Method and system for judging accuracy of advanced geological prediction system
阅读说明：本技术 一种判别超前地质预报体系准确率的方法及系统 (Method and system for judging accuracy of advanced geological prediction system ) 是由 钟果 刘云鹏 张世殊 冉从彦 吴章雷 杨静熙 肖华波 于 2020-08-07 设计创作，主要内容包括：本发明公开了一种判别超前地质预报体系准确率的方法,涉及施工技术及技术标准领域,对开挖工程进行超前地质预报,首先明确预报对象,针对预报对象实施超前地质预报；当施工开挖后,对工程开挖段进行地质对象的复核和评价,获取全段的真实地质条件；将全段实际地质条件与预报结论进行对比,得到该预报体系下各对象的基本预测综合评分及预报准确度,最终经公式运算得到该预报体系的预报准确率。本发明的有益效果是可得到某超前地质预报体系下某工程整体的预报准确率,从而可对同一预报体系在不同地质环境下及不同预报体系在同一地质环境下进行预报整体效果的比较,以评判其优劣；同时可根据不同预报体系各因子的分值,针对性地进行预报体系的改进。(The invention discloses a method for judging the accuracy of an advanced geological prediction system, which relates to the field of construction technology and technical standard, and is used for carrying out advanced geological prediction on excavation projects, wherein firstly, a prediction object is determined, and the advanced geological prediction is carried out aiming at the prediction object; after construction excavation, rechecking and evaluating a geological object on an engineering excavation section to obtain the real geological condition of the whole section; and comparing the whole actual geological condition with the forecast conclusion to obtain the basic comprehensive forecast score and the forecast accuracy of each object under the forecast system, and finally obtaining the forecast accuracy of the forecast system through formula operation. The invention has the advantages that the forecasting accuracy of the whole engineering under a certain advanced geological forecasting system can be obtained, so that the overall forecasting effects of the same forecasting system under different geological environments and the same geological environment can be compared to judge the advantages and disadvantages of the same forecasting system; meanwhile, the forecasting system can be improved in a targeted manner according to the values of all factors of different forecasting systems.)
1. A method for judging the accuracy of a leading geological forecast system is characterized by comprising the following steps: the method comprises the following steps:
s1: respectively naming the existing advanced geological prediction systems needing to be distinguished and compared as a prediction system M and a prediction system N;
s2: adopting the existing forecasting system M to carry out advanced geological forecasting on an object needing advanced geological forecasting, carrying out advanced geological forecasting on an excavation project and obtaining a forecasting conclusion of the forecasting object;
s3: adopting an existing forecasting system N to carry out advanced geological forecasting on an object needing advanced geological forecasting, carrying out advanced geological forecasting on an excavation project, and obtaining a forecasting conclusion of a forecasting object;
s4: after the construction excavation is finished, rechecking and evaluating the geological conditions of the whole engineering excavation section to obtain the real geological conditions of the whole engineering excavation section;
s5: dividing the whole section into a plurality of statistical sections according to different geologic body types according to the real geological conditions of the whole section;
s6: under a forecasting system M, comparing the whole actual geological condition with a forecasting conclusion to obtain a basic forecasting comprehensive score of each object under the forecasting system M, comparing the position of each object to be forecasted under the whole actual geological condition with the corresponding forecasting position to calculate the forecasting accuracy of each statistical section, multiplying the basic forecasting comprehensive score corresponding to each object with the forecasting accuracy, accumulating and summing, and dividing by the total statistical section number to obtain the forecasting accuracy of the forecasting system M;
s7: under a forecasting system N, comparing the whole-section actual geological condition with a forecasting conclusion to obtain a basic forecasting comprehensive score of each object under the forecasting system N, comparing the position of each object to be forecasted under the whole-section actual geological condition with the corresponding forecasting position to calculate the forecasting accuracy of each statistical section, multiplying the basic forecasting comprehensive score corresponding to each object with the forecasting accuracy, performing accumulation and summation, and dividing by the total statistical section number to obtain the forecasting accuracy of the forecasting system N;
and S8, comparing the prediction accuracy under the prediction system M with the prediction accuracy under the prediction system N, comparing the prediction accuracy of the two systems, further comparing the detailed scores of the two systems to obtain key factors influencing the prediction accuracy, and further performing prediction improvement on the corresponding prediction system.
2. The method for judging the accuracy of the advanced geological forecast system according to claim 1, wherein the basic forecast comprehensive score of any statistical section is obtained by multiplying the score of the goodness of fit of the section forecast object type, the score of the goodness of fit of the section forecast geologic body in terms of properties and phenomena, and the score of the goodness of fit of the section forecast geologic body on the construction influence by the corresponding weight coefficients respectively, and then summing up the scores, wherein the formula is as follows: a. thei＝axiAxi+ayiAyi+aziAziWherein A isiFor the basic prediction composite score of a single prediction object in a statistical section, AxiForecasting object type kiss for a single segment in a statistical segmentFitness score, AyiForecasting the object property and phenomenological goodness of fit score for the segment, AziScoring the severity goodness of fit of the forecast object to the construction impact, axi、ayi、aziAre the corresponding weight coefficients.
