Disclosure of Invention
The invention aims to provide an optimized layout method for bridge monitoring nodes, which can dynamically adjust the layout scheme of the monitoring nodes according to the actual running state of a bridge, improve the quality of monitoring data, optimize the performance of a monitoring system and has remarkable innovation and practicability.
In order to achieve the above purpose, the invention provides an optimized layout method for bridge monitoring nodes, comprising the following steps:
Determining key structural information and key monitoring areas of the bridge by utilizing finite element analysis and field test;
Determining an initial monitoring node layout scheme according to key structure information of the bridge and position information of key monitoring areas, wherein the initial monitoring node layout scheme comprises types and layout positions of monitoring nodes, and ensuring main structure parts and key response points of the coverage bridge;
the method comprises the steps of collecting actual running data of a bridge in real time, wherein the actual running data comprises external excitation information and response data of a bridge structure, processing the actual running data through a data analysis module, evaluating the quality of monitoring data under the current layout scheme to obtain an evaluation result, and identifying areas with insufficient monitoring or data redundancy;
and constructing an objective function with comprehensive optimization of monitoring precision, data transmission efficiency and cost as a target, and dynamically adjusting an initial monitoring node layout scheme by using an improved particle swarm optimization algorithm in combination with an evaluation result to obtain an optimized layout scheme.
Preferably, the monitoring node type comprises a strain sensor, an acceleration sensor, a displacement sensor, an inclinometer and a temperature and humidity sensor.
Preferably, the external excitation information comprises traffic load, wind load, temperature change and humidity change, and the response data of the bridge structure comprises strain, acceleration, displacement and inclination angle.
Preferably, in the improved particle swarm optimization algorithm:
Each particle represents one possible monitoring node layout scheme, the position of the particle represents the specific coordinates of the monitoring node on the bridge, and the speed of the particle represents the amplitude and direction of node position adjustment;
the updated formula for particle position and velocity is:
;
;
Wherein,AndRespectively represent the firstThe particles are at the firstSpeed and position at the time of iteration; AndRespectively represent the firstThe particles are at the firstSpeed and position at the time of iteration; representing inertial weights;、 representing a learning factor;、 representing a random number; Represent the firstHistorical optimal positions of individual particles; representing a global optimal position;
;
;
Wherein,、Respectively representing the maximum value and the minimum value of the inertia weight; representing the current iteration number; representing a maximum number of iterations; The learning factor is represented as such,、Respectively representing the maximum value and the minimum value of the learning factor; representing the index variable.
Preferably, the improved particle swarm optimization algorithm designs a mutation operator based on node energy consumption, and perturbs the position of the particles:
;
Wherein,Represent the firstNew positions of the individual particles after disturbance; Representing the variation step length; representing a random number; representing the node position where the energy consumption is minimal; Is a mutation operator.
Preferably, the objective function is as follows:
;
Wherein,Representing an objective function;、、 Representing dynamically adjusted weight coefficients for balancing the relative importance between different targets; representing the number of critical monitoring areas; Represent the firstWeights of the key regions; Represent the firstMonitoring data quality of the key areas; representing the total number of monitoring nodes; Represent the firstThe data transmission rate of the individual nodes; Represent the firstEnergy consumption of individual nodes; Represent the firstThe layout cost of the individual nodes.
Preferably, the method comprises the steps of,、、The dynamic adjustment method of (2) is as follows:
According to the monitoring requirements and the priorities of the specific bridges, determining an initial weight value through expert experience or historical data statistics;
quality of current monitoring dataEfficiency of data transmissionAnd cost ofAs a state input, expressed as a state vector;
Defining a set of actions、、Representing an adjustment operation on the weight coefficient;
designing a reward functionTo evaluate the adjustment effect, the bonus function is expressed as:
;
Wherein,、、Representing a reward factor;、、 respectively representing the variation of the quality, the data transmission efficiency and the cost of the monitoring data;
updating weight coefficients using reinforcement learning algorithm, rewards according to current state and actionsThe weight coefficients are adjusted to maximize the long-term jackpot, and the formula is updated as:
;
;
;
Wherein,、、Representing the updated weight coefficients; The learning rate is represented, and the step length of the weight coefficient adjustment is controlled.
