CN116804534B

CN116804534B - Auxiliary determination method for continuous beam bridge construction process

Info

Publication number: CN116804534B
Application number: CN202310778829.3A
Authority: CN
Inventors: 张大军; 王晓峰; 佟立春; 王欢; 吕振雷; 游艳文; 刘岱鑫
Original assignee: Cccc Changsha Construction Co ltd; CCCC Second Harbor Engineering Co
Current assignee: Cccc Changsha Construction Co ltd; CCCC Second Harbor Engineering Co
Priority date: 2023-06-29
Filing date: 2023-06-29
Publication date: 2024-03-12
Anticipated expiration: 2043-06-29
Also published as: CN116804534A

Abstract

The invention discloses an auxiliary measurement method for a continuous beam bridge construction process, which further provides an integrated control framework on the basis of traditional optical measurement, and can realize more accurate and real-time monitoring and prediction of the sag disturbance problem in the continuous beam bridge construction process through the cooperation of cellular automata, kalman filtering, support vector machines and genetic algorithms; 1. real-time monitoring and early warning capabilities: the deformation state of the bridge can be monitored and analyzed in real time by using a cellular automaton model and a related algorithm. This real-time monitoring capability allows problems to be discovered and identified early, and corresponding measures to repair or maintain in time to prevent potential safety risks. 2. Automated data processing and analysis: the measurement data can be automatically processed and analyzed through the application of algorithms such as Kalman filtering, a support vector machine, a genetic algorithm and the like.

Description

Auxiliary determination method for continuous beam bridge construction process

Technical Field

The invention relates to the technical field of continuous beam bridges, in particular to a process auxiliary measuring method for continuous beam bridge construction.

Background

A continuous beam bridge is a bridge structure characterized by a continuous span construction structure when crossing a river, road or other topographical obstacle. Because of the large design span, the continuous beam bridge is often subjected to the problem of sag disturbance in the construction process.

The sag disturbance refers to deformation of the bridge in the construction process caused by the factors of dead weight, construction load and the like, which is mainly expressed as vertical displacement and inclination of the beam, in a state that part of the construction section is suspended in the air and the bottom of the construction section is free of a supporting structure because the continuous bridge is temporarily not constructed in the construction process. Such deformations can negatively affect the structural stability and the safety of the bridge.

In order to detect and monitor the sag disturbance problem in the construction process of a continuous beam bridge, the prior art generally adopts a continuous optical instrument to perform periodic measurement in the construction process, and feeds data back to a construction decision maker as a reference, and further determines repair and maintenance decisions. The optical instruments can measure the bridge in different construction time periods, acquire deformation data, and judge the stability and deformation condition of the bridge through analysis and comparison.

However, there are some drawbacks to using continuous optical instruments to perform measurements in the prior art. Because of the large volume and large span of the continuous beam bridge, the method needs to measure the bridge for multiple times in different time periods and different construction sections, but the optical measurement can only provide discrete measurement data, and has a certain limit on continuous monitoring of bridge deformation. In addition, the mere reliance on optical measurements does not provide an immediate deformation state estimate, limiting the ability to monitor bridge deformation in real time.

Therefore, an auxiliary measuring method for the construction process of the continuous beam bridge is provided.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a method for assisting in determining a construction process of a continuous beam bridge, so as to solve or alleviate the technical problem existing in the prior art, that is, certain limitations exist in discrete measurement data, and at least provide a beneficial choice for the method;

the technical scheme of the embodiment of the invention is realized as follows:

first aspect

The auxiliary measuring method for the continuous beam bridge construction process further provides an integrated control framework on the basis of traditional optical measurement, and the sag problem in the continuous beam bridge construction process can be accurately and timely monitored and predicted through the cooperation of cellular automata, kalman filtering, a support vector machine and a genetic algorithm. The comprehensive algorithm combines cell state updating rules of bridge behaviors and deformation conditions, utilizes Kalman filtering to estimate local deformation states, carries out anomaly detection through a support vector machine, and optimizes a linear control strategy through a genetic algorithm so as to reduce deformation and improve structural stability.

The logic of the above measurement method is as follows:

STEP1, arranging a plurality of measuring points on the bridge, and simultaneously measuring by using a plurality of optical instruments to obtain deformation data of different positions of the bridge.

In the above-described embodiments, in order to monitor the sag problem of the continuous beam bridge, it is necessary to arrange a plurality of measuring points on the bridge and perform measurement using an optical instrument. The measuring points cover different positions of the bridge, and deformation data of the bridge in the construction process can be obtained.

It should be noted that deformation data of different positions of the bridge can be obtained by arranging a plurality of measuring points on the bridge and measuring by using a plurality of optical instruments. The optical instruments can measure the bridge through technical means such as optical sensors and the like, and deformation data such as displacement, inclination, deflection and the like are obtained.

STEP2, creating a cellular automaton model, wherein each cell represents a location of the bridge.

In the above embodiment, in order to model and analyze the behavior and deformation of the bridge, a cellular automaton model needs to be created, where each cell represents a position of the bridge.

It should be noted that cellular automata is a modeling method based on discrete space and time, where each cell represents a local state of the system. In this step, the invention divides the bridge into several locations, one for each cell. And the cells interact and update the state through the neighbor relation.

STEP3, describing the state updating rule of the cells through a linear function according to the behavior and deformation condition of the bridge and combining the factors of the positions of the cells, the states of the neighbor cells, the material characteristics and the external load.

In the above embodiment, in order to accurately describe the state update rule of the bridge cells, the behavior and deformation of the bridge, and factors such as the cell position, the state of the neighboring cells, the material characteristics, and the external load need to be considered.

It should be noted that, in the cellular automaton model, the state update rule of each cell is described by taking the behavior and deformation condition of the bridge into consideration, and using a linear function in combination with factors such as the position of the cell, the state of the neighbor cell, the material characteristics, and the external load.

CellState(t+1)＝f(CellState(t),NeighborStates(t),BridgeProperties,ExternalLoad)

Wherein, cellState (t) represents the state of a cell at a time t, neighborStates (t) represents the state of a neighbor cell at a time t, bridgeProperties represents the material property of the bridge, and ExternalLoad represents the external load. The function f represents a linear functional relationship for calculating the state of the cell at the next moment.

STEP4: for each cell: and defining the local deformation state of the bridge as the problem of Kalman filtering, filtering the measurement data and obtaining the bridge deformation state measurement.

In the above-described embodiment, in order to measure the local deformation state of the bridge, it is necessary to define the local deformation state as a problem of kalman filtering and perform filtering processing on the measurement data.

It should be noted that the kalman filter is a filtering method for estimating the state of a system, and by fusing the dynamic model of the system and measurement data, the optimal estimation of the state of the system can be obtained. In this step, for each cell, the local deformation state of the bridge is defined as a problem of kalman filtering, and filtering processing is performed using the measurement data to obtain a deformation state measurement of the bridge.

The Kalman filtering algorithm formula may be expressed as:

StateEstimation(t+1)＝A*StateEstimation(t)+B*ControlInput(t)

MeasurementEstimation(t)＝C*StateEstimation(t)

KalmanGain(t)＝P(t)*C^T*(C*P(t)*C^T+R)^(-1)

StateEstimation(t+1)

＝StateEstimation(t)+KalmanGain(t)*(Measurement(t)-MeasurementEstimation(t))

P(t+1)＝(I-KalmanGain(t)*C)*P(t)

where StateEstimation (t) denotes the state estimate at time t, measurement (t) denotes the measurement values at time t, A and B are the state transition matrices, C is the measurement matrix, controlInput (t) is the control input, P (t) is the state covariance matrix, R is the measurement noise covariance matrix, and I is the identity matrix.

STEP5: for each cell: using the filtered data and combining the position, shape and material characteristic information of the measuring points by using a support vector machine as a feature vector; and identifying and detecting abnormal conditions in bridge deformation.

In the above-described embodiments, in order to identify and detect an abnormal situation in bridge deformation, it is necessary to use filtered data and perform abnormality detection in combination with the use of a support vector machine, and classify the abnormal situation by using the position, shape, and material property information of the measurement points as feature vectors.

It should be noted that the support vector machine is a machine learning algorithm that can be used for classification and regression problems. In the step, for each cell, the filtered data is used as input, the position, shape and material characteristic information of the measuring point are used as feature vectors, and the deformation condition of the bridge can be classified, identified and detected by training a support vector machine model.