3. The method according to claim 1, wherein the accuracy of the advanced geological prediction system is determined by determining the prediction accuracy corresponding to the basic comprehensive prediction score of any statistical section, and after the construction of each statistical section is completed, the actual prediction score is measured to obtain the actual position of the geological body of each statistical section, and then the actual position of each statistical section is compared with the prediction position of each section, so as to obtain the prediction accuracy of any statistical section, i.e. the prediction accuracy of any statistical section is obtained by subtracting the length of the missed report section in the prediction result from the length of the hit section in the prediction result of the section, subtracting the length of the false report section in the prediction result from the length of the false report section by the reduction coefficient, and then dividing the true exposure length of the predicted geological body of the section in the tunnel by the false report reduction coefficient, and the accuracy of each statistical section is formulated as:wherein: parameter PiForecasting a parameter A of an object for a respective segmentiAccuracy of LiForecasting the real exposure length l of a single segment in a tunnel for countingiPredicting the length of the hit segment in the result for that segment, piPredicting the length of the missing segment in the result for this segment, diAnd k is the length of the false alarm segment in the segment prediction result, and the k is the false alarm reduction coefficient.
4. The method of claim 1, wherein the accuracy of the construction engineering forecast is determined by summing the products of the basic forecast composite scores of the statistical sections and the corresponding accuracies, and dividing the sum by the total number of the sections to perform a total differentiation, wherein the accuracy of the construction engineering forecast is formulated as:
5. The method for judging the accuracy of the advanced geological prediction system according to claim 2, wherein the score of the goodness of fit of the type of the prediction object is based on the type of the single-section prediction object in the statistical section and the corresponding goodness of fit, and the value range of the score is 0-100; the forecast object property and phenomenon goodness of fit score ranges from 0 to 100 according to the forecast object property and phenomenon goodness of fit of the section; the evaluation of the severity goodness of fit of the forecast object on the construction influence ranges from 0 to 100 according to the severity goodness of fit of the forecast object on the construction influence; the corresponding weight coefficients are determined according to specific construction requirements, and the condition that all the weight coefficients are non-negative numbers and the sum of all the weight coefficients is equal to 1 is required to be met.
6. The method of claim 3, wherein the false positive reduction factor is empirically 0.8.
7. The method for judging the accuracy of the advanced geological prediction system according to claim 2, wherein the object types in the forecast object type goodness of fit score can be classified into surrounding rock types and poor geologic body types, the goodness of fit is graded and graded, the surrounding rock types can be classified into four grades including an identical grade, a prediction difference grade 1 grade, a prediction difference grade 2 grade and a prediction difference grade more than 2 grade, the goodness of fit of the poor geologic body types is graded and graded into three grades of an identical grade, a basic identical grade and a non-identical grade, and each grade is assigned with a value; grading and grading the corresponding forecast object properties and the corresponding phenomenon goodness of fit in the forecast object properties and the phenomenon goodness of fit scores, wherein the grading and grading can be divided into four grades of fit, basic fit, fit and non-fit, and each grade is assigned with a value correspondingly; and (3) grading and grading the severity goodness of fit of the construction influence by the corresponding object in the severity goodness of fit score of the construction influence by the forecast object, wherein the grading and grading can be divided into four grades of coincidence, basic coincidence, more coincidence and non-coincidence, and each grade is assigned with a value correspondingly.
8. The method according to claim 7, wherein the goodness-of-fit grade assignments are assigned from top to bottom according to four grades of fit, basic fit, better fit and no fit, and the goodness-of-fit grade assignments can be assigned to 100 fits, 80 fits, 60 fits and 20 fits.
9. A system for determining the accuracy of an advanced geological prediction system, which is based on the method for determining the accuracy of an advanced geological prediction system as claimed in any one of claims 1 to 8.
Technical Field
The invention relates to the field of construction technology and technical standard, in particular to a method and a system for judging the accuracy of an advanced geological prediction system.
Background
At present, in order to ensure smooth construction, geological prediction and forecast are mostly needed in tunnel and underground engineering excavation, but most of projects only pay attention to the forecasting effect aiming at a single object in the construction process, the whole effect of engineering forecasting is generally difficult to quantitatively evaluate, and meanwhile, a uniform evaluation standard does not exist in the face of a wide-range forecasting system.
