A tunnel safety assessment early warning system and method based on artificial intelligence

Abstract

The present invention discloses a tunnel safety assessment and early warning system and method based on artificial intelligence, which belongs to the technical field of tunnel safety monitoring. A tunnel information database is established, crack detection sensors are arranged, and a time period sequence is initialized; the safety index related to the crack is quantified, and the tunnel is subjected to abnormal generalization processing; the abnormal generalization curve of the tunnel is drawn, and curve intersection processing is performed in a unified coordinate system; based on the curve intersection relationship, the early warning linkage relationship between tunnels is analyzed, and an early warning linkage sequence is generated and sent to a working port. The present invention can monitor tunnel cracks at a fixed point for a long time, and at the same time, by quantifying the safety index value of the crack width, the normal distribution of the crack width is fed back, and the maximum and minimum value measurement methods are introduced to evaluate the generalization scale of each safety index value under the normal distribution, and the changes in the generalization scale are recorded for a long time in the form of a curve, and then the abnormal linkage characteristics of each tunnel under the generalization scale are analyzed to realize the linkage early warning between tunnel groups.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel safety monitoring, and in particular to an artificial intelligence-based tunnel safety assessment and early warning system and method. Background Art

[0002] With the rapid development of transportation infrastructure, tunnels, as key engineering structures that connect geographical space and improve transportation efficiency, are increasing in number and size. However, tunnels are easily affected by various factors during use, such as changes in geological conditions, improper construction quality control, excessive external loads, etc., which lead to the generation and expansion of tunnel cracks, exacerbating problems such as damage to tunnel linings and water leakage, and thus affecting the overall stability and safety of the tunnel.

[0003] In the prior art, crack monitoring sensors are generally used to monitor the generation and expansion of tunnel cracks. Crack monitoring sensors include ultrasonic sensors, resistive strain sensors, and infrared detectors. However, the sensors can only monitor a single location point and feedback the monitoring data. Due to the different lengths of tunnels, the number of sensor monitoring points set up is also different, making the single-point monitoring method unable to meet the abnormal feedback requirements of long-distance tunnels. Summary of the invention

[0004] The purpose of the present invention is to provide a tunnel safety assessment and early warning system and method based on artificial intelligence to solve the problems raised in the above background technology.

[0005] A tunnel safety assessment and early warning system based on artificial intelligence, the system comprises: a data initialization module, a data quantification module, a curve characterization analysis module and an early warning linkage analysis module;

[0006] The data initialization module is used to establish a tunnel information database, arrange crack detection sensors, and initialize a time period sequence;

[0007] The data quantification module is used to quantify the safety index related to the cracks and perform abnormal generalization processing on the tunnel;

[0008] The curve characterization and analysis module is used to draw the abnormal generalization curve of the tunnel and perform curve intersection processing in a unified coordinate system;

[0009] The early warning linkage analysis module analyzes the early warning linkage relationship between tunnels based on the curve intersection relationship, generates an early warning linkage sequence, and sends it to the working port.

[0010] Further, the data initialization module includes a tunnel information base unit and a period statistics unit;

[0011] The tunnel information database unit is used to establish a tunnel information database and compile the serial numbers of crack detection sensors;

[0012] The period statistics unit is used to initialize the time period sequence so that the crack detection sensor collects crack width data at the same frequency according to the time period sequence.

[0013] Further, the data quantification module includes a safety index quantification unit and an abnormal generalization quantification unit;

[0014] The safety index quantification unit is used to quantify the safety index value of the crack width and generate an abnormal feature set;

[0015] The abnormal generalization quantification unit evaluates and feeds back the abnormal generalization degree of the tunnel based on the abnormal feature set.

[0016] Furthermore, the curve characterization and analysis module includes a coordinate system construction unit and a curve intersection processing unit;

[0017] The coordinate system construction unit is used to construct a two-dimensional coordinate system with periodic nodes as abscissa values ​​and anomaly generalization degrees as ordinate values, and draw anomaly generalization curves of different tunnels;

[0018] The curve intersection processing unit marks the intersection points between the abnormal generalization curves based on the two-dimensional coordinate system, encodes the intersection points, maps the intersection points in the minimum continuous cycle segment based on the time period sequence, and generates a mapping relationship. The minimum continuous cycle segment is composed of two consecutive adjacent cycle nodes.

[0019] Further, the early warning linkage analysis module includes a linkage relationship processing unit and an early warning sequence generation unit;

[0020] The linkage relationship processing unit calculates the generalization proximity between tunnels and the tunnel warning linkage relationship degree based on the mapping relationship;

[0021] The warning sequence generation unit is used to take a tunnel as the warning linkage center, coordinate the tunnel warning linkage relationship degrees between each tunnel except the tunnel corresponding to the warning linkage center and the tunnel corresponding to the warning linkage center, and arrange the tunnel warning linkage relationship degrees in descending order to generate a warning linkage sequence.