STEP6: for each cell: and searching an optimal solution by using filtered data and defining a bridge linear control strategy problem, and searching the optimal linear control strategy by using a genetic algorithm through cyclic iteration.

In the above embodiment, in order to determine the optimal linear control strategy of the bridge, the filtered data is required to be used and searched in combination with a genetic algorithm to find an optimal solution.

It should be noted that the genetic algorithm is an optimization algorithm, and continuously optimizes the solution of the problem by simulating the evolution process in nature and through operations such as selection, crossover, mutation and the like. In this step, for each cell, the filtered data is used as input, and in combination with the defined bridge linear control strategy problem, a genetic algorithm is used to search for an optimal solution, and the linear control strategy is gradually optimized by a loop iteration manner, so as to achieve an optimal bridge shape.

Second aspect

A computer device comprising a processor, a memory coupled to the processor, the memory having stored therein program instructions that, when executed by the processor, cause the processor to perform the assay method as described above.

The computer device is a device for executing the above STEP1 to STEP 6 measurement method. It comprises the following main components:

(1) A processor: a computer device is provided with a processor for executing instructions and controlling the operation of the device. The processor is responsible for executing algorithms in STEP 1-6 measuring methods, processing measurement data, and performing calculation operations such as state updating, filtering, anomaly detection, optimization control and the like.

(2) A memory: memory coupled to the processor is a computer device for storing program instructions and data. Program instructions required by the STEP 1-6 measuring method after the STEP are stored in a memory, including the establishment of a cellular automaton model, the definition of a state updating rule, the realization of Kalman filtering, the training and classification of a support vector machine, the search of a genetic algorithm and the like.

(3) Input and output interfaces: the computer device also includes input and output interfaces for data interaction with external devices. Through the input interface, the measurement data may be input into the computer device for processing by the processor. Through the output interface, the computer device may output the processing results to an external device, such as a display or other data storage device.

By providing a processor and a memory and executing stored program instructions, the computer equipment can automatically execute the STEP 1-6 measuring method to realize monitoring and control of bridge deformation. The bridge deformation measuring method and device are high in efficiency, accurate and automatic, flexible parameter setting, optimization and adjustment can be carried out according to actual requirements, and structural safety and operation stability of the bridge are guaranteed.

Third aspect of the invention

A storage medium storing program instructions capable of implementing the above-described measurement method.

The storage medium is a medium for storing program instructions for implementing the STEP1 to STEP 6 measurement method. It may be various types of non-volatile memory such as hard disk drives, solid state drives, flash drives, etc., or programmable memory such as flash memory chips, EEPROM chips, etc.

In the storage medium, programmed and compiled program instructions are stored that describe the STEPs and algorithms in the following STEP 1-6 assay methods. Specifically, the storage medium stores a process of creating a cellular automaton model, a definition of a cellular state update rule, a method of implementing kalman filtering, a training and classifying algorithm of a support vector machine, a searching process of a genetic algorithm, and the like.

By connecting the storage medium to a computer device or other executable device, program instructions may be read from the storage medium and executed by a processor of the device. In this way, the storage medium provides a convenient way for program instructions that enable the following STEP 1-6 assay methods to be identified and executed by a computer device or other executable device.

The use of the storage medium makes the implemented STEP 1-6 measurement method more convenient and flexible. By updating program instructions in the storage medium, the algorithm can be improved and optimized to adapt to different bridge monitoring requirements and technical development. Meanwhile, the reliability and stability of the storage medium can also ensure the safe storage and reading of the program instructions so as to ensure the reliability and accuracy of the measuring method.

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

1. real-time monitoring and early warning capabilities: the deformation state of the bridge can be monitored and analyzed in real time by using a cellular automaton model and a related algorithm. This real-time monitoring capability allows problems to be discovered and identified early, and corresponding measures to repair or maintain in time to prevent potential safety risks.

2. Automated data processing and analysis: the measurement data can be automatically processed and analyzed through the application of algorithms such as Kalman filtering, a support vector machine, a genetic algorithm and the like. The automatic processing and analyzing capability greatly improves the utilization efficiency and accuracy of the data, reduces the mistakes and subjectivity of human intervention and improves the overall data processing efficiency.

3. Fault detection and optimization of line control: by applying an abnormality detection algorithm and a genetic algorithm, abnormal conditions in bridge deformation can be timely identified and detected, and a corresponding linear control strategy is adopted. Therefore, the safety and stability of the bridge can be improved, the linear control strategy is optimized, and the influence of human factors and subjective judgment is reduced.

4. Measurement of comprehensiveness and accuracy: deformation data of different positions of the bridge can be obtained by arranging a plurality of measuring points and simultaneously measuring by using a plurality of optical instruments. The measuring method can provide comprehensive and accurate data, and the deformation condition of the bridge can be better known.

Compared with the prior art, the invention solves the defects of the traditional technology as follows:

1. limitations of conventional measurement methods: the traditional method is often only capable of measuring by adopting limited measuring points, so that the deformation condition of the bridge cannot be comprehensively known. The measuring method solves the defect through the cooperation of multipoint measurement and a plurality of instruments, and provides more comprehensive and accurate data.

2. Limitations of manual handling and analysis: the traditional method needs to process and analyze the data manually, and has the problems of subjectivity and human intervention. The application of the data processing and analyzing algorithm realizes automatic processing and analysis, and improves the accuracy and the utilization efficiency of data.

3. Lack of real-time monitoring and early warning capabilities: the traditional method often lacks real-time monitoring and early warning capability, and can not timely find and identify the abnormal condition of bridge deformation. The application of the algorithm and the model realizes real-time monitoring and early warning, and improves the safety and the maintenance efficiency.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the technical descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a logic diagram of a workflow of the present invention;

FIG. 2 is a control program diagram (first portion) of a seventh embodiment of the present invention;

Fig. 3 is a control program diagram (second portion) of a seventh embodiment of the present invention.

Detailed Description

In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit of the invention, whereby the invention is not limited to the specific embodiments disclosed below;

referring to fig. 1, the following applies to this embodiment:

In the continuous beam bridge construction process, deformation data of different positions of the bridge can be obtained by arranging a plurality of measuring points on the bridge and measuring by using a plurality of optical instruments. The measurement data can reflect deformation conditions such as displacement, inclination, deflection and the like of the bridge at different positions.

For modeling and analysis of bridges, the present embodiment creates a cellular automaton model in which each cell represents a location of a bridge. By dividing the bridge into a series of cells and describing and updating the state of each cell, a global bridge model can be built.

In the cellular automaton model, the specific embodiment describes the cell state updating rule through a linear function according to the behavior and deformation conditions of the bridge and by combining the factors such as the cell position, the neighbor cell state, the material characteristics, the external load and the like. Thus, the state of each cell can be updated in each iteration according to the current state of the bridge and external factors.

For each cell, the present embodiment defines the local deformation state of the bridge as a problem of kalman filtering. The deformation state measurement of the bridge can be obtained by filtering the measurement data. The Kalman filtering can estimate the current bridge deformation state by considering the past measurement data and the prediction of the model, thereby improving the accuracy and stability of the data.

For each cell, the present embodiment uses the filtered data in combination with a Support Vector Machine (SVM) algorithm to identify and detect anomalies. By taking the position, shape and material characteristic information of the measuring point as feature vectors, the specific embodiment can train a support vector machine model for judging whether abnormal conditions exist in bridge deformation. This can help the monitoring personnel to find potential problems in time and take corresponding maintenance measures.

For each cell, the present embodiment uses the filtered data to search for an optimal solution using a genetic algorithm in combination with defining a bridge linear control strategy problem. Through a cyclic iteration mode, the genetic algorithm can gradually optimize the linear control strategy of the bridge so as to achieve better structural stability and performance. The genetic algorithm can find the optimal solution by continuously evolving and selecting excellent linear control strategies, so that the overall performance of the bridge is improved.

Through the application of the STEPs, STEP 1-6 can realize real-time monitoring and early warning in continuous beam bridge construction, and the safety and maintenance efficiency of the bridge structure are improved. By arranging a plurality of measuring points and measuring by using an optical instrument and combining a cellular automaton model and various algorithms, the deformation condition of the bridge can be accurately captured, and the abnormality can be timely found. Meanwhile, the technical means such as filtering, a support vector machine and a genetic algorithm are used, so that the accuracy of data processing and the optimization effect of a linear control strategy can be improved. Therefore, the problems of detection delay, inaccurate data processing, difficulty in linear control strategy and the like in the traditional technology can be effectively solved, and more reliable support is provided for continuous beam bridge construction.