In the related scientific research work, the scientific and quantitative evaluation of the forecasting accuracy is of great significance especially for the comprehensive geological forecasting under the multivariate information fusion which is commonly adopted at present. The evaluation result can guide and compare the aspects of the selection of multivariate information, the formulation of interpretation standards, the fusion analysis and evaluation of information and the like under a certain forecasting system so as to improve the level and the accuracy of comprehensive geological forecasting; meanwhile, the forecasting effects of the same forecasting system in different geological environments and the forecasting effects of different forecasting systems in the same geological environment can be compared transversely and longitudinally to judge the advantages and disadvantages of the forecasting systems.
The existing geological forecast result accuracy rate evaluation mainly has the following defects:
1. if different units carry out prediction and forecast on different tunnels, the result is difficult to obtain the conclusion that the geological forecast accuracy of the tunnel B is 70 percent and the forecast accuracy is better than that of the tunnel C by 60 percent after the result is counted;
2. the statistical objects are unclear, and the invalid data is more (for example, the general tunnel sections with better rock mass quality are added into the statistical range together, and the geological forecast makes correct judgment on the general tunnel sections without obvious influence on the construction relatively easily, so that the final forecast accuracy is obviously higher, and simultaneously, the forecast accuracy of the bad geological tunnel sections with great influence on the construction reflects distortion);
3. the definition of "accuracy" is different, which can cause the statistical difference of accuracy rate to be huge.
Disclosure of Invention
In order to overcome the technical problems, the invention provides a method and a system for judging the accuracy of a leading geological forecast system, wherein the method is used for evaluating, and the level and the accuracy of comprehensive geological forecast are improved by guiding and comparing the aspects of selection of multivariate information, interpretation standard formulation, information fusion analysis evaluation and the like under a certain forecast system; meanwhile, the forecasting effects of the same forecasting system under different geological environments and different forecasting systems under the same geological environment can be compared to judge the advantages and disadvantages of the forecasting systems. The method eliminates invalid data influencing accuracy judgment, enables a statistical result to be closer to a real situation, simplifies the result into a single numerical value, and enables achievement data to be directly compared among different projects, different tunnels and even different standard sections of the same tunnel.
The invention is realized by the following technical scheme:
a method of discriminating the accuracy of a lead geological prediction system, the method comprising the steps of:
s1: respectively naming the existing advanced geological prediction systems needing to be distinguished and compared as a prediction system M and a prediction system N;
s2: adopting the existing forecasting system M to carry out advanced geological forecasting on an object needing advanced geological forecasting, carrying out advanced geological forecasting on an excavation project and obtaining a forecasting conclusion of the forecasting object;
s3: adopting an existing forecasting system N to carry out advanced geological forecasting on an object needing advanced geological forecasting, carrying out advanced geological forecasting on an excavation project, and obtaining a forecasting conclusion of a forecasting object;
s4: after the construction excavation is finished, rechecking and evaluating the geological conditions of the whole engineering excavation section to obtain the real geological conditions of the whole engineering excavation section;
s5: dividing the whole section into a plurality of statistical sections according to different geologic body types according to the real geological conditions of the whole section;
s6: under a forecasting system M, comparing the whole actual geological condition with a forecasting conclusion to obtain a basic forecasting comprehensive score of each object under the forecasting system M, comparing the position of each object to be forecasted under the whole actual geological condition with the corresponding forecasting position to calculate the forecasting accuracy of each statistical section, multiplying the basic forecasting comprehensive score corresponding to each object with the forecasting accuracy, accumulating and summing, and dividing by the total statistical section number to obtain the forecasting accuracy of the forecasting system M;
s7: under a forecasting system N, comparing the whole-section actual geological condition with a forecasting conclusion to obtain a basic forecasting comprehensive score of each object under the forecasting system N, comparing the position of each object to be forecasted under the whole-section actual geological condition with the corresponding forecasting position to calculate the forecasting accuracy of each statistical section, multiplying the basic forecasting comprehensive score corresponding to each object with the forecasting accuracy, performing accumulation and summation, and dividing by the total statistical section number to obtain the forecasting accuracy of the forecasting system N;
and S8, comparing the prediction accuracy under the prediction system M with the prediction accuracy under the prediction system N, comparing the prediction accuracy of the two systems, further comparing the detailed scores of the two systems to obtain key factors influencing the prediction accuracy, and further performing prediction improvement on the corresponding prediction system.
The basic prediction comprehensive score of any statistical section in the forecasting system is obtained by respectively multiplying the score of the type goodness of fit of the section of the forecasting object, the score of the nature and phenomenon goodness of fit of the section of the forecasting geologic body and the score of the goodness of fit of the section of the forecasting geologic body on the construction influence by corresponding weight coefficients, and then summing the scores, wherein the formula is as follows: a. thei＝axiAxi+ayiAyi+aziAziWherein A isiFor the basic prediction composite score of a single prediction object in a statistical section, AxiFor forecasting the type goodness of fit of an object in a single section of a statistical section, AyiForecasting the object property and phenomenological goodness of fit score for the segment, AziScoring the severity goodness of fit of the forecast object to the construction impact, axi、ayi、aziAre the corresponding weight coefficients.
