CN-122020822-A - Building deformation risk prediction method and system

Abstract

The invention belongs to the technical field of buildings, and discloses a method and a system for predicting deformation risk of a building. According to the invention, the effect of reducing the deformation risk of the building by different maintenance schemes can be accurately quantified by establishing the lifting coefficient prediction model, the prediction lifting coefficient is taken as a core threshold, the scheme comparison is completed by combining with the practical constraint of engineering implementation, the destabilization risk of the long-term service of the building can be scientifically predicted by constructing the destabilization probability prediction model, the limitation that the prior art only focuses on the short-term risk is broken, the monitoring period and the management and control strategy are dynamically adjusted based on the prediction destabilization probability, the grading and the fine management and control of the risk are realized, the grading system and the prediction model are continuously optimized based on the practical treatment effect by setting the closed-loop optimization mechanism, and the deformation risk management and control requirements of different building types and different service stages can be adapted.

Inventors

• LIU PENG
• XU LIJUN
• Wu xinshao
• LIU XIAOYONG

Assignees

• 湖南五环创新建筑科技有限公司

Dates

Publication Date

20260512

Application Date

20260413

Claims (10)

1. 1. A system for predicting risk of deformation of a building, comprising the following modules: The data acquisition module is used for acquiring deformation risk data of the building; The deformation risk assessment module is used for generating deformation risk scores of the building according to the deformation risk data, and calculating the deformation risk score change rate of the building according to the deformation risk scores; the deformation risk grading module is used for grading the deformation risk of the building according to the deformation risk score and the deformation risk score change rate, and taking corresponding measures for the buildings with different grades; The maintenance scheme decision module is used for acquiring historical lifting coefficient data, constructing a lifting coefficient prediction model, predicting the lifting coefficient of the deformation risk building by using the lifting coefficient prediction model, deciding and selecting a specific maintenance scheme according to the predicted lifting coefficient, and calculating according to the predicted lifting coefficient to obtain the deformation risk score of the maintained high deformation risk building; The destabilization probability prediction module is used for acquiring historical destabilization probability data and constructing a destabilization probability prediction model, predicting the destabilization probability of the building by using the destabilization probability prediction model, and taking countermeasures for the building according to the predicted destabilization probability.
2. 2. The building deformation risk prediction system according to claim 1, wherein a plurality of buildings form a building group, each building is provided with a monitoring unit, the monitoring units are used for collecting deformation risk data, the deformation risk data comprises vertical settlement amount, horizontal displacement amount, overall inclination rate, crack expansion rate and steel bar corrosion rate, and the deformation risk data is normalized.
3. 3. The building deformation risk prediction system according to claim 2, wherein the process of generating a deformation risk score for a building from deformation risk data comprises: deformation risk score S for building: In the formula, 、 、 、 And Training according to historical data as a weight coefficient; the process for calculating the deformation risk score change rate of the building according to the deformation risk score comprises the following steps: Setting a monitoring period, in a fixed monitoring period, firstly acquiring the latest deformation risk score calculated in the monitoring period for the same building, then acquiring the deformation risk score of the last monitoring period of the building, subtracting the deformation risk score of the last monitoring period from the latest deformation risk score to obtain the absolute change of scores in two monitoring periods, dividing the absolute change by the deformation risk score of the last monitoring period, and simultaneously combining the time intervals of the two monitoring periods, converting the absolute change into the score change proportion in unit time to finally obtain the deformation risk score change rate of the building.
4. 4. A building deformation risk prediction system according to claim 3, wherein the process of ranking the deformation risk of the building according to the deformation risk score and the deformation risk score change rate comprises: setting a first scoring threshold, a second scoring threshold and a third scoring threshold which are suitable according to historical scoring data, wherein the historical scoring data refers to a data set of deformation risk scores of a previous building; setting a first change rate threshold, a second change rate threshold, a third change rate threshold and a fourth change rate threshold which are suitable according to historical change rate data, wherein the historical change rate data is a data set of deformation risk scoring change rates of a previous building; combining the current deformation risk score and the deformation risk score change rate of the building, adopting a two-dimensional coupling mode to carry out risk classification, classifying the buildings in the building group into four classes of low deformation risk, medium deformation risk, higher deformation risk and high deformation risk, and matching corresponding disposal measures; if the current deformation risk score is smaller than or equal to a first score threshold value and the first change rate threshold value is smaller than or equal to the current deformation risk score change rate is smaller than or equal to a second change rate threshold value, judging that the deformation risk building is low, and only maintaining a conventional monitoring period and daily inspection without taking extra treatment measures; if the first scoring threshold value is smaller than the current deformation risk scoring and smaller than or equal to the second scoring threshold value, or the second change rate threshold value is smaller