1) when the circulating cooling water in the cooling pipe leaks, a large amount of water enters the primary dust collector of the dry quenching coke, the water is contacted with hot coke powder to become water vapor and enters a dry quenching circulating system, finally, the water vapor reacts with coke in a dry quenching furnace to generate hydrogen and carbon monoxide, the concentration of combustible components in the circulating gas is increased rapidly, the balance state of the circulating gas is damaged, and the condition that air enters the circulating gas exists due to the leakage of a negative pressure section of the dry quenching coke circulating gas or the stop of a circulating fan during the production stop, namely, the explosion accident is easily caused in the presence of the air.

2) When the circulating cooling water in the cooling pipe leaks, the cooling water invades into the coke breeze layer, a large amount of soaked coke breeze cannot be discharged from the cooling sleeve to block the lower space of the dust remover, the pressure of the circulating gas passing through the lower space is increased, the increase of the circulating air quantity is seriously restricted, and the dry quenching rate of the dry quenching is reduced.

3) When circulating cooling water in the cooling pipe leaks, the phenomenon that coke powder in the cooling sleeve expands is easily caused, the upper material level meter of the cooling sleeve displays a high material level, but the actual coke powder material level in the water cooling sleeve is not high, at the moment, the running time of the lower ash discharge valve is prolonged according to an error high material level signal, the coke powder at the lower part of the cooling sleeve is completely emptied, a large amount of cold air is sucked into the water cooling sleeve from the grid ash discharge valve, severe combustion occurs at the position of the double-fork chute, the castable of the double-fork chute is peeled off seriously, the shell of the double-fork chute is further burnt through by high-temperature oxidation, and meanwhile, the once dedusting material loading instrument is also damaged by continuous high temperature. Causing serious economic loss and being incapable of ensuring the normal continuous production of the dry quenching.

When the cooling system of the primary dust remover leaks, the production of dry quenching needs to be stopped urgently to replace the cooling equipment, the production plan of the dry quenching needs to be adjusted during the overhaul period, and even the production plan of the coke oven can be influenced. The work of timely and accurately judging the occurrence of the leakage phenomenon of the dry quenching primary dedusting cooling system becomes very important.

Disclosure of Invention

In order to solve the technical problems in the prior art, the invention provides a leakage detection system and an intelligent diagnosis method for a coke dry quenching primary dedusting cooling system, which can effectively avoid the problems or reduce adverse effects caused by leakage, and ensure that the leakage can be timely and accurately found in the primary dedusting production process, thereby improving the production efficiency of the coke dry quenching process and playing a role in protecting driving and navigation for the stable and safe production of coke dry quenching.

In order to achieve the purpose, the invention adopts the following technical scheme:

a leakage detection system of a coke dry quenching primary dedusting cooling system comprises a data acquisition module, a leakage intelligent diagnosis module, a diagnosis information output module, a system leakage verification module and an expert self-learning module.

The data acquisition module comprises a water inlet pressure sensor PT01, a water inlet flow sensor F01, a water inlet flow regulating valve FV01, a backwater pressure sensor PT02, a backwater cut-off valve XZV01 and a backwater flow sensor F02;

a water inlet pressure sensor PT01, a water inlet flow sensor F01 and a water inlet flow regulating valve FV01 are arranged on a cooling water inlet pipeline of the primary coke dry quenching dust remover, and cooling inlet water sequentially passes through the pressure sensor PT01, the water inlet flow sensor F01 and the water inlet flow regulating valve FV01 and then enters the dust remover;

the return water pressure sensor PT02, the return water shut-off valve XZV01 and the return water flow sensor F02 are arranged on a cooling water outlet pipeline of the primary coke dry quenching dust remover, and cooling outlet water sequentially passes through the return water pressure sensor PT02, the return water shut-off valve XZV01 and the return water flow sensor F02 and then enters a return water pipeline;

the water inlet pressure sensor PT01, the water inlet flow sensor F01, the water inlet flow regulating valve FV01, the water return pressure sensor PT02, the water return cut-off valve XZV01 and the water return flow sensor F02 are all connected to a controller of the dry quenching control system;

the data acquisition module also comprises a dust collector bottom temperature sensor T01, and the dust collector bottom temperature sensor T01 is also connected to a controller of the dry quenching control system;

the data acquisition module further comprises a circulating gas H2 content sensor A01 and a circulating gas CO content sensor A02, the circulating gas H2 content sensor A01 and the circulating gas CO content sensor A02 are arranged on a rear end pipeline of a heat exchanger of the dry quenching circulating system, and the circulating gas H2 content sensor A01 and the circulating gas CO content sensor A02 are also connected into a controller of the dry quenching control system.