Preferably, the constraint of the objective function is as follows:
communication distance constraint between nodes:
;
Wherein,Represent the firstPersonal node and the firstCommunication distance between individual nodes; representing a maximum communication distance;
Coverage constraints:
;
Wherein,Representing the coverage of the monitoring node; representing the minimum coverage required;
Energy consumption constraint:
;
Wherein,Representing the maximum allowable energy consumption of the node.
Therefore, the bridge monitoring node optimizing layout method has the beneficial technical effects that:
(1) The method has strong dynamic adaptability, and the method can collect the actual running data of the bridge in real time by introducing a dynamic feedback mechanism, and dynamically adjust the layout scheme of the monitoring nodes according to the data evaluation result. The dynamic adjustment capability enables the monitoring system to adapt to the monitoring requirements of the bridge under different running states, and overcomes the defect that the layout scheme in the prior art is static and fixed and cannot adapt to the change of the actual running state of the bridge.
(2) The invention constructs an objective function with the objective of comprehensive optimization of monitoring precision, data transmission efficiency and cost, and dynamically adjusts by using an improved particle swarm optimization algorithm. Compared with the single-target optimization method in the prior art, the method can simultaneously consider a plurality of key factors, realize the improvement of monitoring precision, the optimization of data transmission efficiency and the effective control of cost, and avoid the problem of unbalanced system performance caused by single-target optimization.
(3) The optimization algorithm is efficient and reliable, the improved particle swarm optimization algorithm introduces a self-adaptive parameter adjustment mechanism and a mutation operator based on node energy consumption, so that the searching efficiency and the global convergence capacity of the algorithm are improved, and the problems that the existing optimization algorithm is easy to fall into local optimum, the convergence speed is low and the like are solved. In the optimized layout of the bridge monitoring nodes, the method means that a better node layout scheme can be found more quickly, and the response speed and the adaptability of the monitoring system are improved. Through self-adaptive adjustment of inertia weight and learning factors, the algorithm can better balance global searching and local searching capacity, so that a better and reliable optimization result is obtained, and the arrangement of monitoring nodes can be ensured to accurately meet the requirements of bridge structure health monitoring.
(4) The invention continuously collects and analyzes data in the monitoring process, can timely find out areas with insufficient monitoring or redundant data, and adjusts the layout scheme according to the areas. The problem that an effective dynamic adjustment mechanism is lacking in the prior art is solved, so that a monitoring system can continuously optimize a layout scheme according to real-time feedback information, and good monitoring performance is always maintained.
Detailed Description
The technical scheme of the invention is further described below through the attached drawings and the embodiments.
Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs.
Example 1
As shown in fig. 1, the invention provides an optimized layout method for bridge monitoring nodes, which comprises the following steps:
And step 1, modeling the bridge by utilizing finite element analysis software ANSYS. In the modeling process, factors such as actual geometric dimensions, material characteristics, load working conditions and the like of the bridge are considered, so that accuracy of the model is ensured. The key structural information of the bridge is obtained through field test, wherein the key structural information comprises a stress concentration area, a larger deformation area, vibration mode characteristics and the like. The field test adopts various means, such as strain gauge measurement stress, acceleration sensor measurement vibration response, etc. And determining the structural weak points and key monitoring areas of the bridge.
And 2, preliminarily determining the type and the approximate layout position of the monitoring node according to the structural characteristics of the bridge and the conventional monitoring requirements. The monitoring node type comprises a strain sensor, an acceleration sensor, a displacement sensor, an inclinometer, a temperature and humidity sensor and the like. The arrangement positions of the different types of sensors are arranged in a targeted manner according to the characteristics of the monitored objects. For example, strain sensors are arranged in stress concentration areas in a focused manner, and acceleration sensors are arranged at key vibration mode nodes of the bridge.
In the initial layout scheme, 120 monitoring nodes are arranged on the bridge, wherein 40 strain sensors, 30 acceleration sensors, 20 displacement sensors, 20 inclinometers and 10 temperature and humidity sensors are arranged on the bridge. These sensors cover the main structural parts of the bridge, but the preliminary layout scheme may have problems of monitoring blind areas or data redundancy.