The method comprises the following steps of determining the forecasting accuracy corresponding to the basic forecasting comprehensive score of any statistical section in the forecasting system, after construction of each statistical section is completed, respectively measuring the statistical sections in the field to obtain the real position of a geologic body of each statistical section, then comparing the real position of each statistical section with the forecasting position of each section to obtain the forecasting accuracy of any statistical section, namely subtracting the length of a missed report section in the forecasting result of the section from the length of a hit section in the forecasting result of the section, subtracting the length of a false report section in the forecasting result of the section and multiplying the length of the false report section by a false report reduction coefficient, and then dividing the true exposure length of the forecasting geologic body of the section in a tunnel to obtain the forecasting accuracy of each statistical section, wherein the accuracy of each statistical section is formulated as:wherein: parameter PiForecasting a parameter A of an object for a respective segmentiAccuracy of LiForecasting the real exposure length l of a single segment in a tunnel for countingiPredicting the length of the hit segment in the result for that segment, piPredicting the length of the missing segment in the result for this segment, diAnd k is the length of the false alarm segment in the segment prediction result, and the k is the false alarm reduction coefficient.
In the forecasting system, the accuracy of the construction engineering forecasting is determined by summing the products of basic forecasting comprehensive scores of a plurality of statistical sections and corresponding accuracies of the basic forecasting comprehensive scores, dividing the sum by the total number of the sections, and performing hundred-differentiation to obtain the accuracy of the construction engineering advanced geological forecasting, wherein the accuracy is formulated as:
The score of the goodness of fit of the forecast object types is based on the single forecast object type and the corresponding goodness of fit in the statistical section, and the value range of the score is 0-100; the forecast object property and phenomenon goodness of fit score ranges from 0 to 100 according to the forecast object property and phenomenon goodness of fit of the section; the evaluation of the severity goodness of fit of the forecast object on the construction influence ranges from 0 to 100 according to the severity goodness of fit of the forecast object on the construction influence; the corresponding weight coefficients are determined according to specific construction requirements, and the condition that all the weight coefficients are non-negative numbers and the sum of all the weight coefficients is equal to 1 is required to be met.
The false alarm reduction coefficient is 0.8 according to experience.
The object types in the forecast object type goodness of fit score can be divided into surrounding rock types and poor geologic body types, the goodness of fit is graded and graded, the surrounding rock types can be divided into four grades of fit,
The goodness of fit grade assignment is assigned from top to bottom according to four grades of fit, basic fit, better fit and no fit, and the goodness of fit grade assignment can be assigned to fit 100, basic fit 80, better fit 60 and no fit 20.
A system for judging the accuracy of a leading geological forecast system is based on any of the above methods for judging the accuracy of a leading geological forecast system.
Compared with the prior art, the invention has the following advantages and beneficial effects:
according to the comprehensive advanced geological forecast method, the aspects of selection, interpretation standard formulation and information fusion analysis and evaluation of multivariate information under a certain construction project forecast system are guided and compared, and the level and accuracy of the comprehensive advanced geological forecast of the construction project are improved by the adopted forecast system; meanwhile, the forecasting effects of the same forecasting system under different geological environments and different forecasting systems under the same geological environment can be compared to judge the advantages and disadvantages of the forecasting systems. The method eliminates invalid data influencing accuracy judgment, enables a statistical result to be closer to a real situation, simplifies the result into a single numerical value, and enables achievement data to be directly compared among different projects, different tunnels and even different standard sections of the same tunnel.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
fig. 1 is a schematic diagram illustrating a relationship between an exposed segment of an actual object and a
FIG. 2 is a schematic diagram illustrating a relationship between an exposed segment of an actual object and a
FIG. 3 is a diagram illustrating the relationship between the exposure of the actual object and the distribution of the predicted
FIG. 4 is a diagram illustrating a relationship between an exposed segment of an actual object and a distribution 4 of positions of predicted objects in an embodiment of example 1;
fig. 5 is a schematic diagram illustrating a relationship between an actual object exposure segment and a predicted object position distribution 5 in one embodiment of the following example 1.
In the figure, 1, an actual object is exposed; 2. a tunnel; 3. the object position is predicted.
Detailed Description
The method and system for determining the accuracy of advanced geological prediction according to the present invention will be described in detail below with reference to the accompanying drawings.
- 上一篇：一种医用注射器针头装配设备
- 下一篇：一种基于深度神经网络的强对流天气识别方法