Further, the intelligent diagnosis method in the intelligent diagnosis module comprises the following steps:

the leakage intelligent diagnosis module judgment condition comprises the following four inputs:

1) the difference F between the inlet flow and the outlet flow is F01-F02 overrun and the duration thereof is overrun;

2) the inlet-outlet pressure difference P is P01-P02 overrun and the duration thereof is overrun;

3) the hydrogen H2 content a01 of the recycle gas increased while the CO content a02 of the recycle gas increased;

4) the bottom temperature T01 of the dust remover is suddenly changed in the normal heat exchange state.

1. The conditions for judging that the inlet-outlet flow difference F is F01-F02 overrun and the duration time is overrun are as follows:

F＝F01-F02

if > delta F and t > TF

Wherein: δ F — reasonable deviation of flow difference;

t is the condition satisfying the holding time;

TF-minimum value of hold time;

the delta F and TF value initial values are set by an operator according to the measurement precision and experience of the instrument, and are acquired from an expert knowledge base in the later period.

2. The conditions for judging that the inlet-outlet pressure difference P is P01-P02 overrun and the duration time is overrun are as follows:

P＝P01-P02

p > delta P and t > TP

Wherein: δ P — reasonable deviation of differential pressure;

t is the condition satisfying the holding time;

TP-minimum value of retention time;

the initial values of the delta P and TP values are set by the experience of operators, and are acquired from an expert knowledge base in the later period.

3. The judgment conditions for increasing the hydrogen H2 content A01 of the circulating gas and increasing the CO content A02 of the circulating gas are as follows:

if a leak condition occurs in the cooling system, water enters the circulating gas as water vapor, and the following reactions occur when the water vapor encounters hot coke in the coke dry quenching furnace:

C+H2O＝CO+H2----(1)

as shown in the formula (1), water gas is generated by the reaction of water vapor and coke, and the components are mainly CO and H2, so that the condition of water leakage can be indirectly known by observing the real-time values and the change rates of CO and H2 in the circulating gas;

the judgment conditions based on the H2 measurement are:

A_H＝α₁(f_H(t)-A_Hh)+α₂f_H(t)

A_H＞A_HS

wherein:

A_H-a comprehensive judgment value calculated from the H analysis data;

α₁-H2 analyzer data deviation weighting coefficients;

f_H(t) - - -H2 analyzer real-time data a 01;

A_Hh-an upper limit of the normal value of H2;

α₂-a rate of change weighting for H2 analyzer data;

f_H(t) - - -H2 analyzer rate of change over a period of n;

f_H(t-n) -H2 analyzer data n time periods prior to the current time;

A_HS-H2 analyzing the data to calculate a synthetic judgment value set-point;

where α is₁、A_Hh、n、α₂And A_HSSetting a value initial value by experience of an operator, and acquiring the value initial value from an expert knowledge base in a later period;

the judgment conditions based on the CO measurement are:

A_CO＝α₃(f_co(t)-A_coh)+α₄f_co(t)

A_CO＞A_COS

wherein:

A_CO-a comprehensive judgment value calculated from the CO analysis data;

α₃-CO analyzer data deviation weighting coefficients;

f_co(t) - - -CO analyzer real time data a 02;

A_coh-upper limit of normal value of CO;

α₄-a rate of change weighting for CO analyzer data;

f_co(t) - - -CO analyzer rate of change over a period of n;

f_co(t-n) -data of the CO analyzer n time periods prior to the current time;

A_COS-a set value of a comprehensive judgment value calculated from the CO analysis data;

where α is₃、A_coh、n、α₄And A_COSThe initial value of the value is set by the experience of operators, and is acquired from the expert knowledge base in the later period.

