CN-121976864-A - Intelligent soft start control method and system for turbine jigger system

CN121976864ACN 121976864 ACN121976864 ACN 121976864ACN-121976864-A

Abstract

The invention discloses an intelligent soft start control method and system of a turbine jigger system, and mainly relates to the technical field of system control. The method comprises the steps of dividing a jigger control process into a plurality of dividing stages according to time evolution characteristics of a steam turbine jigger operation process, respectively establishing state monitoring and control logic of corresponding stages in a distributed control system, acquiring a plurality of multi-source operation state signal sequences, triggering a jigger soft start process, adaptively selecting control actions of jigger motors and controlling a meshing execution process, executing asynchronous double monitoring to acquire an asynchronous double monitoring result, and controlling the jigger motors to stop and driving jigger devices to automatically disconnect to a safe position when a jigger command is detected or the asynchronous double monitoring result is abnormal. The control system has the beneficial effects that the technical problem that the control discretization of the turbine jigger system is low in intelligent degree and cannot cope with complex working conditions in the prior art is solved, and the technical effect of improving the soft start control is achieved.

Inventors

WANG YUANZHENG
DONG XUEWU
WANG YIBANG
LIAO GUO
XU AIQING
LU YUANLI

Assignees

广东粤电新会发电有限公司

Dates

Publication Date: 20260505
Application Date: 20260205

Claims (9)