The technical features of the above-described embodiments may be combined in any manner, and for brevity, all of the possible combinations of the technical features of the above-described embodiments may not be described, however, they should be considered as the scope of the present description as long as there is no contradiction between the combinations of the technical features.

Example 1

In order that the above-recited embodiments of the invention may be understood in detail, a more particular description of the invention, briefly summarized below, may be had by way of example. The present invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit of the invention, so that the invention is not limited to the embodiments disclosed below.

In this embodiment, the application deduction description and the display are performed based on the auxiliary measurement method for the continuous beam bridge construction process provided by the specific embodiment, wherein:

STEP1, arranging a plurality of measuring points on a bridge, and simultaneously measuring by using optical instruments corresponding to the number of the measuring points:

1.1, selecting positions on a bridge to arrange measuring points, wherein the distance between each measuring point is uniformly distributed;

1.2, configuring an optical instrument, and measuring a measuring point at the same time;

1.3, measuring each measuring point by using an optical instrument to obtain the state data of the measuring point.

In this embodiment, the purpose of this step is to arrange a plurality of measuring points and measure them using an optical instrument in continuous bridge construction to obtain deformation data of different positions of the bridge. Through arrangement and measurement of the measuring points, deformation conditions such as displacement, inclination, deflection and the like of the bridge can be monitored in real time, so that potential problems can be found in time, and the safety and quality of the construction process are ensured.

In this embodiment, the principle of this step is to select a location on the bridge to arrange the measuring point, and the structural characteristics and construction requirements of the bridge are generally considered to ensure the representativeness and effectiveness of the measuring point. The site locations are typically selected at key locations or representative areas of the bridge in a uniformly distributed manner.

In this embodiment, the specific operation steps are as follows:

a. the measuring points are arranged at selected positions on the bridge, so that the key parts or representative areas are covered.

b. And the optical instruments corresponding to the number of the measuring points are configured, so that each measuring point can be measured by the optical instruments at the same time. The optical instrument can adopt devices such as a laser range finder, a total station and the like, and has high-precision and high-efficiency measurement capability.

c. And measuring each measuring point by using an optical instrument to acquire the state data of the measuring point. Deformation information such as displacement, inclination, deflection and the like of the measuring point can be obtained through an optical measurement technology. The measurement data may be recorded in digital form for subsequent data processing and analysis.

It will be appreciated that in this embodiment, by arranging a plurality of measuring points and measuring using an optical instrument during continuous bridge construction, the following functions can be achieved:

(1) And (3) real-time monitoring: through the arrangement of a plurality of measuring points and the simultaneous measurement of optical instruments, the deformation data of different positions of the bridge can be obtained in real time, and the deformation condition of the bridge can be known in time.

(2) Deformation analysis: the deformation conditions such as displacement, inclination and deflection of the bridge can be analyzed through the state data of the measuring points, and the change trend and characteristics of the bridge structure can be known.

(3) Abnormality detection: by comparing and analyzing the measuring point data, abnormal conditions in bridge deformation, such as deformation or speed exceeding a preset range, can be found, so that measures can be taken in time to repair or adjust the bridge.

(4) And (3) quality control: by collecting and monitoring the measuring point data, the quality of bridge construction can be evaluated, problems in construction can be found and corrected in time, and the safety and stability of the bridge are ensured.

The above examples merely illustrate embodiments of the invention that are specific and detailed for the relevant practical applications, but are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Example two

The present embodiment is linked to embodiment one:

STEP2, cellular automaton initialization:

2.1 initializing a grid structure of a cellular automaton, taking each measuring point as a cell, and determining neighbor relations among the cells;

2.2 setting an initial state value and an attribute value:

for each station i:

initialization state: s_i=0;

initializing the state of the optical instrument: s_i≡k=0 for all k;

cell state update:

for each station i: calculating states and attributes of neighbor measuring points:

N_i＝{S_j,P_j}

neighbor measurement points where j is i;

calculating an external load: l_i=f (p_i).

The aim of this embodiment is to initialize cellular automata, including initializing a mesh structure and determining neighbor relations between cells. Meanwhile, an initial state value and an attribute value are set, so that preparation is made for subsequent state updating and attribute calculation.

The operation flow of this embodiment is as follows:

a. initializing a grid structure: and taking each measuring point as a cell, and arranging the cells on the bridge according to a certain rule to form a grid structure. The lattice structure may be two-dimensional or three-dimensional, depending on the geometry of the bridge and the dimensions to be monitored. Specifically, if a bridge structure in a 'face form' such as a bridge deck is to be detected, two dimensions are selected. If a bridge mechanism in a 'body form' such as a bridge pier is to be detected, three-dimensional bridge mechanisms are selected. But the principle is the same whether two or three dimensions are chosen.

b. Determining neighbor relation between cells: from the grid structure, the neighbor cells of each cell, typically the cells adjacent thereto, are determined. The determination of the neighbor relation may be defined in terms of the spatial distance or relative position between the cells.

In this embodiment, the logic of the algorithm formula is as follows:

a. initialization state value and attribute value:

initialization state: the state s_i of each cell is initialized to 0, representing the initial state.

Initializing the state of the optical instrument: for each measurement point i, initializing its corresponding optical instrument state S_i≡k to 0, where k represents the number of the optical instrument.

b. Cell state update: for each station i, its own state update is calculated from the state and attributes of its neighbor stations.

Neighbor state and attributes: n_i is used to represent the neighbor set of station i, where the state and attribute of each neighbor station j is denoted as S_j and P_j.

External load calculation: according to the attribute P_i of the measuring point i, calculating to obtain an external load L_i through a function f (P_i).

The function of this step is to initialize the cellular automaton, providing initial values for subsequent state updates and attribute calculations. The specific functions include:

(1) Initializing a grid structure: and constructing a cellular automaton model of the continuous beam bridge by taking the measuring points as cells and determining neighbor relations, and providing a basis for subsequent state updating.

(2) State value and attribute value initialization: by setting the initial state value and the attribute value, initial conditions are provided for the state update and attribute calculation of the cells.

(3) Neighbor state and attribute calculation: and acquiring the state and attribute information of the neighbor cells according to the neighbor relation among the cells, and taking the state and attribute information as references for state updating and attribute calculation.

(4) External load calculation: according to the attribute information of the cells, calculating to obtain external load through a function f (P_i) for simulating the influence of external influence on the state of the cells.

By executing STEP2, accurate initial state and attribute information can be provided for subsequent STEPs, and a foundation is provided for deformation monitoring and control of continuous beam bridge construction.

Exemplary: it is assumed that there is a continuous bridge with 6 stations numbered 1 to 6. This embodiment will use cellular automata to simulate this scenario.

Step 1: initializing a grid structure and a neighbor relation:

in this step the mesh structure of the stations on the bridge is initialized and the neighbour relation of each station is determined. As can be seen from the description of the first embodiment, the distances between the set points are equal, and the neighbor relation is as follows:

step 2: setting an initial state value and an attribute value:

In this step, an initial state value and an attribute value are set for each measurement point.

Assume that the initial state value (S) and the attribute value (P) are as follows:

measuring point number	Initial state value (S)	Initial attribute value (P)
			1	0	1
2	0	2
			3	0	3
4	0	4
			5	0	5
6	0	6

Step 3: cell state update:

in which a status update for each station is calculated. The cell state update rule is set as follows (using a linear function):

S_i＝a*S_j+b*P_j

wherein a and b are constants. Let a=0.5 and b=0.2. The cell state update results are as follows:

cell simulation diagram example:

measuring point number	Update status value (S)	Neighbor state value (S_j)	Neighbor attribute value (P_j)
				1	0.2	0	1
2	0.8	0.2	2
				3	1.4	0.8	3
4	2.2	1.4	4
				5	3.2	2.2	5
6	4	3.2	6

In this example, the present embodiment shows the state value and attribute value of each measurement point, and the state update value calculated by the cell state update rule.

Example III

The present embodiment is linked to embodiment two:

STEP3, cell state update rule:

S_i(t)＝α*S_i(t-1)+β*∑(S_j(t-1)*w_ij)+j_i+δ*F_i

wherein S_i (t) is the state value of cell i at time t;

s_j (t-1) is the state value of the neighbor cell j at the time t-1;

w_ij is a neighbor relation weight;

m_i is the material property of cell i;

f_i is an external load;

alpha, beta, gamma and delta are weight coefficients;

setting a threshold value of a cell convergence condition: epsilon.