4. The judgment condition that the bottom temperature T01 of the dust remover changes suddenly in the normal heat exchange state is as follows:

T₀₁＝αf_T(t)

T₀₁<T_01S

wherein:

T₀₁-a comprehensive judgment value calculated from the analysis data of T01;

α - -T01 temperature drop coefficient;

f_T(T) - - -T01 real-time data;

f_T(T) - - -T01 rate of change over a period of n;

f_T(T-n) - - -T01 data n time periods before the current time;

T_01S-a comprehensive judgment value set value calculated by T01;

here T_01SAnd the initial value of the value n is set by experience of operators, and is acquired from an expert knowledge base in the later period.

Further, the diagnostic information output module is specifically:

after the expert system diagnoses that the system is leaked, the diagnosis information is sent to a picture of the control system to be displayed, the expert system participates in judgment manually, and if the diagnosis result of the expert system diagnosis is not verified, the expert system directly enters into the interlocking control operation; the operation reaction is timely, and the processing speed is improved; if the diagnosis result is required to be verified, the program enters a leakage system verification module;

interlock control operation: and the water inlet flow regulating valve FV01 is closed in an interlocking manner, the water supply control of the cooling system is stopped, and the backwater cut-off valve XZV01 is closed in an interlocking manner, so that the backwater backflow generated by a negative pressure system at the cooling sleeve is prevented from occurring, the corresponding discharging operation of the dust removal bin is not allowed by the interlocking, and the alarm output is given.

Furthermore, the system leakage verification module is used for rapidly cutting off the water inlet valve by turning off the return valve of the cooling water pipe, observing the change condition of the pressure in the cooling water pipe, judging the leakage condition of the cooling system if the pressure loss condition occurs in unit time, and quantitatively judging the leakage rate according to the pressure loss time; in order to improve the verification accuracy, the above operation needs to be repeated for many times.

Furthermore, the input of the expert self-learning module is used for comparing and learning the input information of the system leakage expert diagnosis system and the output information of the leakage verification system in real time, and the learning result is sent to an expert knowledge base to be used for diagnosing the system leakage expert diagnosis system model; the expert self-learning module calculates an expert knowledge result according to the diagnosis conclusion output by the intelligent leakage diagnosis module and the verification conclusion output by the system leakage verification module and the self-learning mechanism of the expert self-learning module, and stores the knowledge into the relational database by the contents of the diagnosis data and the diagnosis rules for future intelligent leakage diagnosis.

Compared with the prior art, the invention has the beneficial effects that:

firstly, an instrument monitoring means is added on a traditional primary dedusting cooling system, instrument data related to leakage diagnosis in the traditional dry quenching control system is uniformly applied to a diagnosis system, an information acquisition and interlocking control means of the cooling system is added, remote online monitoring is carried out on the dry quenching control system, the aim of on-site maintenance and humanization is fulfilled, the control level of the system is improved, and a solid foundation is laid for stable and continuous production of the process.

Secondly, the leakage intelligent diagnosis system is adopted, the intelligent diagnosis method with the fusion of various judgment conditions is adopted, the comprehensive diagnosis information source is integrated, the real-time state of process data is considered for each judgment condition, the change rate of the data is increased, the comprehensive diagnosis information is given through the calculation of a weighting model, and the accuracy and the timeliness of the leakage diagnosis of the cooling system are improved.

Thirdly, various set parameters in the intelligent leakage diagnosis method can adopt expert design values or result data of the expert knowledge base after multiple practical learning, so that the intelligent leakage diagnosis model is dynamically adjusted and automatically corrected, and the applicability of the intelligent leakage diagnosis model is improved.

The invention also comprises a function module for verifying the diagnosis result, an operator has the authority to select further verification operation on the diagnosis result output by the intelligent diagnosis model, the verification operation is fully automatically executed, the correctness of the diagnosis result can be corrected through verification, and the verification conclusion is also the necessary input condition of the expert self-learning system.