[0022] A tunnel safety assessment and early warning method based on artificial intelligence, the method comprises the following steps:

[0023] S1: Establish a tunnel information database, arrange crack detection sensors, and initialize the time cycle sequence;

[0024] S2: quantify crack-related safety indices and generalize anomalies to tunnels;

[0025] S3: Draw the abnormal generalization curve of the tunnel and perform curve intersection processing in the unified coordinate system;

[0026] S4: Based on the curve intersection relationship, the early warning linkage relationship between tunnels is analyzed, and an early warning linkage sequence is generated and sent to the working port.

[0027] Furthermore, the S1 comprises the following steps:

[0028] Establish tunnel information database TU = {TU i |i∈[1,m]}, where TU i represents the i-th tunnel, m is the total number of tunnels; in tunnel TU i Crack detection sensors are arranged at different positions in the crack detection system to generate a sensor registration database CDS. i ={CDS in |n∈[1,N]}, where CDS in represents the nth crack detection sensor in the ith tunnel, and N is the total number of crack detection sensors;

[0029] Define the time period sequence t={t y |y∈[1,T]}, where t y is the yth periodic node, T is the total number of periodic nodes; at periodic node t y When the crack detection sensor is used, the crack width data is collected at the same frequency, and the crack data set C(t y ,TU i ,CDS in )={CW(CDS in |t y )|n∈[1,N]}, where CW(CDS in |t y ) indicates that at the period node t y When the crack detection sensor CDS in Measured crack width.

[0030] Further, the S2 comprises the following steps:

[0031] At cycle node t y When the crack width CW (CDS in |t y ) is calculated using the following formula:

[0032]

[0033] Among them, CSI[CW(CDS in |t y)] represents the crack width CW (CDS in |t y ), μ is the expected safety value, σ 2 is the variance, and is the mean crack width;

[0034] According to the above method, the safety index value is based on the normal distribution characteristics generated by the crack width. When the crack width is closer to the preset expected safety value, the risk probability of the crack width is lower;

[0035] Preset safety index threshold, if the safety index value CSI[CW(CDS in |t y )] is less than or equal to the safety index threshold, the safety index value CSI[CW(CDS in |t y )] recorded in the abnormal feature set AC(TU i , t y ), otherwise the safety index value CSI[CW(CDS in |t y )] removed;

[0036] Based on the abnormal feature set AC[TU i , t y ], at period node t y When evaluating and providing feedback on tunnel TU i The abnormal generalization degree is as follows:

[0037]

[0038] Among them, AC(TU i , t y ) indicates that at the period node t y Time Tunnel TU i The abnormal generalization degree, max{CSI[CW(CDS in |t y )]∈AC[TU i , t y ]} means in the abnormal feature set AC[TU i , t y ], select the maximum safety index value, min{CSI[CW(CDS in |t y )]∈AC[TU i , t y ]} means in the abnormal feature set AC[TU i , t y ] Select the minimum safety index value;

[0039] According to the above method, the anomaly generalization degree reflects the degree of deviation of the maximum and minimum values ​​from the preset expected safety value. The smaller the anomaly generalization degree, the narrower the range of the safety index value, and the more concentrated the data points are near the preset expected safety value. The anomaly generalization degree is used to record the abnormal conditions of cracks monitored by each sensor for a long time.

[0040] Further, the S3 comprises the following steps:

[0041] With the periodic node as the horizontal coordinate value and the abnormal generalization degree as the vertical coordinate value, a two-dimensional coordinate system is constructed to draw the abnormal generalization curve of the tunnel; i The corresponding abnormal generalization curve is denoted as GC(TU i ), the rth tunnel TU r The corresponding abnormal generalization curve is denoted as GC(TU r ), and i≠r;

[0042] In the two-dimensional coordinate system, the abnormal generalization curve GC (TU i ) and abnormal generalization curve GC(TU r ) between them; the intersections are numbered in sequence, and each intersection is mapped to a unique minimum continuous period fragment in the time period sequence t, and the fth intersection is recorded as PI f , then the intersection point PI f The unique minimum continuous period segment corresponding to the mapping is denoted as [t y , t y+1 ], then the mapping relationship PI f :[t y , t y+1 ].