The purpose of this embodiment is to update the state value of each cell in the continuous beam bridge to reflect its interactions with neighboring cells, material properties and external loads. By applying the cell state updating rule, the deformation condition of the bridge at different moments can be simulated, and further deformation prediction and control can be performed.

In this embodiment, each cell represents a position of the bridge during construction of the continuous beam bridge, and the cell state update rule is based on the state, neighbor relation weight, material characteristics, and external load of the neighbor cells of the position. The influence degree of different factors on the cell state can be controlled by adjusting the weight coefficient, so that the behavior and deformation condition of the bridge can be described more accurately.

Specifically, in the cell state update rule, the logic of the algorithm formula is as follows:

the first term (α×s_i (t-1)) represents the maintenance of the state of the cell itself, and the state value at the previous time is maintained by multiplying the weight coefficient α.

The second term (β Σ (s_j (t-1) w_ij)) represents the effect of the neighbor cell on the current cell state, taking into account the contribution of the neighbor cell by adding to the state value of the neighbor cell (multiplied by the weight coefficient β).

The third term (γ×m_i) represents the influence of the material properties of the cells on the state, and factors such as rigidity and elasticity of the material are taken into consideration by multiplying the material property coefficient γ.

The fourth term (δ×f_i) represents the influence of external load on the state, and the load condition during the construction process is considered by multiplying the external load factor δ.

The cell state update rules integrate the effects of self state, neighbor state, material properties, and external loading to more fully describe the state changes of the cells.

Further, the definition rules of the weight coefficient α, the rigidity and elasticity of the material characteristic coefficient γ, and the external load coefficient δ can be defined according to specific engineering requirements and bridge characteristics.

(1) Weight coefficient α:

rule 1: the state at the last moment is kept unchanged. α=1.

Rule 2: the state at the previous moment is gradually attenuated to reflect the evolution of the cell state. Alpha may vary between 0 and 1, e.g., alpha decreases with time, indicating that the cell state is increasingly affected by the state at the current time.

(2) Material characteristic coefficient γ:

rule 1: an elastic material. γ=1, which indicates that the material properties have an overall effect on the cell state.

Rule 2: a rigid material. γ=0, indicating that the material properties have no effect on the cell state.

(3) External load factor δ:

rule 1: the external load is uniformly applied. δ=1, which indicates that the external load has an overall effect on the cell state.

Rule 2: no external load. δ=0, indicating that the external load has no effect on the cell state.

It should be noted that the specific definition rules should be determined according to actual engineering situations and design requirements. The appropriate coefficient values can be selected to meet specific engineering requirements based on the material properties, load conditions and design goals of the bridge, or in combination with the material properties described in the technical manuals commonly used in the construction arts, in combination with the form of finite element modeling.

Exemplary: the present embodiment is provided with a continuous beam bridge consisting of 5 measuring points, and is derived according to the following steps:

The following parameter values were set:

neighbor relation weight w_ij:

w_12＝0.5

w_23＝0.8

w_34＝0.6

w_45＝0.7

other neighbor relation weight of 0

Material properties m_i:

M_1＝1.2

M_2＝1.5

M_3＝1.0

M_4＝1.3

M_5＝1.1

external load f_i:

F_1＝2.0

F_2＝1.8

F_3＝1.5

F_4＝1.7

F_5＝1.6

the weight coefficient α=0.6, β=0.4, γ=0.8, δ=0.2

Convergence Condition threshold ε=0.001

Initializing a state value:

S_1(0)＝0.0

S_2(0)＝0.0

S_3(0)＝0.0

S_4(0)＝0.0

S_5(0)＝0.0

calculating a cell state value according to a cell state updating rule:

iterating until convergence: t=1:

S_1(1)＝α*S_1(0)+β*(S_2(0)*w_12)+γ*M_1+δ*F_1

S_2(1)＝α*S_2(0)+β*(S_1(0)*w_21+S_3(0)*w_23)+γ*M_2+δ*F_2

S_3(1)＝α*S_3(0)+β*(S_2(0)*w_32+S_4(0)*w_34)+γ*M_3+δ*F_3

S_4(1)＝α*S_4(0)+γ*(S_3(0)*w_43+S_5(0)*w_45)+γ*M_4+δ*F_4

S_5(1)＝α*S_5(0)+β*(S_4(0)*w_54)+γ*M_5+δ*F_5

in the table, each column represents a time, and each row represents a site cell. The initial state S_i (0) is 0, then the iterative computation is carried out according to a formula to obtain S_i (1), the second iterative computation is carried out to obtain S_i (2), and so on. Until the convergence condition is satisfied, a final state value can be obtained.

t＝2:

Repeating the above calculation until meeting the convergence condition

Checking and calculating: the change in the state value of each cell is checked until the state value change of all cells does not exceed the convergence condition threshold epsilon. If the condition is satisfied, stopping iteration, otherwise continuing iteration.

Analysis of results: according to the final state value calculated in the iterative process, the states of the continuous beam bridge at different measuring points can be analyzed. A larger state value indicates that there is a larger deformation or stress at that location, and a smaller state value indicates that it is relatively stable.

The function of this embodiment is to update the state value of each cell in the continuous bridge according to the cell state update rule. By applying the rule through multiple iterations, the deformation process of the bridge at different moments can be simulated, and deformation data of different positions of the bridge can be obtained. The data can be used for deformation prediction, anomaly detection, linear control and other applications, so that the safety and reliability of continuous beam bridge construction are improved.

Example IV

This embodiment is linked to embodiment three:

STEP4, kalman filtering is matched with cellular automata:

4.1 defining a problem state vector:

X_i＝[d_i,theta_i,phi_i]

where d_i is displacement, theta_i is inclination angle, phi_i is deflection;

4.2 defining an observation vector:

Z_i＝[M_i^k]

wherein k is the number of optical instruments of all measuring points i;

4.3 define a state transition matrix: f_i;

defining an observation matrix: h_i;

4.4 defining a process noise covariance matrix: q_i;

defining an observed noise covariance matrix: r_i;

4.5 define an initial state estimation vector: x_i (0);

defining an initial state covariance matrix: p_i (0);

4.6, kalman filtering estimation and prediction:

4.6.1 prediction:

X_i(t)＝F_i*X_i(t-1)P_i(t)＝F_i*P_i(t-1)*F_i^T+Q_i

4.6.2 updating:

K_i(t)＝P_i(t)*H_i^T*(H_i*P_i(t)*H_i^T+R_i)^-1

X_i(t)＝X_i(t)+K_i(t)*(Z_i-H_i*X_i(t))

P_i(t)＝(I-K_i(t)*H_i)*P_i(t)

4.7, obtaining an estimated value of the local deformation state of the bridge: x_i (t).

In the embodiment, the main principle is to predict the local deformation state of the bridge, including displacement, inclination angle and deflection, and dynamically monitor and control the deformation behavior in the construction of the continuous bridge by combining with cellular automata.

Further, kalman filtering is a recursive, optimal state estimation algorithm that predicts and updates states using a linear dynamic model and a linear observation model. In continuous beam bridge construction, the local deformation state of the bridge can be estimated and predicted by using Kalman filtering by defining a state vector and an observation vector and combining a state transition matrix, an observation matrix, a noise covariance matrix, an initial state estimation vector and a covariance matrix.

In this embodiment, the principle thereof includes: a problem state vector x_i is defined, which includes information of displacement, inclination angle and deflection, in order to describe the local deformation state of the bridge.

An observation vector z_i is defined, comprising the optical instrument measurements of the measurement points, for observing and updating the state vector.

A state transition matrix f_i and an observation matrix h_i are defined for describing the relationship between the state vector and the observation vector. The specific definition rules can be determined according to the physical characteristics and the motion equation of the bridge. For example, a finite element method based on elastic mechanics theory may be used to build a state transition matrix, and the state vector of the previous time is mapped to the state vector of the current time.

A process noise covariance matrix q_i and an observation noise covariance matrix r_i are defined for describing noise in the state transition and observation processes. The process noise covariance matrix describes the uncertainty and noise of the bridge local deformation state due to external factors during state transition. The specific definition rules can be determined according to the actual scene and the statistical characteristics of the measurement errors.