Fifthly, the interlocking control output of the invention is fully automatically executed, the operation response is timely, the processing speed is improved, the occurrence of serious safety accidents is prevented, the maintenance is timely, the efficiency is improved, and the economic loss is reduced.

And sixthly, the expert self-learning function of the invention can continuously perfect the knowledge base, provide accurate empirical data and diagnosis rules for the intelligent diagnosis model and improve the intelligent degree of the diagnosis system.

And seventhly, the invention can transplant the knowledge base of the learning result to other leakage diagnosis systems of the dry quenching primary dedusting cooling system through a large amount of diagnosis operation practices and expert self-learning, thereby omitting the verification step and improving the accuracy and timeliness of leakage diagnosis.

And eighthly, the application method of the system can be used for intelligent leakage diagnosis of the heat exchange cooling system in other fields through simple expansion, and has strong popularization.

Drawings

FIG. 1 is a system block diagram of a leak detection system for a coke dry quenching primary dust removal cooling system of the present invention;

FIG. 2 is a schematic diagram of an automatic intelligent diagnosis method for leakage of a dry quenching primary dedusting cooling system of the invention;

FIG. 3 is a block diagram of an implementation process of the automatic intelligent diagnosis method for leakage of the dry quenching primary dedusting cooling system.

In the figure: 1-water inlet pressure measurement PT 012-water inlet flow measurement F013-water inlet flow regulating valve FV 014-backwater pressure measurement PT 025-backwater cut-off valve XZV 016-backwater flow measurement F027-dust remover bottom temperature measurement T038-circulating gas H2 content measurement A019-circulating gas H content measurement A02.

Detailed Description

The following detailed description of the present invention will be made with reference to the accompanying drawings.

As shown in the figures 1-2, the system for detecting the leakage of the coke dry quenching primary dedusting cooling system and the automatic intelligent diagnosis method thereof are composed of a data acquisition module, an intelligent leakage diagnosis module, a diagnosis information output module, a system leakage verification module and an expert self-learning module. The system is realized on the basis of the existing control system of the dry quenching, wherein the storage of the expert information base is arranged in a relational database for management at the upper stage of the control system.

First, data acquisition module

As shown in FIG. 1, the cooling system of the dry quenching primary dust collector is divided into a bin A and a bin B. The cooling system of the A cabin and the cooling system of the B cabin are independent and have the same configuration, and only the cooling system of the A cabin is explained. The water inlet pipe and the water return pipe of the cooling system are supplemented with instrument detection and control items. The data acquisition module comprises a water inlet pressure sensor PT01(1), a water inlet flow sensor F01(2), a water inlet flow regulating valve FV01(3), a water return pressure sensor PT02(4), a water return cut-off valve XZV01(5) and a water return flow sensor F02 (6).

The water inlet pressure sensor PT01(1), the water inlet flow sensor F01(2) and the water inlet flow regulating valve FV01(3) are arranged on a cooling water inlet pipeline of the primary dry quenching dust remover, and cooling inlet water enters the dust remover after sequentially passing through the pressure sensor PT01(1), the water inlet flow sensor F01(2) and the water inlet flow regulating valve FV01 (3).

The return water pressure sensor PT02(4), the return water cut-off valve XZV01(5) and the return water flow sensor F02(6) are arranged on a cooling water outlet pipeline of the primary dry quenching dust remover, and cooling outlet water sequentially enters the return water pipeline after passing through the return water pressure sensor PT02(4), the return water cut-off valve XZV01(5) and the return water flow sensor F02 (6).

The coke dry quenching system is characterized in that the water inlet pressure sensor PT01(1), the water inlet flow sensor F01(2), the water inlet flow regulating valve FV01(3), the water return pressure sensor PT02(4), the water return cut-off valve XZV01(5) and the water return flow sensor F02(6) are all connected to a controller of the coke dry quenching control system.

The data acquisition module also includes a dust collector bottom temperature sensor T01(7), the dust collector bottom temperature sensor T01(7) also being connected to the controller of the dry quenching control system.

