Main oil pump boundary element multi-unit intelligent cooperative operation and maintenance system

Abstract

The invention relates to a main oil pump boundary element multi-unit intelligent cooperative operation and maintenance system, and belongs to the technical field of intelligent operation and maintenance of industrial equipment. By fusing a boundary element method and a multi-unit collaborative optimization algorithm, a dynamic operation and maintenance model of a main oil pump cluster is constructed, and intelligent scheduling and resource optimization configuration of multi-unit maintenance tasks are achieved. The system comprises three core components including a main oil pump state monitoring network, a boundary element characteristic analysis module and a multi-unit collaborative decision engine, wherein the boundary element characteristic analysis module establishes a mapping relation between a main oil pump operation state and mechanical loss based on a three-dimensional field domain discretization method; and the multi-unit collaborative decision engine generates an optimal maintenance strategy sequence by adopting a dynamic programming algorithm. According to the system, the non-planned downtime can be reduced, the operation and maintenance cost is saved, and meanwhile, the overall reliability index of a multi-unit system is effectively improved.

Description

TECHNICAL FIELD

[0001] The application relates to a main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system and belongs to the technical field of intelligent operation and maintenance of industrial equipment. BACKGROUND

[0002] In the current main oil pump operation and maintenance field, a maintenance strategy based on empirical rules is generally adopted, and there are prominent problems of low multi-unit collaborative efficiency and insufficient resource optimization. The traditional method mainly relies on manual experience to formulate a maintenance plan, and it is difficult to accurately evaluate the equipment operation state and residual life, resulting in large deviation in maintenance opportunity selection, and the phenomenon of redundant or insufficient downtime frequently occurs. In the existing technology, when dealing with multi-unit collaborative optimization, a finite element analysis method is usually used for stress field and temperature field coupling calculation, and the calculation amount increases exponentially with the increase of the number of units, which cannot meet the real-time requirement. The conventional system lacks dynamic response capability to boundary conditions and cannot effectively predict and collaboratively dispose sudden failures such as oil film rupture and shaft vibration. In addition, the existing mathematical model has precision defects in describing the coupling effect of multiple physical fields, especially in the analysis of non-linear flow of lubricating medium and rotor dynamics characteristics, and the error is significant, resulting in reduced credibility of maintenance decision. The current solution has not realized intelligent optimization iteration based on real-time operation data, and cannot dynamically adjust the load distribution and maintenance priority among multiple units, ultimately causing the dual dilemma of low equipment utilization and high operation and maintenance cost. SUMMARY

[0003] The application aims to solve the problems of low multi-unit collaborative efficiency and insufficient collaboration in the current main oil pump operation and maintenance field, and proposes a main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system.

[0004] The technical scheme of the application is as follows:

[0005] A main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system comprises a data acquisition module for acquiring real-time operation state data of a main oil pump unit group, including vibration, temperature and oil pressure parameters; a boundary element analysis module connected with the data acquisition module for establishing a main oil pump boundary element model and solving characteristic parameters; a collaborative optimization module connected with the boundary element analysis module for generating a multi-unit collaborative maintenance strategy according to unit state data and boundary element analysis results; a decision execution module connected with the collaborative optimization module for executing maintenance instructions and feeding back execution results; and a man-machine interaction module for data interaction with the data acquisition module, the boundary element analysis module, the collaborative optimization module and the decision execution module, for providing visual display and interactive control of system operation state.

[0006] The feedback output end of the decision execution module is further connected with the input end of the collaborative optimization module, forming a closed-loop optimization loop for strategy execution.

[0007] Specifically, the data acquisition module comprises:

[0008] a vibration sensor group arranged at the main oil pump rotating shaft and bearing position, with a sampling frequency range of Hz; a temperature sensor array arranged at the surface of the main oil pump shell and the lubricating oil circuit, with a measurement accuracy of C; and a pressure transmitter group arranged at the oil inlet and outlet pipe sections, with a range covering MPa.