An initial state estimation vector x_i (0) and an initial state covariance matrix p_i (0) are defined for initializing state estimation and prediction. The initial state estimation vector represents an initial estimation of the local deformation state of the bridge at the start time. Specific definition rules are determined from historical data. The initial state estimation vector can be used as an initial condition of the bridge to provide a reasonable set of initial estimation values.

According to the estimation and prediction steps of Kalman filtering, carrying out state prediction and updating:

and a prediction step: and predicting the state at the last moment according to the state transition matrix and the noise covariance matrix.

Updating: and updating the state through the observation matrix, the noise covariance matrix and the predicted state to obtain a more accurate state estimation value.

Finally, an estimated value X_i (t) of the local deformation state of the bridge is obtained, wherein the estimated value X_i represents the displacement, the inclination angle and the deflection of the bridge at different measuring points, and the estimated value X_i (t) is specifically:

(1) The estimated value X_i (t) of the local deformation state of the bridge consists of three components of a state vector X_i, which respectively represent the displacement, the inclination angle and the deflection of the beam at different measuring points.

(2) The specific representation may be determined according to specific measurement methods and station arrangements. For example, the displacement may be represented by horizontal and vertical displacement of the measurement point, the inclination angle may be represented by an inclination angle of the measurement point, and the deflection may be represented by a deflection value of the measurement point.

It will be appreciated that in this embodiment, STEP4 combines a kalman filter algorithm with a cellular automaton, and provides an accurate method for monitoring and controlling local deformation behavior in continuous bridge construction by performing state estimation and prediction on observed data of measurement points.

By applying the Kalman filtering to continuous bridge construction, parameters such as displacement, inclination angle, deflection and the like of the bridge can be estimated in real time, accurate deformation monitoring information is provided, and the bridge deformation trend prediction and timely measure adjustment and repair are facilitated. Meanwhile, for the embodiment, the definition rules and the representation modes are general, and specific definition and selection are needed according to actual situations in specific application in continuous beam bridge construction.

Exemplary: the embodiment is provided with a continuous beam bridge in construction, the current bridge consists of three measuring points, and the state vector of each measuring point is as follows:

X_i＝[d_i,theta_i,phi_i]

where d_i represents displacement, theta_i represents inclination angle, phi_i represents deflection. The embodiment hopes to use a method of combining Kalman filtering and cellular automaton to estimate the state of local deformation of the bridge according to the observation value of the optical instrument of the measuring point:

defining a problem state vector:

X_i＝[d_i,theta_i,phi_i]

where i represents the index of the measurement point.

Defining an observation vector:

Z_i＝[M_i^k]

where k represents the number of optical instruments at the measurement point i, and M_i≡k represents the moment value measured by the optical instruments.

Defining a state transition matrix: f_i is an identity matrix because the state vector of the present embodiment does not change in unit time.

Defining an observation matrix: h_i is an identity matrix because the measurements of this embodiment directly correspond to the individual components of the state vector.

Defining a process noise covariance matrix: q_i is a diagonal matrix, each element representing the process noise variance of the corresponding state component. The process noise variance of the displacement and the inclination angle of the continuous beam bridge of the embodiment is equal, and the process noise variance of the deflection is slightly larger.

Where q_d, q_theta, q_phi represent the process noise variance of displacement, tilt angle, and deflection, respectively.

Defining an observed noise covariance matrix: r_i is a diagonal matrix, and each element represents the noise variance of the corresponding observed component. The measurement errors of the optical instrument in the construction of the continuous beam bridge of the embodiment are independent and have equal variances at each measuring point.

Where r_1, r_2,..r_k represent the variance of the optical instrument measurement error of station i, respectively.

Defining an initial state estimation vector and an initial state covariance matrix: the initial state estimation vector can be set as a zero vector, and the initial state covariance matrix can be set as a larger diagonal matrix:

X_i(0)＝[0,0,0]

The initial state estimation vector is set to a zero vector, which indicates that there is no prior information on the local deformation state of the bridge at the beginning.

Initial state covariance matrix p_i (0):

p1, carrying out Kalman filtering estimation and prediction:

based on the state transition matrix f_i and the state estimate x_i (t-1) at the previous time, the state estimate x_i (t) at the current time is predicted. Based on the state transition matrix f_i and the state covariance matrix p_i (t-1) of the previous time, the state covariance matrix p_i (t) of the current time is predicted.

P2, updating:

based on the observation matrix h_i, the observation noise covariance matrix r_i and the predicted state estimate x_i (t), a kalman gain k_i (t) is calculated.

Based on the kalman gain k_i (t), the observed value z_i and the predicted state estimate x_i (t), the state estimate x_i (t) at the current time is updated. The state covariance matrix P_i (t) at the current moment is updated according to the Kalman gain K_i (t), the observation matrix H_i and the predicted state covariance matrix P_i (t).

P3, obtaining an estimated value of the local deformation state of the bridge:

the estimated value X_i (t) of the local deformation state of the bridge is an updated state estimation vector, wherein each component corresponds to the displacement, the inclination angle and the deflection at the measuring point.

Example five

The present embodiment is linked to embodiment four:

STEP5, the support vector machine is matched with the cellular automaton;

5.1 defining a training dataset supporting a vector machine:

D＝{(x_i,y_i)}

wherein x_i is a feature vector and y_i is a label;

5.2 defining feature vectors:

x_i＝[p_i,s_i,m_i]

wherein p_i is the measuring point position, s_i is the measuring point shape, and m_i is the measuring point material characteristic;

5.3 training a support vector machine model to obtain a classifier:

C＝f(D)

5.4, predicting the state of the measuring point: y_i=c (x_i).

The purpose is as follows:

training the training data set by a support vector machine (Support Vector Machine, SVM) to obtain a classifier. And predicting the state of the measuring point by using a support vector machine classifier so as to realize monitoring and prediction of the local deformation state of the bridge.

In this embodiment, the SVM classifies the data by constructing a hyperplane in the feature space to find the optimal decision boundary. The training process of the SVM determines the parameters of the classifier by finding the support vector (the point of the sample point closest to the classification hyperplane), thereby realizing the classification of unknown data.

The operation steps of this embodiment are as follows:

p1, defining a training data set D of a support vector machine, wherein each sample consists of a feature vector x_i and a label y_i.

P2, feature vector x_i contains site location p_i, site shape s_i, and site material property m_i.

And P3, training the support vector machine by using the training data set D to obtain a classifier C.

P4, for a new measurement point, by inputting its feature vector x_i into classifier C, its state y_i can be predicted.

And P5, the predicted state value can be used for monitoring and predicting deformation states such as displacement, inclination angle, deflection and the like of the continuous beam bridge at different measuring points.

It should be noted that:

(1) Feature vector x_i:

station position p_i: the specific position of the measuring point in the continuous beam bridge structure is expressed by coordinates or relative positions.

Station shape s_i: curvature, slope, structural shape characteristics around the measurement point.

Station material properties m_i: the material characteristics of the part where the measuring point is located include the elastic modulus or density of the material.

(2) Tag y_i:

the tags represent the status of the measurement points, typically as a result of classification or regression for a particular problem.

Classification labels can be used to indicate the status of a measurement point as normal, with minor distortion or with severe distortion. Regression labels may also be used to represent specific deformation values of the measurement points, such as values of displacement, tilt angle, or deflection. For the present embodiment, the definition rules and the representation modes are general, and specific definition and selection are required for specific application in continuous beam bridge construction according to actual situations.

In this embodiment, it can be understood that the support vector machine is used as a powerful classifier, and can predict the state of a new measurement point through learning a training data set. The support vector machine can be used for monitoring and predicting the local deformation state of the bridge in real time by combining the framework of the cellular automaton. The support vector machine can classify the measuring points into different state categories according to the characteristic vectors of the different measuring points, so that the deformation condition of the bridge can be better understood.

Further, in the present embodiment and the fourth embodiment: in the framework of cellular automata, the kalman filter in STEP4 is used to estimate and predict the local deformation state of the bridge. The support vector machine in STEP5 is matched with cellular automata, and state prediction is carried out through a training data set and feature vectors, so that more comprehensive information is provided.

The prediction result of the support vector machine can be used as an observed value of the kalman filter in STEP4 to update the state estimation and covariance matrix. By comprehensively using STEP4 and STEP5, the deformation state of the continuous beam bridge can be more accurately monitored and predicted through iterative updating of cellular automata and state prediction of a support vector machine.