The data acquisition module further comprises a circulating gas H2 content sensor A01(8) and a circulating gas CO content sensor A02(9), wherein the circulating gas H2 content sensor A01(8) and the circulating gas CO content sensor A02(9) are arranged on a rear end pipeline of a heat exchanger of the dry quenching circulating system, and the circulating gas H2 content sensor A01(8) and the circulating gas CO content sensor A02(9) are also connected into a controller of the dry quenching control system.

The inflow flow control valve FV01(3) is a control valve with a quick cut-off function, generally a pneumatic control valve, and the shutoff of the inflow flow control valve FV01 is realized by shutting off the solenoid valve FSV01 arranged on the gas supply loop of the control valve. The water inlet flow can be automatically adjusted at any time according to the heat exchange requirement. Because the return water shut-off valve XZV01(5) does not have the requirement of quick shut-off, an electric shut-off valve is generally recommended. The leakage grades of the inlet water flow regulating valve FV01(3) and the backwater cut-off valve XZV01(5) are above ANSI CLASS V, the inlet water flow regulating valve FV01(3) is provided with a full-open and full-close limit switch ZAO01 and ZAC01, and the backwater cut-off valve XZV01(5) is provided with a full-open and full-close limit switch ZAO02 and ZAC 02. The instrument signal is sent to a dry quenching control system.

As shown in fig. 2-3, it is an automatic intelligent diagnosis method and an overall flow chart for leakage of the dry quenching primary dedusting cooling system of the invention.

Second, intelligent diagnosis module

The intelligent diagnosis method in the intelligent leakage diagnosis module comprises the following steps:

the leakage intelligent diagnosis module judgment condition comprises the following four inputs:

2) the difference F between the inlet flow and the outlet flow is F01-F02 overrun and the duration thereof is overrun;

2) the inlet-outlet pressure difference P is P01-P02 overrun and the duration thereof is overrun;

3) the hydrogen H2 content a01 of the recycle gas increased while the CO content a02 of the recycle gas increased;

4) the bottom temperature T01 of the dust remover is suddenly changed in the normal heat exchange state.

1. The conditions for judging that the inlet-outlet flow difference F is F01-F02 overrun and the duration time is overrun are as follows:

F＝F01-F02

if > delta F and t > TF

Wherein: δ F — reasonable deviation of flow difference;

t is the condition satisfying the holding time;

TF-minimum value of hold time;

the delta F and TF value initial values are set by an operator according to the measurement precision and experience of the instrument, and are acquired from an expert knowledge base in the later period.

Because the flowmeter is affected by the measurement accuracy, the inlet and outlet flow rates are generally not equal under the condition that the cooling water pipe does not leak, certain errors exist, and the influence of the accuracy of the meter is removed as much as possible during judgment.

2. The conditions for judging that the inlet-outlet pressure difference P is P01-P02 overrun and the duration time is overrun are as follows:

P＝P01-P02

p > delta P and t > TP

Wherein: δ P — reasonable deviation of differential pressure;

t is the condition satisfying the holding time;

TP-minimum value of retention time;

the initial values of the delta P and TP values are set by the experience of operators, and are acquired from an expert knowledge base in the later period.

3. The judgment conditions for increasing the hydrogen H2 content A01 of the circulating gas and increasing the CO content A02 of the circulating gas are as follows:

if a leak condition occurs in the cooling system, water enters the circulating gas as water vapor, and the following reactions occur when the water vapor encounters hot coke in the coke dry quenching furnace:

C+H2O＝CO+H2----(1)

as shown in the formula (1), water gas is generated by the reaction of water vapor and coke, and the components are mainly CO and H2, so that the condition of water leakage can be indirectly known by observing the real-time values and the change rates of CO and H2 in the circulating gas;

the judgment conditions based on the H2 measurement are:

A_H＝α₁(f_H(t)-A_Hh)+α₂f_H(t)

A_H＞A_HS

wherein:

A_H-a comprehensive judgment value calculated from the H analysis data;

α₁-H2 analyzer data deviation weighting coefficients;

f_H(t) - - -H2 analyzer real-time data a 01;