than or equal to the third change rate threshold value, judging that the deformation risk building is middle deformation risk building, performing special investigation aiming at indexes with abnormal scores, determining causes of deformation risks, and taking targeted prevention and control measures; If the second grading threshold value is smaller than the current deformation risk grading value and smaller than the third grading threshold value or the third change rate threshold value is smaller than the current deformation risk grading change rate and smaller than the fourth change rate threshold value, judging that the deformation risk building is higher, compiling a special emergency treatment plan, taking temporary reinforcing measures on abnormal deformation parts, synchronously checking structural damage conditions, and formulating a preliminary maintenance reinforcing plan; If the current deformation risk score is greater than or equal to the third score threshold value or the current deformation risk score change rate is greater than the fourth change rate, judging that the structure is a high deformation risk building, immediately starting emergency response, stopping unnecessary operation in the building, limiting entry of irrelevant personnel, arranging 24 hours for uninterrupted real-time monitoring, synchronizing organization specialists to develop risk assessment, immediately making and implementing an emergency maintenance scheme, and preventing and controlling the structure instability risk.
5. 5. The system of claim 4, wherein the process of obtaining historical lifting coefficient data and constructing a lifting coefficient prediction model, and predicting lifting coefficients of a deformation risk building using the lifting coefficient prediction model comprises: The emergency maintenance scheme is required to be immediately formulated for the high-deformation-risk building, and the lifting coefficient is the reduction ratio of the deformation risk score after the high-deformation-risk building is maintained after the selected maintenance scheme is implemented on the high-deformation-risk building by the pointer; Factors influencing the lifting coefficient comprise current deformation risk score, current deformation risk score change rate, building age of a building, historical engineering case data of similar maintenance schemes and implementation range and implementation precision of the maintenance schemes; Acquiring historical lifting coefficient data of a single high-deformation risk building in different monitoring periods, wherein the historical lifting coefficient data comprises current deformation risk scores of the single high-deformation risk building in different monitoring periods, current deformation risk score change rates, building years of the building, historical engineering case data of similar maintenance schemes, implementation range and implementation precision of the maintenance schemes and historical lifting coefficients of the single high-deformation risk building in the monitoring periods; Generating a lifting coefficient prediction set according to the current deformation risk score of a corresponding high-deformation risk building in different historical lifting coefficient data, the current deformation risk score change rate, the building age of the building, the historical engineering case data of the similar maintenance scheme, the implementation range and implementation precision of the maintenance scheme and the corresponding historical lifting coefficient, and dividing the prediction set into a first training set and a first test set; constructing a first one-dimensional convolutional neural network, taking current deformation risk scores, current deformation risk score change rates, building years, historical engineering case data of similar maintenance schemes and implementation ranges and implementation precision of the maintenance schemes in different historical lifting coefficient data in a first training set as input data of the first one-dimensional convolutional neural network, and taking corresponding historical lifting coefficients in the first training set as output data of the first one-dimensional convolutional neural network; training the first one-dimensional convolutional neural network to obtain a first initial convolutional neural network, performing model verification on the first initial convolutional neural network by using a first test set, and outputting the first initial convolutional neural network which is smaller than or equal to a preset first test error threshold value as a lifting coefficient prediction model; And inputting the current deformation risk score, the current deformation risk score change rate, the building year of the building, the historical engineering case data of the similar maintenance scheme and the implementation range and implementation precision of the maintenance scheme of each high deformation risk building in the monitoring period to a lifting coefficient prediction model every time a monitoring period is reached, so as to obtain the prediction lifting coefficient of each high deformation risk building.
6. 6. The building deformation risk prediction system according to claim 5, wherein the process of selecting a particular maintenance scheme based on the predicted lift coefficient decision comprises: Aiming at a high-deformation risk building, firstly, core causes, key risk points, structural damage degree and risk development trend of the building deformation risk are accurately locked through site special detection, full-period deformation monitoring data review, building structure and geological data review, the structural safety bottom line requirements which must be met by maintenance disposal are definitely met, and then, aiming at the defined core risk problem, at least 3 sets of alternative maintenance reinforcement schemes with complete feasibility are compiled by structural engineering and geotechnical engineering professional staff with corresponding qualification, wherein each set of scheme definitely corresponds to core contents such as disposal targets, core construction technology, risk disposal coverage, key technical parameters, management and control measures for the deformation causes, implementation period, field operation conditions and the like; Then, aiming at each set of alternative schemes, calculating to obtain a predicted lifting coefficient corresponding to the scheme through a pre-established lifting coefficient prediction model, setting the lowest lifting coefficient threshold of a maintenance scheme according to the risk level