Furthermore, in continuous beam bridge construction, STEP5 can train a support vector machine model according to the position, shape, material characteristics and other characteristics of the measuring points to obtain a classifier. Then, by inputting the feature vector of the new measurement point, the support vector machine classifier can predict the state of the measurement point, such as displacement, inclination angle and deflection. Therefore, the local deformation state of the bridge can be monitored and predicted in real time in the bridge construction process, and important information is provided for engineers and monitoring staff so as to take necessary measures in time for adjustment and repair.

Exemplary:

the present embodiment is to use a support vector machine and cellular automata to predict states (normal, tiny deformation, serious deformation) at different measuring points of the continuous beam bridge. The position, shape and material property data of some measuring points are currently collected and marked accordingly to form a training data set.

Training dataset D:

x_1= [ p_1, s_1, m_1], y_1= normal

x_2= [ p_2, s_2, m_2], y_2= micro deformation

x_3= [ p_3, s_3, m_3], y_3= normal

...

x_n= [ p_n, s_n, m_n ], y_n= severe deformation

Step 5.1: defining a training data set of a support vector machine:

a set of training data sets D is collected, wherein each sample contains a feature vector x_i and a corresponding label y_i.

Step 5.2: defining a feature vector:

the feature vector x_i consists of the site location p_i, site shape s_i and site material property m_i.

Step 5.3: training a support vector machine model:

the training data set D is used to train the support vector machine model to obtain a classifier C.

Step 5.4: predicting the state of a measuring point:

for any measuring point, the characteristic vector x_i of the measuring point can be input into a support vector machine model, and the state label y_i of the measuring point can be obtained through prediction by a classifier C.

The states at different measuring points of the continuous beam bridge can be predicted through the steps.

In the cellular automaton framework combined with the step 4, the purpose of the step 5 is to predict the state of the measuring point by using a support vector machine model, and take a prediction result as the input of the cellular automaton model to further change the cellular state so as to reflect the local deformation condition of the bridge. The Kalman filtering in the step 4 and the cellular automaton can be matched to provide estimation and prediction of the bridge state, and the support vector machine in the step 5 and the cellular automaton can be matched to predict the state of the measuring point according to the feature vector. The two complement each other and interact with each other, so that the whole simulation system is more comprehensive and accurate.

The following is an exemplary cellular simulation illustrating the state prediction of multiple stations in a continuous bridge:

measuring point	Status of
		1	Normal state
2	Micro deformation
		3	Normal state
4	Severe deformation

In this example, the state information at different points may be obtained by predicting the points of the continuous bridge with a support vector machine model. These predictions can be further applied to the Kalman filtering in step 4 in combination with cellular automata for the estimation and prediction of bridge states. Summarizing, the purpose of the cooperation of the support vector machine and the cellular automaton in the step 5 is to predict the state of the measuring point through the support vector machine model, and use the prediction result for inputting the cellular automaton model to further simulate and predict the deformation state of the continuous beam bridge. The two steps complement each other and interact, so that the accuracy and the reliability of the simulation system can be improved.

Example six

The present embodiment is connected to the fifth embodiment:

STEP6, the genetic algorithm is matched with cellular automata:

6.1 definition of individual representation of genetic algorithm:

G_i＝[w_i,h_i,m_i]

wherein w_i is the width of the bridge, h_i is the height of the bridge, and m_i is the material property of the bridge;

6.2 defining fitness function:

F(G_i)

6.3 initializing population:

P(0)＝{G_i(0)}

wherein i is a cell number;

6.4 evolution process:

6.4.1 selection operation: selecting an excellent individual using the fitness function;

6.4.2 crossover operation: generating new individuals by crossover operations;

6.4.3 variation operations: performing mutation operation on an individual;

6.4.4 update population: obtaining a new population according to the selection, crossing and mutation operations;

6.4.5 judging convergence conditions: judging whether to terminate evolution according to the convergence condition;

6.4.6 output results: and outputting the optimal individual in the evolution process as an optimal bridge design.

In the embodiment, the optimization design of the continuous beam bridge is realized through the cooperation of a genetic algorithm and a cellular automaton. Through the evolution process of the genetic algorithm, searching for potential excellent bridge design individuals, and further obtaining an optimal bridge design scheme.

In this embodiment, the genetic algorithm is an optimization algorithm that simulates the natural evolution process. It iteratively searches the optimal solution in the solution space from generation to generation by simulating selection, crossover and mutation operations in biological genetics. Cellular automata is a discrete dynamics model based on local interaction rules, and can describe the dynamic behavior of a complex system by simulating state evolution among cells. The combination of the two is operated according to the following steps:

Step 6.1: defining an individual representation of the genetic algorithm:

individual is denoted as width w_i, height h_i and material property m_i of the bridge. These parameters determine the structure and performance of the bridge. The material properties m_i include:

(1) Strength characteristics: including flexural strength, compressive strength, tensile strength, etc. of the material. These characteristics describe the load carrying capacity of the material under stress.

(2) Stiffness characteristics: including the elastic modulus, shear modulus, etc. of the material. These characteristics describe the degree of deformation of a material under stress.

(3) Stability characteristics: including stability of the material, yield strength, critical stress, etc. These characteristics describe the change and stability of the material over long periods of use.

(4) Durability characteristics: including corrosion resistance, fatigue resistance, etc. of the material. These characteristics describe the durability of the material under ambient conditions.

The specific definition rules can be selected and determined according to the design requirements of engineers and the actual performance of materials, or simulation determination can be directly performed by using finite elements. Appropriate materials are selected according to the design requirements, the structural type, the use environment and other factors of the beam bridge, and specific values of the material characteristics m_i are defined according to experimental data of the materials or technical specifications provided by manufacturers.

Step 6.2: defining a fitness function:

the fitness function F (g_i) evaluates the goodness of each individual. The method can be defined according to design requirements and targets, for example, indexes such as structural strength, economy and the like of the bridge are considered.

Exemplary: f (G_i) considers the structural strength, then its definition rules are as follows:

p1, defining stress conditions of a structure: given the moment forces of a continuous bridge, each individual g_i may be denoted as [ w_i, h_i, m_i ], representing the width, height and material properties of the bridge, respectively.

P2, calculating bending moment of the structure: according to the geometric shape and loading condition of the continuous beam bridge, the bending moment distribution of each measuring point can be calculated.

P3, calculating the safety coefficient or damage index of the structure: based on the calculated bending moment distribution and the strength characteristics of the material, a proper method can be adopted to calculate the safety coefficient or damage index of the structure. For example, the ratio between bending moment and section bending strength may be used to represent a safety factor, or a failure criterion (e.g., ultimate bending moment) may be used to determine a failure indicator.

P4, defining a fitness function: the fitness function F (g_i) may be defined according to a safety factor or a damage indicator of the structure. In general, the fitness function may be chosen as the inverse of the safety factor, namely:

F (g_i) =1/safety factor

Such that a larger fitness function value indicates a safer structure. Of course, the specific definition mode of the fitness function can be adjusted according to actual requirements. For example, other structural performance metrics (e.g., stiffness, vibration characteristics, etc.) and multi-objective optimization problems, etc. may be considered. Designing a proper fitness function requires combining specific engineering requirements and design goals, and reasonably balancing and adjusting the fitness function according to practical situations.

Step 6.3: initializing a population:

the initialization population P (0) contains initial individuals G_i (0). Individuals in the population represent different bridge designs.

Step 6.4: evolution process:

p1, selection operation: and selecting an individual with higher fitness as a parent individual for crossover and mutation operation by using the fitness function.

P2, cross operation: new individuals are generated by crossover operations. The genetic information of the parent individuals may be combined into new individuals using different crossover means, such as single point crossover, multi-point crossover, etc., and by way of example, single point crossover is used, which includes the following:

p2.1, randomly selecting two parent individuals for crossing operation.

P2.2, one crossover point (i.e.one locus) is selected in the chromosome of the parent individual.

P2.3 replication of the gene sequence before the crossover point from one parent individual to a child individual.

P2.4, copying the gene sequence after the crossover point from another parent individual to a offspring individual.

P2.5, generating two child individuals, wherein the gene sequence of one child individual is the combination of the partial gene of the first parent individual and the partial gene of the second parent individual, and the gene sequence of the other child individual is the combination of the partial gene of the second parent individual and the partial gene of the first parent individual.

P3, mutation operation: the mutation operation of individuals introduces certain randomness. Variation can be achieved by modifying certain gene values of an individual, thereby creating a new individual.

P4, updating population: and obtaining a new population according to the selection, crossing and mutation operation, and taking the new population as a father of the next generation.