A_Hh-an upper limit of the normal value of H2;

α₂-a rate of change weighting for H2 analyzer data;

f_H(t) - - -H2 analyzer rate of change over a period of n;

f_H(t-n) -H2 analyzer data n time periods prior to the current time;

A_HS-H2 analyzing the data to calculate a synthetic judgment value set-point;

where α is₁、A_Hh、n、α₂And A_HSSetting a value initial value by experience of an operator, and acquiring the value initial value from an expert knowledge base in a later period;

the judgment conditions based on the CO measurement are:

A_CO＝α₃(f_co(t)-A_coh)+α₄f_co(t)

A_CO＞A_COS

wherein:

A_CO-a comprehensive judgment value calculated from the CO analysis data;

α₃-CO analyzer data deviation weighting coefficients;

f_co(t) - - -CO analyzer real time data a 02;

A_coh-upper limit of normal value of CO;

α₄-a rate of change weighting for CO analyzer data;

f_co(t) - - -CO analyzer rate of change over a period of n;

f_co(t-n) -data of the CO analyzer n time periods prior to the current time;

A_COS-a set value of a comprehensive judgment value calculated from the CO analysis data;

where α is₃、A_coh、n、α₄And A_COSThe initial value of the value is set by the experience of operators, and is acquired from the expert knowledge base in the later period.

4. The judgment condition that the bottom temperature T01 of the dust remover changes suddenly in the normal heat exchange state is as follows:

T₀₁＝αf_T(t)

T₀₁<T_01S

wherein:

T₀₁-a comprehensive judgment value calculated from the analysis data of T01;

α - -T01 temperature drop coefficient;

f_T(T) - - -T01 real-time data;

f_T(T) - - -T01 rate of change over a period of n;

f_T(T-n) - - -T01 data n time periods before the current time;

T_01S-integrated judgment value set value calculated by T01；

Here T_01SAnd the initial value of the value n is set by experience of operators, and is acquired from an expert knowledge base in the later period.

The intelligent leakage diagnosis system carries out intelligent judgment through the input of the information, judges whether the system has leakage according to the judgment rule of the intelligent leakage diagnosis system, returns the program to the data acquisition module if the judgment result has no leakage, and repeatedly carries out data acquisition work. And if the system has leakage according to the judgment result, calculating the leakage rate and giving the leakage probability of the system in real time.

Third, diagnosis information output module

The diagnosis information output module is specifically as follows:

after the expert system diagnoses that the system is leaked, the diagnosis information is sent to a picture of the control system to be displayed, the expert system participates in judgment manually, and if the diagnosis result of the expert system diagnosis is not verified, the expert system directly enters into the interlocking control operation; the operation reaction is timely, and the processing speed is improved; if the diagnosis result is required to be verified, the program enters a leakage system verification module;

interlock control operation: the water inlet flow regulating valve FV01(3) is closed in an interlocking manner, the water supply control of the cooling system is stopped, and the return water cut-off valve XZV01(5) is closed in an interlocking manner, so that the return water reflux generated by a negative pressure system at the cooling sleeve is prevented, the corresponding discharging operation of the dust removal bin is not allowed by the interlocking, and the alarm output is given.

Fourth, system leakage verification module

In order to make the automatic leakage diagnosis system more intelligent and make the diagnosis result more accurate, the invention is provided with the system leakage verification module to verify the operation of the diagnosis system, because the water inlet valve and the water return valve need to be operated during the verification operation, the working continuity of the coke dry quenching primary dust removal cooling system is influenced, the module does not need to be frequently operated under the general condition, and an operator generally selects whether to operate the leakage verification module according to the actual requirement after the system gives a leakage alarm. The verification frequency of the system verification module is increased, more expert knowledge information can be provided for the expert self-learning system, and the diagnosis accuracy of the system leakage expert diagnosis system model is facilitated.

The system leakage verification module is used for rapidly cutting off the water inlet valve by turning off the return valve of the cooling water pipe, observing the change condition of the pressure in the cooling water pipe, judging that the cooling system has a leakage condition if the pressure loss condition occurs in unit time, and quantitatively judging the leakage rate according to the pressure loss time; in order to improve the verification accuracy, the above operation needs to be repeated for many times.