[0009] Specifically, the boundary element analysis module comprises: a characteristic parameter extraction unit for performing boundary integral equation calculation:

[0010] wherein, is a Green function, is a field variable, is a normal derivative, is a pump body boundary discrete unit; and a state evaluation unit for constructing a three-dimensional boundary element model and calculating characteristic parameters wherein is an eigenvalue of the th order, is a mass matrix.

[0011] Specifically, the collaborative optimization module comprises: a multi-unit data aggregation unit for constructing a unit state matrix wherein denotes the th performance index of the th unit; a dynamic programming unit for solving a target function by using an improved genetic algorithm:

[0012] wherein, is the operation and maintenance cost of the th period, is the energy efficiency deviation of the th unit, is a weight coefficient; and a maintenance decision unit for outputting an optimal maintenance sequence that satisfies a constraint condition .

[0013] Specifically, the decision execution module comprises: an instruction distribution unit for transmitting control instructions to each unit PLC through an industrial Ethernet protocol; and a feedback adjustment unit for monitoring execution deviation and triggering a compensation mechanism.

[0014] Specifically, the human-computer interaction module comprises: a three-dimensional visualization unit for rendering a boundary element grid model and a stress cloud map; and a warning prompt unit for prompting when a characteristic parameter exceeds a threshold value A multi-stage alarm signal is generated.

[0015] Specifically, the system modules interact with each other through the OPC UA protocol, and the boundary element discrete unit size meets the accuracy requirement of , wherein is the minimum stress wave wavelength.

[0016] Specifically, the wavelet packet decomposition algorithm is used for vibration signal processing in the data acquisition module, and the decomposition layer number , and the characteristic frequency resolution Hz.

[0017] Specifically, the fitness function change rate for 5 iterations.

[0018] Advantages of the present application:

[0019] (1) The boundary element accurate modeling greatly reduces the calculation dimension, and realizes efficient reconstruction of the internal flow field of the main oil pump;

[0020] (2) The multi-objective collaborative optimization algorithm improves the utilization rate of maintenance resources and shortens the downtime;

[0021] (3) The intelligent decision system constructed supports simultaneous processing of more than 10 unit groups for collaborative operation and maintenance;

[0022] (4) It is especially suitable for the intelligent operation and maintenance scene of the main oil pump group in the field of thermal power generation, petroleum chemical industry, etc., and the intelligent level of equipment management is significantly improved through multi-dimensional technology integration. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 is the implementation flowchart of the present application;

[0024] Figure 2 is the module architecture relationship diagram of the present application. DETAILED DESCRIPTION

[0025] The technical solutions of the present application will be described in detail below with reference to the accompanying drawings. The implementation flowchart of the present embodiment is shown in FIG. Figure 1 , and the system module architecture is shown in FIG. Figure 2 .

[0026] The hardware architecture and software module specific configuration of the present system in implementation are as follows:

[0027] The data acquisition module 100 is used for real-time acquisition of the running state data of the main oil pump unit group, including vibration, temperature and oil pressure parameters, including: a vibration sensor group 101 installed on the main oil pump shaft and bearing position, and the sampling frequency range is set to Hz; the temperature sensor array 102 is arranged on the surface of the main oil pump housing and the lubricating oil circuit, and the measurement accuracy reaches C. The pressure transmitter group 103 is installed on the oil inlet and oil outlet pipe sections, and the measurement range covers MPa. The boundary element analysis module 200 is connected with the data acquisition module 100, and is used for establishing a boundary element model of the main oil pump and solving characteristic parameters, including: a characteristic parameter extraction unit 210, which is used for performing boundary integral equation calculation: , wherein, is a Green function, is a field variable, is a normal derivative, is a pump body boundary discrete unit; a state evaluation unit 220, which is used for constructing a three-dimensional boundary element model and calculating characteristic parameters , wherein is the order eigenvalue, is a mass matrix. The module performs the following processing steps: first, a three-dimensional boundary element discrete model of the main oil pump is established, and a fluid domain is divided into a boundary element grid; then, the distribution characteristics of a velocity potential function on the boundary element are calculated; next, a numerical solution of a pressure wave equation is solved; finally, a time-domain distribution of a dynamic pressure load acting on the structure boundary is extracted. In modeling, the boundary element discrete model uses quadratic isoparametric elements, each element contains 8 nodes, and the shape function uses a double quadratic polynomial: , wherein, is a local coordinate parameter, is a normalized coordinate of node m.