P5, judging convergence conditions: judging whether to terminate the evolution according to the convergence condition. The convergence condition may be reaching a predetermined number of iterations, the fitness reaching a certain threshold, etc. Illustratively, employing the number of iterations includes:

p5.1, rule of thumb: a reasonable number of iterations may be empirically set based on the size and complexity of the problem. This setting may be based on previous experience or practice of similar problems.

P5.2, convergence analysis: and observing the convergence property of the genetic algorithm, and setting the iteration times according to the convergence speed of the algorithm and the requirement of the stopping condition. Whether the algorithm has tended to stabilize can be determined according to the change of population fitness.

P5.3, self-adaptive adjustment: by dynamically adjusting the number of iterations, the algorithm is enabled to adaptively decide when to terminate based on the current evolving state. For example, a convergence criterion may be set, and the algorithm is terminated when the criterion satisfies a certain condition.

P6, outputting a result: and outputting the optimal individual in the evolution process as an optimal bridge design.

In this embodiment, the genetic algorithm of step 6 is matched with a cellular automaton, so that an optimal design scheme of the continuous bridge can be searched. The genetic algorithm searches the design space through the evolution process, new individuals are generated through selection, crossing and mutation operation, and the design of the girder bridge is gradually optimized. The cellular automaton is used as a model for simulating bridge behaviors, and can perform state evolution simulation through an optimal individual generated by a genetic algorithm to evaluate the performance and response characteristics of the cellular automaton.

Furthermore, the kalman filtering and the cellular automaton in the fourth embodiment are used in combination to estimate and predict the local deformation state of the bridge, and provide state information for optimizing design. The support vector machine of the fifth embodiment is used for predicting the state of the measuring point in cooperation with the cellular automaton, and provides state information for optimizing design. The genetic algorithm of the embodiment is matched with the cellular automaton to further search the optimal design scheme of the bridge through optimizing and designing the genetic operation and state evolution simulation of the individual. The three steps complement each other under the framework of the cellular automaton, interact, and jointly realize the targets of optimal design and state prediction in continuous beam bridge construction through feedback of state information and iteration of optimal design.

Exemplary:

p1, define questions and goals: a continuous beam bridge is designed to have minimal deformation and stress under given external load conditions.

P2, preparing an initial population: a set of parameters of the bridge is initialized, including width, height and material properties. These parameters constitute the individual representation of the genetic algorithm.

P3, calculating a fitness function: and according to the parameters of the current bridge, using cellular automata to simulate and calculate the deformation and stress of the bridge. The deformation and stress are used as an evaluation index of the fitness function, i.e., fitness function F (g_i).

P4, selecting: and selecting an excellent individual as a parent individual according to the evaluation result of the fitness function.

P5, cross operation: and performing single-point crossover operation on the selected parent individuals to generate new individuals.

P6, mutation operation: and carrying out mutation operation on the newly generated individuals, and introducing certain randomness.

P7, updating population: a new population is obtained based on the selection, crossover and mutation operations.

P8, judging convergence conditions: setting iteration times or judging convergence condition of the fitness function, and terminating evolution when convergence condition is met.

P9, outputting a result: and outputting the optimal individual in the evolution process as an optimal bridge design.

In the scene, the genetic algorithm in the step 6 complements with the cellular automaton and interacts with the cellular automaton to achieve the beneficial effect. The cellular automaton simulates and calculates the deformation and stress of the bridge, and provides an evaluation index of the fitness function; and the genetic algorithm evolves the individual through selection, crossover and mutation operations, searching the bridge parameter space to find the optimal solution. By iterative means, genetic algorithms continue to optimize designs, while cellular automata offer the ability to rapidly evaluate designs, enabling excellent individuals to be preserved and developed during evolution.

Example seven

In this embodiment, embodiments one to six are linked:

the present embodiment provides a hardware carrier for implementing embodiments one to six, including two schemes:

scheme one: a computer device comprising a processor, a memory coupled to the processor, the memory having stored therein program instructions which, when executed by the processor, cause the processor to perform a method of determining as described above.

Scheme II: a storage medium storing program instructions capable of implementing the above-described measurement method.

The control program shown in fig. 2 to 3 can be executed by the above-described scheme and completes the method STEPs of STEP1 to 6 and logic thereof, and the principle thereof is as follows:

(1) This function is used to initialize the population, i.e., to generate an initial set of individuals. Specific implementations refer to randomly generated individual characteristic values, material properties, etc., and storing them in a suitable data structure.

(2) This function is used to calculate the fitness of the individual, i.e., measure the individual's goodness in terms of solving the problem. The specific implementation will calculate the fitness value of each individual according to the requirements of the problem. In continuous beam bridge construction, the fitness function can consider factors such as structural strength, vibration characteristics and the like.

(3) The stateestimate (). This function is used to perform state estimation, based on the measured data and the model, to estimate the local deformation state of the bridge. The specific implementation refers to updating the state estimation value by using methods such as Kalman filtering and the like and combining measured data and a state transition model.

(4) the function is used for training a support vector machine model to realize prediction of the measuring point state. Specific implementations will use support vector machine algorithms to build classifier models based on existing training data sets.

(5) The predictState () function is used for predicting the state of the measuring point, and classifying and predicting new measuring point data according to the trained support vector machine model. The specific implementation uses a trained classifier model to predict the feature vector of the measuring point to obtain a corresponding state value.

(6) evolvePoplation (). This function is used to perform evolutionary operations of genetic algorithms, i.e., by selecting, crossing, and mutating, etc., to generate new individuals and to update populations. The specific implementation will operate on the current population according to the principles of genetic algorithm, generating a new set of individuals.

(7) This function is used to determine if the evolution process reaches a convergence condition, i.e., if the evolution is terminated. Specific implementation will set a convergence judgment standard according to the requirement of the problem, for example, the iteration number reaches the upper limit or the change of the fitness value is no longer significant, etc.

(8) This function is used to output the results of the current iteration, including the current iteration number, population information, fitness value, etc. The specific implementation will output the relevant information to a screen, file or other media as desired.

(9) This function is used to output the result of the optimal individual, i.e. the individual with the best fitness value during evolution. The specific implementation outputs relevant information such as characteristic values, material characteristics and the like of the optimal individual according to the requirements of the problem.

It can be understood that the control program shown in this embodiment is a c++ pseudo code, and the pseudo code only shows the control logic of STEP1 to STEP 6 described above; any program in other assembly language can be superseded on the logical premise presented by the pseudocode.

Application example 1

In order to make the above-described embodiments of the present invention more comprehensible, the present invention will be described in detail using examples of application. The present invention may be embodied in many other forms than described herein, and similar modifications may be made by those skilled in the art without departing from the spirit of the invention, so that the invention is not limited to the examples of application disclosed below.

The present application example includes specific embodiments and examples one to seven:

the application example is designed into a continuous beam bridge, the span of the bridge is 30 meters, and a total of 5 measuring points are used for monitoring the deformation state of the bridge. The application example aims to optimize the width, the height and the material characteristics of the bridge so as to improve the structural strength and the vibration characteristics of the bridge.

Parameter values:

bridge width range: 5m-10m

Beam bridge height range: 10m-20m

Material characteristics range: 1000-5000

STEP1, initializing the state vector of cellular automaton the width of the bridge was set to 7m, the height was set to 12m, and the material properties were set to 3000 in the initial application example.

STEP2 State transition matrix and Process noise covariance matrix in this example, the present application sets the State transition matrix and Process noise covariance matrix as follows:

State transition matrix a:

process noise covariance matrix Q:

STEP3 update of State estimation vector X_i (t) at each time STEP, the State estimation vector X_i (t) is updated based on the measurement data of the measurement points and the state transition model. The initial state estimation vector is set as [000], and the measurement data of the measuring point is set as [123].

STEP4, training a support vector machine model:

a training data set D is used, which includes feature vectors x_i and corresponding labels y_i. The method of the fifth embodiment is set to be adopted in the application example, and the following training data are provided:

D＝{([5101000],0)

([7123000],1)

([8152000],0)

([9184000],1)

([6111500],0)}

training the training data by using a support vector machine algorithm, and obtaining a classifier C.

STEP5, predicting the state of the measuring point:

and predicting the feature vector of the measuring point according to the trained support vector machine model to obtain the state of the measuring point. The feature vector to be predicted in this application example is set as [7123000], and the prediction result is:

y_i＝C([7123000])

STEP6, genetic algorithm optimization design:

and (5) carrying out optimal design by using a genetic algorithm.