of a building in advance, wherein the coefficient threshold of a high-deformation risk building is not lower than 0.6, firstly completing rigid threshold screening of the scheme, directly removing the alternative schemes which have the predicted lifting coefficient lower than the coefficient threshold and cannot meet the lowest risk management and control requirement, and then, for the rest schemes which pass the threshold screening, taking the predicted lifting coefficient as a core priority judgment basis, preferentially selecting the scheme with higher predicted lifting coefficient and larger reduction amplitude of the deformation risk; Aiming at the conditions that the difference value of the predicted lifting coefficients of two sets and more schemes is not more than 0.1 and the risk management and control effects are similar, the implementation period of the scheme, the field operation difficulty, the interference degree of the existing normal use functions of the building, the one-time engineering investment, the long-term stability of the treatment effect, the safety management and control difficulty of the construction process and other engineering actual factors are selected in an auxiliary mode, the predicted lifting coefficients are finally selected to reach standards, the risk management and control effect is optimal, and meanwhile, the maintenance scheme which is adaptive to the field implementation conditions and the use requirements of the building is used as a formal implementation scheme, meanwhile, the side station supervision requirements, the key process quality acceptance standard and the dynamic deformation monitoring requirements of the construction period of the whole scheme implementation process are clarified, and the effect rechecking process after the implementation is finished, so that the risk reduction effect realized after the scheme is actually landed is ensured to be matched with the predicted lifting coefficients.
7. 7. The system of claim 6, wherein the step of calculating a deformation risk score for the maintained high deformation risk building based on the predicted lift coefficient comprises: After obtaining a predicted lifting coefficient of a to-be-maintained scheme, combining a current deformation risk score of a building and the management and control capability of the maintenance scheme on a deformation core cause, finely calculating the predicted deformation risk score after maintenance in steps, firstly obtaining the predicted lifting coefficient of the scheme through a lifting coefficient prediction model, subtracting the predicted lifting coefficient by 1 to obtain a proportion coefficient of residual risk after maintenance, and multiplying the residual risk proportion coefficient by the current deformation risk score of the building to obtain a basic predicted deformation risk score after maintenance; Then carrying out fine tuning on a basic prediction deformation risk score according to the management and control degree of the core inducement causing the building deformation by a maintenance scheme, if the maintenance scheme comprises measures for completely eliminating the deformation core inducement from the root, multiplying a long-term effect optimization coefficient of 0.85 on the basis of the basic prediction risk score, wherein the specific coefficient is determined according to the comprehensiveness of the inducement management and control and the long-term effect of the measures; if the maintenance scheme only carries out surface repair on structural damage caused by deformation, no management and control measures are taken on the core causes, and a core risk source continuously exists, multiplying a risk retention coefficient by 1.05 on the basis of a basic prediction risk score so as to embody residual risk and later rebound possibility caused by continuous action of the causes; if the maintenance scheme can only partially control the core causes and cannot completely eliminate the risk sources, no additional adjustment coefficient is set, and finally the comprehensive prediction deformation risk score of the building after the maintenance is finished is obtained, and meanwhile, the confidence interval of the prediction score is given by combining the construction process reliability of the maintenance scheme and the historical effect data of the similar engineering, so that a complete basis is provided for the follow-up operation and maintenance control of the building.
8. 8. The system for predicting risk of deformation of a building according to claim 7, wherein the process of acquiring historical destabilization probability data and constructing a destabilization probability prediction model, and predicting the destabilization probability of the building using the destabilization probability prediction model comprises: Five monitoring periods are set as an evaluation period, and each time an evaluation period is reached, the instability probability of the building in the building group is predicted, wherein the instability probability refers to the probability of unstable structure of the building after twenty years in the future; factors influencing the instability probability comprise a lifting coefficient change rate, historical maintenance times, historical maintenance frequency, current deformation risk score change rate and the design service life of a building; acquiring historical instability probability data of a single building in different evaluation periods, wherein the historical instability probability data comprises a lifting coefficient change rate, historical maintenance times, historical maintenance frequency, current deformation risk scores, current deformation risk score change rates, a building design service life and historical instability probability of the single building in the evaluation periods; Generating a destabilizing probability prediction set according to the lifting coefficient change rate, the historical maintenance times, the historical maintenance frequency, the current deformation risk score change rate, the building design service life and the corresponding historical destabilizing probability of the corresponding building in different historical destabilizing probability data, and dividing the destabilizing probability prediction set into a second training set and a second test set; Constructing a second one-dimensional convolutional neural network, taking the lifting coefficient change rate, the historical maintenance times, the historical maintenance frequency, the current deformation risk score change rate and the building design service life in different