[0028] The collaborative optimization module 300 is connected with the boundary element analysis module 200, and is used for generating a multi-unit collaborative maintenance strategy according to unit state data and boundary element analysis results, including: a multi-unit data aggregation unit 310, which is used for establishing a unit state matrix , wherein represents the performance index of the unit; a dynamic programming unit 320, which is used for solving an optimization objective function: , wherein, is the operation and maintenance cost of the time period, is the energy efficiency deviation of the unit, is a weight coefficient; a maintenance decision unit 330, which outputs an optimal maintenance sequence and satisfies a constraint condition; the module further includes the following units: a unit health degree evaluation unit, which calculates residual life indicators of each unit ,in The failure rate coefficient; the maintenance cost calculation unit, generating the resource constraint matrix. , Represent the cost of the j-th type of maintenance operation for the i-th unit; use a dynamic programming solver to construct the objective function. ,in This refers to the downtime. To maintain costs, These are the weighting coefficients; the dynamic programming solver employs an improved NSGA-II algorithm, setting the crossover probability. Probability of mutation Population size Furthermore, this module also employs a dynamic decision-making model based on reinforcement learning: ,in, For the first Revenue from periodic maintenance. For resource consumption costs, As a penalty factor, The discount factor is used; and a multi-agent negotiation mechanism based on Nash equilibrium is designed. The decision execution module 400, connected to the collaborative optimization module 300, is used to execute maintenance commands and provide feedback on the execution results. It includes: a command distribution unit 410, which transmits control commands to the PLCs of each unit via an industrial Ethernet protocol; and a feedback adjustment unit 420, which monitors execution deviations in real time. And trigger a compensation mechanism; this module also includes: an adaptive adjustment unit, which adjusts according to the actual oil pressure deviation. Adjust valve opening : ,in, This is the proportional gain coefficient. The integral time constant; Abnormal operating condition identification unit, setting the vibration acceleration threshold. and temperature gradient threshold The adaptive adjustment unit also includes a feedforward compensator with the following transfer function: This is used to offset pressure fluctuations caused by sudden changes in the main oil pump load. The human-machine interface module 500 interacts with the data acquisition module 100, boundary element analysis module 200, collaborative optimization module 300, and decision execution module 400 to provide a visual display and interactive control of the system's operating status. This includes: a 3D visualization unit 510 for rendering the boundary element mesh model and stress cloud diagram; and an early warning unit 520 that alerts the system when characteristic parameters exceed a threshold. Multiple alarm signals are generated in real time.

[0029] When implementing the system, follow these steps:

[0030] Step one: Real-time acquisition of vibration, temperature, and pressure data of each unit by data acquisition module 100. Data interaction between system modules is achieved through OPC UA protocol.

[0031] Step two: Modeling and analysis by boundary element analysis module 200. In this process, the boundary element discrete unit size meets , where is the minimum stress wave wavelength. Wavelet packet decomposition algorithm is used for vibration signal processing, with decomposition level , and characteristic frequency resolution reaching Hz. The iteration termination condition of the optimization algorithm is set to the change rate of fitness function for 5 consecutive iterations.

[0032] Step three: Multi-unit cooperative optimization process includes:

[0033] (1) Real-time acquisition of vibration, temperature, and pressure data of each unit by sensor group;

[0034] (2) Construction of boundary element model and calculation of characteristic parameter matrix;

[0035] (3) Establishment of comprehensive evaluation index including equipment state, energy consumption, and maintenance cost;

[0036] (4) Solution of optimal maintenance strategy using mixed integer programming algorithm.