Defining an individual is expressed as:

G_i＝[w_i,h_i,m_i]

wherein the width w_i, the height h_i and the material property m_i are the genes of the individual, respectively. A fitness function F (g_i) is defined to evaluate the merits of the individuals. According to the specific requirements of the problem, the fitness function can be designed, for example, the fitness value is calculated by taking the structural strength of the bridge into consideration and using a structural analysis method. Genetic algorithms include selection operations, crossover operations, mutation operations, and population updates until convergence conditions are met. And finally outputting the optimal individual in the convergence process as an optimal bridge design.

The above examples of application only represent embodiments of the invention which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.

Application example two

The application example is based on the simulation scene of the application example one, wherein the application example is further: the application example introduces two malignant conditions, and the first application example is further deduced and the effectiveness of the application example is shown:

(1) Abnormal material properties: the method is set in the original material characteristic range, wherein the material characteristic of one measuring point is abnormal and is far lower than the normal range. Other parameters remain unchanged. This will result in that the support vector machine model trained in STEP4 may not accurately predict the state of outlier points. The application example will see whether the system can detect abnormal measuring points and optimize the design of the bridge as much as possible to enhance the structural strength through the synergistic effect of STEPs 1-6 in the iterative process.

(2) Extreme deformation state: an extreme deformation state is set in the continuous beam bridge construction process, and the extreme deformation state exceeds a normal range. This will result in an increase in the difference of the measurement data from the model at the time of the state estimation vector update in STEP3, possibly resulting in a deviation of the state estimation. The application example observes whether the system can timely detect the deformation state and take corresponding optimization measures through the synergistic effect of STEPs 1-6 in the iterative process.

Specific:

p1, abnormal material properties: the design is such that at the 3 rd measuring point an abnormal material property is introduced, which material property is far below the normal range. The material properties of this station were set to [100,200,10], and the normal range was [1000,2000,100]. Other parameters remain unchanged.

P2, during the iteration of STEP 1-6, the system will detect and optimize this anomaly by co-operation.

P3, extreme deformation state: the extreme deformation state is set at the 5 th measuring point, and the displacement exceeds the normal range. The displacement of the measuring point is set to 10, and the normal range is [ -5,5]. Other parameters remain unchanged.

P3, in the iterative process of STEP 1-6, the system detects and takes corresponding optimization measures through the synergistic effect. The following is a complete derivation and demonstration of the above-described simulation scenario:

P3.1, initializing: setting parameters and measuring point arrangement of a continuous beam bridge, and defining an initial state estimation vector and a covariance matrix.

P3.2, iterative procedure: the following steps are repeated until convergence or the maximum number of iterations is reached:

a. collecting measurement data: and measuring each measuring point by using an optical instrument to obtain displacement data.

b. State estimation and prediction: and according to the measurement data, performing state estimation and prediction by using a Kalman filtering algorithm to obtain an estimated value of the local deformation state of the bridge.

c. Training a support vector machine model: and constructing a training data set by using the historical data and the current state estimation value, and training a support vector machine model.

d. Predicting the state of a measuring point: and predicting the state of each measuring point by using a support vector machine model.

e. Anomaly detection and optimization: detecting the measuring points of the abnormal material characteristics, and judging whether the measuring points exceed the normal range. If the design parameters are out of range, optimizing the design parameters of the bridge through a genetic algorithm.

f. Judging convergence conditions: and judging whether to terminate the iteration according to the convergence condition.

Outputting a result: and outputting an optimal individual in the iterative process as an optimal bridge design, and displaying the final state and the optimization result of the continuous bridge.

By introducing abnormal material characteristics and extreme deformation states, the application example can observe the effectiveness of STEP 1-6 and show the synergistic effect of cellular automaton, kalman filtering, support vector machine and genetic algorithm.

Claims

1. The auxiliary measuring method for the construction process of the continuous beam bridge is characterized by comprising the following steps of:

STEP1, arranging a plurality of measuring points on a bridge, and simultaneously measuring by using a plurality of optical instruments to obtain deformation data of different positions of the bridge;

STEP2, creating a cellular automaton model, wherein each cell represents one position of the bridge;

STEP3, describing a cell state updating rule through a linear function according to the behavior and deformation condition of the bridge and combining the factors of the cell position, the neighbor cell state, the material characteristics and the external load;

STEP4: for each cell: defining the local deformation state of the bridge as the problem of Kalman filtering, filtering the measurement data and obtaining the bridge deformation state measurement; kalman filtering is matched with cellular automata:

4.1 defining a problem state vector:

X_i＝[d_i,theta_i,phi_i]

where d_i is displacement, theta_i is inclination angle, phi_i is deflection;

4.2 defining an observation vector:

Z_i＝[M_i^k]

wherein k is the number of optical instruments of all measuring points i; m_i≡k is a moment value measured by an optical instrument;

4.3 define a state transition matrix: f_i;

defining an observation matrix: h_i;

4.4 defining a process noise covariance matrix: q_i;

defining an observed noise covariance matrix: r_i;

4.5 define an initial state estimation vector: x_i (0);

defining an initial state covariance matrix: p_i (0);

4.6, kalman filtering estimation and prediction:

4.6.1 prediction:

X_i(t)＝F_i*X_i(t-1)P_i(t)＝F_i*P_i(t-1)*F_i^T+Q_i

4.6.2 updating:

K_i(t)＝P_i(t)*H_i^T*(H_i*P_i(t)*H_i^T+R_i)^-1

X_i(t)＝X_i(t)+K_i(t)*(Z_i-H_i*X_i(t))

P_i(t)＝(I-K_i(t)*H_i)*P_i(t)

k_i (t) is the Kalman gain; i is an identity matrix;

4.7, obtaining an estimated value of the local deformation state of the bridge: x_i (t);

STEP5: for each cell: using the filtered data and combining the position, shape and material characteristic information of the measuring points by using a support vector machine as a feature vector; identifying and detecting abnormal conditions in bridge deformation;

STEP6: for each cell: using the filtered data, combining with the problem of defining the bridge linear control strategy, searching an optimal solution by using a genetic algorithm, and searching the optimal linear control strategy by using the genetic algorithm through cyclic iteration;

STEP6, the genetic algorithm is matched with cellular automata:

6.1 definition of individual representation of genetic algorithm:

G_i＝[w_i,h_i,m_i]

6.2 defining fitness function:

F(G_i)

6.3 initializing population:

P(0)＝{G_i(0)}

wherein i is a cell number;

6.4 evolution process:

6.4.2 crossover operation: generating new individuals by crossover operations;

6.4.3 variation operations: performing mutation operation on an individual;

2. The method according to claim 1, wherein:

3. The method according to claim 1, wherein:

STEP2, cellular automaton initialization:

2.2 setting an initial state value and an attribute value:

for each station i:

initialization state: s_i=0;

initializing the state of the optical instrument: s_i≡k=0 for all k;

cell state update:

for each station i: calculating states and attributes of neighbor measuring points: n_i= { s_j, p_j }

Neighbor measurement points where j is i; p_j is the attribute of neighbor measurement point j;

calculating an external load: l_i=f (p_i);

p_i is the attribute of station i.

4. A method according to claim 3, wherein:

STEP3, cell state update rule:

S_i(t)＝α*S_i(t-1)+β*∑(S_j(t-1)*w_ij)+γ*M_i+δ*F_i

wherein S_i (t) is the state value of cell i at time t;

s_j (t-1) is the state value of the neighbor cell j at the time t-1;

w_ij is a neighbor relation weight;

M_i is the material property of cell i;

f_i is an external load;

alpha, beta, gamma and delta are weight coefficients;

setting a threshold value of a cell convergence condition: epsilon.

5. The method according to claim 4, wherein:

STEP5, the support vector machine is matched with the cellular automaton;

5.1 defining a training dataset supporting a vector machine:

D＝{(x_i,y_i)}

wherein x_i is a feature vector and y_i is a label;

5.2 defining feature vectors:

x_i＝[p_i,s_i,m_i]

5.3 training a support vector machine model to obtain a classifier:

C＝f(D)

5.4, predicting the state of the measuring point: y_i=c (x_i).

6. A computer device comprising a processor, a memory coupled to the processor, the memory having stored therein program instructions that, when executed by the processor, cause the processor to perform the assay method of any of claims 1-5.

7. A storage medium storing program instructions for implementing the measurement method according to any one of claims 1 to 5.