And step 3, setting a feedback link in the monitoring system, and collecting actual running data of the bridge in real time, wherein the actual running data comprise external excitation information such as traffic load, wind load, temperature change, humidity change and the like, and response data such as strain, acceleration, displacement, inclination angle and the like of the bridge structure. And processing the data through a data analysis module, evaluating the quality and the representativeness of the monitored data under the current layout scheme, and identifying areas with insufficient monitoring or redundant data.
The specific process of the data analysis module for processing the data comprises the following steps:
filtering, denoising, normalizing and the like are carried out on the collected monitoring data so as to improve the quality and comparability of the data;
checking whether the monitored data has the condition of missing, abnormality or inconsistency, and ensuring the integrity of the data;
By comparing the monitoring data of different sensors, the accuracy and reliability of the data are evaluated;
Analyzing the acquisition and transmission time of the monitoring data, and ensuring the real-time property of the data;
Giving out comprehensive quality scores of the monitored data by combining the data integrity, accuracy and real-time evaluation results;
identifying areas with insufficient monitoring data according to the data quality scores and the monitoring requirements;
By analyzing the correlation of the different sensor data, areas of data redundancy are identified.
In actual operation, the system collects data every 30 minutes, generating about 96 sets of data per day. Through data analysis, the monitoring data of certain areas in the initial layout scheme are obviously insufficient, for example, the data fluctuation of the strain sensor is large at certain parts of the bottom of the bridge tower, and the actual stress state of the structure cannot be accurately reflected. Meanwhile, data redundancy exists in partial areas, for example, the data correlation of acceleration sensors in certain main beam areas is higher, and repeated information is more.
Step 4, constructing an objective function with comprehensive optimization of monitoring precision, data transmission efficiency and cost as targets, and combining the evaluation result, dynamically adjusting an initial monitoring node layout scheme by using an improved particle swarm optimization algorithm to obtain an optimized layout scheme;
In the improved particle swarm optimization algorithm:
Each particle represents one possible monitoring node layout scheme, the position of the particle represents the specific coordinates of the monitoring node on the bridge, and the speed of the particle represents the amplitude and direction of node position adjustment;
the updated formula for particle position and velocity is:
;
;
Wherein,AndRespectively represent the firstThe particles are at the firstThe speed and position at the time of the iteration,AndRespectively represent the firstThe particles are at the firstThe speed and position at the time of the iteration,The weight of the inertia is represented by the weight of the inertia,、The learning factor is represented as such,、The random number is represented by a number,Represent the firstThe historical optimal position of the individual particles,Representing a global optimal position;
;
;
Wherein,、Respectively representing the maximum value and the minimum value of the inertia weight; representing the current iteration number; representing a maximum number of iterations; The learning factor is represented as such,、Respectively representing the maximum value and the minimum value of the learning factor; representing the index variable.
The improved particle swarm optimization algorithm designs a mutation operator based on node energy consumption, and the position of particles is disturbed:
;
Wherein,Represent the firstNew positions of the individual particles after disturbance; Representing the variation step length; representing a random number; representing the node position where the energy consumption is minimal; Is a mutation operator.
The objective function is as follows:
;
Wherein,Representing an objective function;、、 Representing dynamically adjusted weight coefficients for balancing the relative importance between different targets; representing the number of critical monitoring areas; Represent the firstWeights of the key regions; Represent the firstMonitoring data quality of the key areas; representing the total number of monitoring nodes; Represent the firstThe data transmission rate of the individual nodes; Represent the firstEnergy consumption of individual nodes; Represent the firstThe layout cost of the individual nodes.
In the optimization process, the improved particle swarm optimization algorithm parameter is set as follows, the population scale is 50, the maximum iteration number is 200, the inertia weight is linearly decreased from 0.9 to 0.4, and the learning factor is calculated、The initial value was 2.0, linearly decreasing to 0.5, respectively. Through repeated iterative optimization, the algorithm gradually adjusts the positions and the number of the monitoring nodes so as to achieve an optimal layout scheme.
Constraints of the objective function include communication distance constraints, coverage constraints, and energy consumption constraints between nodes.