[0037] (5) Dynamic adjustment of unit load distribution and verification of strategy effectiveness;

[0038] (6) Generation of visual report and updating of equipment life prediction model.

[0039] Step four: System hardware configuration requirements include: industrial-grade embedded processor with main frequency ≥ 2.4 GHz, memory capacity ≥ 16 GB, and storage space ≥ 1 TB. Software environment requirements include: boundary element solver supporting parallel computing, with Python and C++ mixed programming interface, and real-time database response time ≤ 50 ms.

Claims

1. A multi-unit intelligent collaborative operation and maintenance system for the main oil pump boundary element, characterized in that, include: The data acquisition module (100) is used to collect real-time operating status data of the main oil pump unit group, including vibration, temperature and oil pressure parameters; A boundary element analysis module (200), connected to the data acquisition module (100), is used to establish a boundary element model of the main oil pump and solve for characteristic parameters; a collaborative optimization module (300), connected to the boundary element analysis module (200), is used to generate a multi-unit collaborative maintenance strategy based on the unit status data and boundary element analysis results; a decision execution module (400), connected to the collaborative optimization module (300), is used to execute maintenance instructions and provide feedback on the execution results; a human-machine interaction module (500), which interacts with the data acquisition module (100), boundary element analysis module (200), collaborative optimization module (300), and decision execution module (400) respectively, is used to provide a visual display and interactive control of the system's operating status; The feedback output of the decision execution module (400) is further connected to the input of the collaborative optimization module (300) to form a closed-loop optimization circuit for strategy execution.

2. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The data acquisition module (100) includes: Vibration sensor group (101) is arranged at the main oil pump shaft and bearing positions, with a sampling frequency range of [missing information]. Hz; a temperature sensor array (102) is arranged on the surface of the main oil pump housing and in the lubricating oil circuit, with a measurement accuracy of Hz. C; Pressure transmitter group (103), configured in the inlet and outlet pipe sections, with a range covering MPa.

3. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The boundary element analysis module (200) includes: a feature parameter extraction unit (210) for performing boundary integral equation calculations. ; in, For Green's function, For field variables, For the normal derivative, The boundary element is a discrete element for the pump body; the state evaluation element (220) is used to construct the three-dimensional boundary element model and calculate the characteristic parameters. ,in For the first eigenvalues ​​of order 1 This is the quality matrix.

4. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The collaborative optimization module (300) includes a multi-unit data aggregation unit (310) for constructing a unit status matrix. ,in Indicates the first Unit No. Performance metrics; dynamic programming unit (320), used to solve the objective function using an improved genetic algorithm: ; in, For the first Time-based maintenance costs For the first Unit energy efficiency deviation The weighting coefficients are used for the maintenance decision unit (330), which outputs the condition that the constraints are met. Optimal maintenance sequence .

5. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The decision execution module (400) includes: an instruction distribution unit (410) for transmitting control instructions to each unit's PLC via an industrial Ethernet protocol; and a feedback adjustment unit (420) for monitoring execution deviations. And trigger the compensation mechanism.

6. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The human-computer interaction module (500) includes: a three-dimensional visualization unit (510) for rendering the boundary element mesh model and stress cloud map; and an early warning unit (520) for displaying warnings when feature parameters exceed a threshold. Multiple alarm signals are generated in real time.

7. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The various modules of the system exchange data via the OPC UA protocol, and the boundary element discrete element size satisfies The accuracy requirements, among which It is the minimum stress wave wavelength.

8. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The vibration signal processing in the data acquisition module employs wavelet packet decomposition algorithm, with a decomposition level of [number missing]. Characteristic frequency resolution Hz.

9. The main oil pump boundary element multi-unit intelligent collaborative operation and maintenance system according to claim 1, characterized in that, The iteration termination condition of the optimization algorithm in the collaborative optimization module is the rate of change of the fitness function. It was iterated for 5 consecutive times.

Images