1. The intelligent soft start control method of the turbine jigger system is characterized by comprising the following steps: Dividing a jigger control process into a plurality of dividing stages according to the time evolution characteristics of the jigger operation process of the steam turbine, and respectively establishing state monitoring and control logic of corresponding stages in a distributed control system, wherein the plurality of dividing stages are a preparation stage before starting, a soft start execution stage and a jigger operation stage; Respectively acquiring multisource operation state signals of the turbine body equipment, the lubrication and top shaft system, the jigger device and the executing mechanism in a plurality of dividing stages to obtain a plurality of multisource operation state signal sequences, and taking the multisource operation state signal sequences as control inputs of the dividing stages; After receiving a jigger starting instruction, judging jigger starting permission conditions based on a multisource running state signal sequence in a preparation stage before starting, and triggering a jigger soft starting process when the permission conditions are met; in the soft start execution stage, according to the real-time feedback signal interaction results of the preparation stage before starting and the soft start execution stage, the control action of the jigger motor is adaptively selected, and the meshing execution process of the jigger device is controlled; And in the jigger operation stage, performing asynchronous dual monitoring on the corresponding multi-source operation state signal sequence to obtain an asynchronous dual monitoring result, and controlling the jigger motor to stop and driving the jigger device to automatically disconnect to a safe position when a jigger instruction is detected or the asynchronous dual monitoring result is abnormal.
2. The intelligent soft start control method of a steam turbine turning system according to claim 1, wherein after receiving a turning start command, determining a turning start permission condition based on a multi-source running state signal sequence of a preparation stage before starting, and triggering a turning soft start flow when the permission condition is satisfied, comprising: Performing multiscale stability time sequence interaction analysis on the multisource running state signal sequence to obtain state interaction characteristics of a preparation stage before starting; judging whether the state interaction characteristic of the preparation stage before starting meets the jigger starting permission condition or not, and obtaining a judging result; when the judgment result is that the permission condition is met, the soft start permission signal sent by the DCS side is received by the soft start starter of the jigger, and the jigger soft start flow is triggered.
3. The intelligent soft start control method of a steam turbine turning gear system according to claim 2, wherein performing multi-scale stability time sequence interaction analysis on the multi-source running state signal sequence to obtain pre-start preparation stage state interaction characteristics comprises: A preset multiscale feature analysis module is obtained, stability feature analysis is carried out on the multisource running state signal sequence, and a plurality of multiscale stability features are obtained, wherein each multiscale stability feature comprises a lubricating oil main pipe pressure signal stability feature, a top shaft main pipe oil pressure signal stability feature, a turbine rotating speed signal stability feature, a jigger device position feedback signal stability feature and a jigger motor protection state signal stability feature; Carrying out serialization treatment on the multiple multi-scale stability characteristics according to the size of the characteristic analysis scale to obtain a multi-scale stability characteristic sequence; And executing multi-scale stability time sequence interaction analysis based on the multi-scale stability feature sequence to obtain the pre-starting preparation stage state interaction feature.
4. The intelligent soft start control method of a steam turbine turning gear system according to claim 3, wherein performing multi-scale stability time sequence interaction analysis based on the multi-scale stability feature sequence to obtain the pre-start preparation stage state interaction feature comprises: performing time sequence interaction analysis on the first multiscale stability feature and the second multiscale stability feature in the multiscale stability feature sequence to obtain a stage stability interaction feature; Performing time sequence interaction analysis on a third multiscale stability feature in the multiscale stability feature sequence by using the stage stability interaction feature, and updating the stage stability interaction feature according to an analysis result; And continuing to execute time sequence interaction analysis and phase stability interaction feature updating on the multi-scale stability features in the multi-scale stability feature sequence by using the updated phase stability interaction features until the end of the sequence is reached, so as to obtain the pre-start preparation phase state interaction features.
5. The intelligent soft start control method of a steam turbine turning gear system according to claim 3, wherein performing a time sequence interaction analysis on a first multiscale stability feature and a second multiscale stability feature in the multiscale stability feature sequence to obtain a phase stability interaction feature comprises: Performing mapping fine-granularity feature similarity recognition on the first multi-scale stability feature and the second multi-scale stability feature to obtain a mapping fine-granularity feature similarity set, wherein the mapping fine-granularity feature similarity recognition comprises a lubricating oil main pipe pressure signal stability feature, a top shaft main pipe oil pressure signal stability feature, a turbine rotating speed signal stability feature, a jigger device position feedback signal stability feature and a jigger motor protection state signal stability feature; Preprocessing based on the mapping fine-grained feature similarity set to construct a time sequence interaction analysis matrix; And utilizing the time sequence interaction analysis matrix to interact the second multiscale stability characteristics to obtain the stage stability interaction characteristics.
6. The intelligent soft start control method of a turbine jigger system according to claim 5, wherein the preprocessing is performed based on the mapping fine-grained feature similarity set, and the constructing of the time sequence interaction analysis matrix comprises: normalizing the mapping fine-grained feature similarity set through a softmax function to obtain a mapping fine-grained feature similarity normalization value set; And filling the mapping fine-granularity feature similarity normalization value set into an initially empty diagonal matrix, and constructing the time sequence interaction analysis matrix.
7. The intelligent soft start control method of a steam turbine jigger system according to claim 1, wherein, in a jigger operation stage, asynchronous double monitoring is performed on a corresponding multi-source operation state signal sequence to obtain an asynchronous double monitoring result, and when a jigger command or the asynchronous double monitoring result is detected to be abnormal, the jigger motor is controlled to stop and the jigger device is driven to be automatically disconnected to a safe position, comprising: acquiring a historical abnormal operation log set of a turbine turning system during soft start under historical time; Analyzing the abnormal intervals of the historical abnormal operation log set, and determining a first-step monitoring period and a second-step monitoring period, wherein the time length of the first-step monitoring period is smaller than that of the second-step monitoring period; constructing an asynchronous dual monitor based on the step-by-step monitoring period and the step-by-step monitoring period; And executing asynchronous double monitoring on the multi-source running state signal sequence by using the asynchronous double monitor to obtain an asynchronous double monitoring result.
8. The intelligent soft start control method of a turbine jigger system according to claim 7, wherein the performing the abnormal interval analysis on the history abnormal operation log set to determine a step-by-step monitoring period and a step-by-step monitoring period comprises: Extracting time intervals of occurrence of two adjacent types of anomalies in the historical abnormal operation log set to obtain an abnormal interval set; taking an abnormal interval corresponding to the minimum value in the abnormal interval set as a step-by-step monitoring period; And taking the abnormal interval corresponding to the maximum value in the abnormal interval set as a later monitoring period.
9. An intelligent soft start control system for a turbine jigger system, characterized in that the system is used for implementing an intelligent soft start control method for a turbine jigger system according to any one of claims 1-8, the system comprising: the control logic construction module is used for dividing the jigger control process into a plurality of dividing stages according to the time evolution characteristics of the jigger operation process of the steam turbine, and respectively establishing state monitoring and control logic of corresponding stages in the distributed control system, wherein the plurality of dividing stages are a preparation stage before starting, a soft start execution stage and a jigger operation stage; The signal acquisition module is used for respectively acquiring multisource operation state signals of the turbine body equipment, the lubrication and jackshaft system, the jigger device and the executing mechanism in a plurality of dividing stages to obtain a plurality of multisource operation state signal sequences, and the multisource operation state signal sequences are used as control inputs of the dividing stages; the barring starting permission condition judging module is used for judging barring starting permission conditions based on a multi-source running state signal sequence in a preparation stage before starting after receiving a barring starting instruction, and triggering a barring soft starting process when the permission conditions are met; The control action self-adaptive selection module is used for self-adaptively selecting the control action of the jigger motor and controlling the meshing execution process of the jigger device according to the real-time feedback signal interaction results of the preparation stage before starting and the soft start execution stage; and the monitoring control module is used for executing asynchronous double monitoring on the corresponding multi-source running state signal sequence in the jigger running stage to obtain an asynchronous double monitoring result, and controlling the jigger motor to stop and driving the jigger device to automatically disconnect to a safe position when a jigger command is detected or the asynchronous double monitoring result is abnormal.