The verification method of the system leakage verification module comprises the following steps:

a) closing the backwater cut-off valve XZV-01 (5);

b) judging that the backwater cut-off valve XZV-01(5) is completely closed;

c) the following operations were repeated N times:

1) a water inlet flow regulating valve FV-01(3) is closed quickly;

2) judging that the valve inlet flow regulating valve FV-01(3) is completely closed;

3) observing the pressure change trend of a backwater pressure sensor PT-02 (4);

4) if the return water pressure sensor PT-02(4) is subjected to rapid pressure loss, continuing the next step 5, if the return water pressure sensor PT-02(4) is not subjected to rapid pressure loss, finishing verification, and feeding back verification information to the expert self-learning system;

5) recording the pressure loss time;

6) opening a water inlet flow regulating valve FV-01(3) until a water outlet return pressure sensor PT-02(4) recovers to be close to a pressure value before the water inlet flow regulating valve FV-01(3) is closed;

7) returning to 1 to restart, carrying out verification operation, and carrying out the next step to d after the system records N times;

d) calculating the average value Tavg of the pressure loss time of N times;

e) determining that a cooling system has a leak, and interlockingly controlling operation of the diagnostic information output module;

f) and (4) quantifying the leakage rate according to the length of the average value Tavg of the pressure loss time, and then sending the calculation result to an expert self-learning module.

Expert self-learning module

In order to improve the accuracy of the diagnosis result of the system leakage expert diagnosis system, the invention provides an expert self-learning module, the input of which is used for comparing and learning the input information of the system leakage expert diagnosis system and the output information of the leakage verification system in real time and sending the learning result to an expert knowledge base for diagnosing the system leakage expert diagnosis system model; the expert self-learning module calculates an expert knowledge result according to the diagnosis conclusion output by the intelligent leakage diagnosis module and the verification conclusion output by the system leakage verification module and the self-learning mechanism of the expert self-learning module, and stores the knowledge into the relational database by the contents of the diagnosis data and the diagnosis rules for future intelligent leakage diagnosis.

The invention provides an automatic intelligent diagnosis method for leakage of a primary dust removal cooling system of dry quenching coke. The invention improves the traditional dry quenching dust removal cooling system, adds flow detection, pressure detection and control valves on a water inlet pipe and a water return pipe of the cooling system, and simultaneously utilizes the original instrument detection means in the dry quenching process system, wherein the detection means comprises the component analysis and detection of circulating gas and the bottom temperature detection of a primary dust remover. The existing control system of the dry quenching is utilized to complete the work of detection information collection, model calculation and control output. The collected information is analyzed in the control system and input into the intelligent diagnosis system of the leakage expert of the cooling system for diagnosis and treatment, and a leakage diagnosis result is given. And the diagnosis result is displayed on an operation station of the control system to guide the operator to carry out timely maintenance. If the judgment result is that the leakage condition occurs, the control system can close the water inlet valve and the water return valve in an interlocking manner at once, so that the leakage phenomenon is stopped at once, the process system is in a safe state, and an operator is informed to enter the maintenance process at once. In order to improve the accuracy of the diagnosis result of the automatic diagnosis expert system, the invention is provided with a system leakage verification function, and an operator selects whether to perform the verification function of the leakage diagnosis result. The verification information given by the verification function module can automatically enter the expert self-learning module for analysis and learning, and the learning result is stored in the knowledge base for the future system leakage diagnosis expert system to use. The use of the expert self-learning system can continuously enrich the content of the expert knowledge base and improve the diagnosis accuracy of the intelligent diagnosis system. After long-time learning accumulation and practice inspection, the expert knowledge base can be continuously updated and can be conveniently transplanted to other intelligent leakage judgment applications of the dry quenching primary dedusting cooling system.

The above embodiments are implemented on the premise of the technical solution of the present invention, and detailed embodiments and specific operation procedures are given, but the scope of the present invention is not limited to the above embodiments. The methods used in the above examples are conventional methods unless otherwise specified.