historical instability probability data in a second training set as input data of the second one-dimensional convolutional neural network, and taking the corresponding historical instability probability in the second training set as output data of the second one-dimensional convolutional neural network; Training the second one-dimensional convolutional neural network to obtain a second initial convolutional neural network, performing model verification on the second initial convolutional neural network by using a second test set, and outputting the second initial convolutional neural network which is smaller than or equal to a preset second test error threshold value as a destabilization probability prediction model; And inputting the change rate of the lifting coefficient, the historical maintenance times, the historical maintenance frequency, the current deformation risk score change rate and the service life of the building design of each building in the evaluation period into a instability probability prediction model every time an evaluation period is reached, so as to obtain the predicted instability probability of each building.
9. 9. The system for predicting risk of deformation of a building of claim 8, wherein taking countermeasures to the building based on the predicted probability of instability comprises: After the predicted instability probability of a building after twenty years of a preset year is obtained, the structural safety level, the use function and the personnel density of the building are combined, a four-level control threshold value of the instability probability is preset, wherein the predicted instability probability is lower than 5% and is low, 5% -20% is middle instability probability, 20% -50% is high instability probability, 50% and above are extremely high instability probability, and then according to different instability probability levels, step control measures and evaluation period dynamic adjustment are synchronously implemented, and the specific process is as follows: For a building with low instability probability, maintaining the conventional operation and maintenance control requirement, and keeping the original 30-day conventional period in the evaluation period, wherein long-term risk change conditions only need to be checked in annual structure inspection; For a building with medium instability probability, firstly, calculating an adjustment evaluation period according to the prediction instability probability, wherein an adjustment formula is that the adjustment evaluation period=the conventional evaluation period (1-prediction instability probability) is adopted, meanwhile, a future operation and maintenance management and control plan is optimized, the annual special detection frequency of a structure is increased by 1 time, a preventive maintenance scheme is formulated in advance, advanced prevention and control measures are adopted for solving the problems of material degradation, deformation induction and the like, and the change situation of the instability probability is rechecked every half year; For a building with high instability probability, the assessment period is greatly compressed, meanwhile, structural engineering experts immediately develop building structure safety special identification, hidden structural hazards and deformation core causes are comprehensively checked, a systematic reinforcement reconstruction scheme is formulated and implemented, after implementation of the scheme, the model is input again to calculate the prediction instability probability after maintenance, and the prediction instability probability after twenty years of reinforcement is ensured to be reduced to be within 5%; For a building with extremely high instability probability, the building is immediately started for 24 hours without interruption and real-time monitoring, building use management and control measures are synchronously adopted, the use functions and personnel loads of the building are limited, emergency reinforcement treatment is implemented by organizations, measures for thoroughly eliminating risks such as dismantling and reconstruction are adopted when necessary, safety accidents caused by structural instability are avoided, meanwhile, when the maintenance treatment or the monitoring data are obviously mutated once, the model is input again to update the prediction instability probability, and the evaluation period and the management and control measures are synchronously adjusted, so that dynamic closed loop control of the whole life cycle of the building is realized.
10. 10. A method of predicting risk of deformation of a building, based on the system of predicting risk of deformation of a building according to any one of claims 1 to 9, comprising the steps of: s1, aiming at each building in a building group, acquiring deformation risk data in real time through corresponding monitoring equipment, and carrying out normalization processing on the deformation risk data; S2, calculating to obtain a deformation risk score of the corresponding building by combining a preset weight coefficient based on the preprocessed index normalization values in a multi-index linear weighting summation mode; s3, setting a fixed monitoring period, acquiring deformation risk scores of the current and last monitoring periods for the same building, calculating the absolute change of scores of the two periods, and converting to obtain the deformation risk score change rate in unit time by combining the time intervals of the monitoring periods; S4, combining deformation risk scores and score change rates of the building, dividing the building into four deformation risk grades, namely low, medium, higher and high, by adopting a preset grading threshold, and matching corresponding disposal measures aiming at different risk grades; S5, aiming at the high-deformation risk building, locking the deformation core causes and compiling a plurality of sets of alternative maintenance schemes, calculating the predicted lifting coefficient of each set of scheme through a lifting coefficient prediction model, and combining threshold screening and engineering actual factors to finish comparison and final determination of the maintenance schemes; And S6, calculating the predicted instability probability of the building after a preset period by using the fixed number of monitoring periods as an evaluation period through an instability probability prediction model, and adjusting a management and control strategy based on the instability probability level to realize the closed-loop safety management and control of the whole life cycle of the building.