Description

Intelligent soft start control method and system for turbine jigger system Technical Field The invention relates to the technical field of system control, in particular to an intelligent soft start control method and system of a turbine jigger system. Background The turbine jigger system is key auxiliary equipment in the start-stop process of large-scale rotary units such as thermal power and nuclear power, and has the function of driving the rotor at a low speed after the unit is stopped or before the unit is started so as to prevent the rotor from being bent due to uneven heating. The traditional jigger system mostly adopts a simple control loop based on a relay, and the operation mode is mainly manual in situ, namely, an operator is required to manually operate a mechanical handle in situ to enable a jigger gear to be meshed with a turbine gear wheel, and then the jigger motor is started after switching on. With the popularization of a Distributed Control System (DCS) in a power station, a part of jigger systems realize the function of remote start-stop motors, but the key engagement and disengagement processes still depend on manual intervention and judgment. At present, the control process is discrete and split, and a coherent and staged intelligent control framework cannot be constructed according to the internal sequential logic of jigger operation, so that state monitoring of each link is incomplete and interlocking protection is imperfect. And secondly, the control input signal is single, and mainly depends on a few switching values, and the fusion analysis and the cooperative judgment on the multi-source state information of the turbine body, the auxiliary system and the like are lacked, so that the reliability of automatic control is insufficient, and complex working conditions are difficult to deal with. Disclosure of Invention The application provides an intelligent soft start control method and system of a turbine jigger system, which are used for solving the technical problems that the control discretization of the turbine jigger system is low in intelligent degree and cannot cope with complex working conditions in the prior art. In view of the above problems, the application provides an intelligent soft start control method and system for a turbine jigger system. In a first aspect of the present application, an intelligent soft start control method for a turbine jigger system is provided, the method comprising: According to the time evolution characteristics of the steam turbine turning operation process, the turning control process is divided into a plurality of dividing stages, state monitoring and control logic of corresponding stages are respectively established in a distributed control system, the dividing stages are a preparation stage before starting, a soft start execution stage and a turning operation stage, multisource operation state signals of a steam turbine body device, a lubrication and top shaft system, a turning device and an execution mechanism are respectively collected in the dividing stages, a plurality of multisource operation state signal sequences are obtained, the multisource operation state signal sequences are used as control inputs of the dividing stages, after a turning starting instruction is received, a turning starting permission condition is judged based on the multisource operation state signal sequences of the preparation stage before starting, a turning soft start flow is triggered when the permission condition is met, in the soft start execution stage, the control action of a turning motor is adaptively selected according to real-time feedback signal interaction results of the preparation stage before starting and the soft start execution stage, the engagement execution process of the turning device is controlled, double monitoring is carried out on the corresponding multisource operation state signal sequences, and when the asynchronous turning operation command is received, the asynchronous detection result is automatically detected to the asynchronous detection result or the asynchronous detection command is stopped, and the automatic safety control device is stopped when the asynchronous detection command is stopped. In one possible implementation manner, after receiving a jigger starting instruction, judging jigger starting permission conditions based on a multisource operation state signal sequence of a preparation stage before starting, and triggering a jigger soft starting process when the permission conditions are met, wherein the method comprises the steps of executing multiscale stability time sequence interaction analysis on the multisource operation state signal sequence to obtain state interaction characteristics of the preparation stage before starting; and when the judging result is that the permission condition is satisfied, judging that the barring soft starter receives a soft start permission signal sent by the DCS side, and triggering