Communication distance constraint between nodes:
;
Wherein,Represent the firstPersonal node and the firstCommunication distance between individual nodes; representing a maximum communication distance;
Coverage constraints:
;
Wherein,Representing the coverage of the monitoring node; representing the minimum coverage required;
Energy consumption constraint:
;
Wherein,Representing the maximum allowable energy consumption of the node.
、、The dynamic adjustment method of (2) is as follows:
According to the monitoring requirements and the priorities of the specific bridges, determining an initial weight value through expert experience or historical data statistics;
quality of current monitoring dataEfficiency of data transmissionAnd cost ofAs a state input, expressed as a state vector;
Defining a set of actions、、Representing an adjustment operation on the weight coefficient;
designing a reward functionTo evaluate the adjustment effect, the bonus function is expressed as:
;
Wherein,、、Representing a reward factor;、、 Respectively representing the variation of the quality, the transmission efficiency and the cost of the monitored data;
updating weight coefficients using reinforcement learning algorithm, rewards according to current state and actionsThe weight coefficients are adjusted to maximize the long-term jackpot, and the formula is updated as:
;
;
;
Wherein,、、Representing the updated weight coefficients; The learning rate is represented, and the step length of the weight coefficient adjustment is controlled.
And 5, applying the optimized layout scheme to actual bridge monitoring, and verifying the effectiveness of the optimized layout method by comparing the quality of monitoring data before and after implementation and the system operation efficiency.
As shown in table 1, in the optimized layout scheme, the total number of monitoring nodes is reduced to 100, wherein 30 strain sensors, 25 acceleration sensors, 20 displacement sensors, 15 inclinometers and 10 temperature and humidity sensors are arranged. Through practical application, the optimized scheme obviously improves the quality of the monitored data of the key area, the data integrity is improved from 85% to 95%, and the data accuracy is improved from 90% to 97%. Meanwhile, the data transmission efficiency is improved by 30%, the system energy consumption is reduced by 25%, and the cost is reduced by 20%.
Table 1 comparison before and after optimization
;
As shown in fig. 2, the bridge monitoring node optimizing layout system includes:
The finite element analysis module is used for modeling and analyzing the bridge and determining key structural information of the bridge;
the field test module is used for acquiring actual structural information of the bridge, including a stress concentration area, a larger deformation area, vibration mode characteristics and the like;
The initial layout scheme generation module preliminarily determines the type and the approximate layout position of the monitoring node according to the structural characteristics of the bridge and the conventional monitoring requirements;
the monitoring node layout module is responsible for arranging various types of monitoring nodes on the bridge according to an initial layout scheme, wherein the monitoring nodes comprise a strain sensor, an acceleration sensor, a displacement sensor, an inclinometer and a temperature and humidity sensor;
The data acquisition module is used for acquiring data of the monitoring nodes in real time, wherein the data comprise external excitation information and response data of the bridge structure;
The data analysis module is used for processing the acquired data and evaluating the quality and representativeness of the monitored data;
The data quality evaluation module evaluates the quality of the monitoring data under the current layout scheme according to the integrity, accuracy and real-time property of the data;
The optimization decision module decides whether the layout scheme needs to be optimally adjusted according to the data quality evaluation result;
the improved particle swarm optimization algorithm module is used for optimizing and adjusting an initial monitoring node layout scheme by combining a dynamic adjustment weight coefficient by using an improved particle swarm optimization algorithm so as to improve monitoring precision, data transmission efficiency and control cost;
the layout scheme adjusting module dynamically adjusts the layout scheme of the monitoring node according to the instruction of the optimization decision module and the calculation result of the improved particle swarm optimization algorithm module;
and the comparison and verification module is used for verifying the effectiveness of the optimized layout method by comparing the quality of the monitoring data and the running efficiency of the system before and after implementation.
It is noted that what is not described in detail in the present invention is well known to those skilled in the art.
Therefore, the bridge monitoring node optimal layout method can dynamically adjust the monitoring node layout scheme according to the actual running state of the bridge, improve the quality of monitoring data and optimize the performance of a monitoring system.
It should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted by the same, and the modified or substituted technical solution may not deviate from the spirit and scope of the technical solution of the present invention.