Description

Building deformation risk prediction method and system Technical Field The invention relates to the technical field of buildings, in particular to a method and a system for predicting building deformation risk. Background The building deformation is a core index reflecting the safety states of the foundation and the main structure of the building, whether the construction period of the building is controlled or the whole life cycle operation and maintenance of the existing building is carried out, the deformation problems of uneven settlement, structure inclination, crack expansion and the like can possibly cause cracking of structural members, bearing capacity reduction and even serious safety accidents such as building collapse and the like, and the building deformation is directly related to the life and property safety of people and is a core link of the safety control of the building engineering. The prior art is mostly limited by risk assessment and treatment decision-making, and can not quantify the risk management effects of different maintenance schemes, so that the maintenance scheme is blind in model selection, the problem of 'bounce of risk due to insufficient treatment' or 'cost waste due to excessive treatment' easily occurs, the prior art is lack of long-term risk pre-judgment capability, and can not combine with the operation history and structure degradation rule of a building, so that the unstability risk of long-term service of the building is scientifically predicted, and the safety management and control of the whole life cycle of the building are difficult to support. Disclosure of Invention The invention provides a method and a system for predicting building deformation risk in order to solve the problems in the background technology. In order to achieve the purpose, the invention provides a building deformation risk prediction system, which comprises the following modules: The data acquisition module is used for acquiring deformation risk data of the building; The deformation risk assessment module is used for generating deformation risk scores of the building according to the deformation risk data, and calculating the deformation risk score change rate of the building according to the deformation risk scores; the deformation risk grading module is used for grading the deformation risk of the building according to the deformation risk score and the deformation risk score change rate, and taking corresponding measures for the buildings with different grades; The maintenance scheme decision module is used for acquiring historical lifting coefficient data, constructing a lifting coefficient prediction model, predicting the lifting coefficient of the deformation risk building by using the lifting coefficient prediction model, deciding and selecting a specific maintenance scheme according to the predicted lifting coefficient, and calculating according to the predicted lifting coefficient to obtain the deformation risk score of the maintained high deformation risk building; The destabilization probability prediction module is used for acquiring historical destabilization probability data and constructing a destabilization probability prediction model, predicting the destabilization probability of the building by using the destabilization probability prediction model, and taking countermeasures for the building according to the predicted destabilization probability. Further, a plurality of buildings form a building group, each building is provided with a monitoring unit, and the monitoring units are used for collecting deformation risk data, wherein the deformation risk data comprise vertical settlement amount, horizontal displacement amount, overall inclination rate, crack expansion rate and steel bar corrosion rate; Through laying the hydrostatic level gauge and the GNSS receiver at the bottoms of the building foundation and the bearing column, taking the site stability datum point as reference acquisition data, calculating to obtain vertical settlement, and carrying out normalization processing on the vertical settlement: In the formula, In order to normalize the vertical settlement amount,For the measured vertical settlement amount,Is an allowable upper limit value of the vertical settlement amount; The method comprises the steps of acquiring plane coordinate data through GNSS receivers and total stations arranged on a building roof and vertical bearing members, obtaining horizontal displacement after error correction, and carrying out normalization processing on the horizontal displacement: In the formula, In order to normalize the amount of horizontal displacement,As the amount of horizontal displacement to be measured,An allowable upper limit value for the horizontal displacement amount; Through the combination measurement and calculation of three modes of vertical measurement and sedimentation difference inversion of the inclination sensor and the total station, the maximum value of