[0043] Further, the S4 comprises the following steps:

[0044] Based on the mapping relationship PI f :[t y , t y+1 ], calculate the tunnel early warning linkage relationship degree, the formula is as follows:

[0045]

[0046] Among them, RV(TU i →TU r ) indicates tunnel TU i and Tunnel TU r The tunnel warning linkage relationship between i →TU r )|PI f ] indicates that in the mapping relationship PI f :[t y , ty+1 ], tunnel TU i and Tunnel TU r The generalization closeness between them, F represents the abnormal generalization curve GC (TU i ) and abnormal generalization curve GC(TU r ) between the total number of intersections;

[0047] According to the above method, the tunnel early warning linkage relationship can feedback the overlap degree between the curves, and the intersection position feedbacks the abnormal correlation between the tunnels. The present invention uses the minimum continuous period fragment of the intersection mapping to segmentally quantify the overlap of each curve segment, and then accurately analyzes the abnormal linkage degree between the tunnels. The larger the tunnel early warning linkage relationship, the smaller the abnormal linkage between the tunnels.

[0048] Tunnel TU i For the early warning linkage center, calculate the tunnel TU i Each tunnel outside the tunnel TU i The tunnel warning linkage relationship degrees between them are calculated, and the tunnel warning linkage relationship degrees are arranged in descending order to generate a warning linkage sequence, which is sent to the working port.

[0049] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: in a tunnel safety assessment and early warning system and method based on artificial intelligence provided by the present invention, a tunnel information database is established, crack detection sensors are arranged, and a time period sequence is initialized; the safety index related to the crack is quantified, and the tunnel is subjected to abnormal generalization processing; the abnormal generalization curve of the tunnel is drawn, and curve intersection processing is performed in a unified coordinate system; based on the curve intersection relationship, the early warning linkage relationship between tunnels is analyzed, and an early warning linkage sequence is generated and sent to the working port. The present invention can monitor the cracks of the tunnel at a fixed point for a long time, and at the same time, by quantifying the safety index value of the crack width, the normal distribution of the crack width is fed back, and the maximum and minimum value measurement methods are introduced to evaluate the generalization scale of each safety index value under the normal distribution, and the changes in the generalization scale are recorded for a long time in the form of a curve, and then the abnormal linkage characteristics of each tunnel under the generalization scale are analyzed to realize the linkage early warning between tunnel groups. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0051] Figure 1 It is a structural schematic diagram of a tunnel safety assessment and early warning system based on artificial intelligence of the present invention;

[0052] Figure 2It is a schematic diagram of the steps of an artificial intelligence-based tunnel safety assessment and early warning method of the present invention. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] See also Figure 1 , in the first embodiment: a tunnel safety assessment and early warning system based on artificial intelligence is provided, the system comprising: a data initialization module, a data quantification module, a curve characterization analysis module and an early warning linkage analysis module;

[0055] A data initialization module is used to establish a tunnel information database, arrange crack detection sensors, and initialize the time period sequence;

[0056] Exemplarily, the data initialization module includes a tunnel information base unit and a cycle statistics unit;

[0057] A tunnel information database unit is used to establish a tunnel information database and compile the serial numbers of crack detection sensors;

[0058] The periodic statistical unit is used to initialize the time period sequence so that the crack detection sensor collects crack width data at the same frequency according to the time period sequence;

[0059] Data quantification module, used to quantify crack-related safety indices and generalize anomalies for tunnels;

[0060] Exemplarily, the data quantification module includes a safety index quantification unit and an abnormal generalization quantification unit;

[0061] A safety index quantification unit is used to quantify the safety index value of the crack width and generate an abnormal feature set;

[0062] The anomaly generalization quantification unit evaluates and provides feedback on the anomaly generalization of the tunnel based on the anomaly feature set;

[0063] The curve characterization and analysis module is used to draw the abnormal generalization curve of the tunnel and perform curve intersection processing in a unified coordinate system;

[0064] Exemplarily, the curve characterization analysis module includes a coordinate system construction unit and a curve intersection processing unit;

[0065] A coordinate system construction unit is used to construct a two-dimensional coordinate system with periodic nodes as horizontal coordinate values ​​and anomaly generalization degree as vertical coordinate values, and draw anomaly generalization curves of different tunnels;

[0066] The curve intersection processing unit marks the intersection points between the abnormal generalization curves based on the two-dimensional coordinate system, encodes the intersection points, maps the intersection points into the minimum continuous cycle segment based on the time period sequence, and generates a mapping relationship. The minimum continuous cycle segment is composed of two consecutive adjacent cycle nodes.

[0067] The early warning linkage analysis module analyzes the early warning linkage relationship between tunnels based on the curve intersection relationship, generates an early warning linkage sequence, and sends it to the working port;

[0068] Exemplarily, the early warning linkage analysis module includes a linkage relationship processing unit and an early warning sequence generation unit;

[0069] A linkage relationship processing unit calculates the generalization proximity between tunnels and the tunnel warning linkage relationship degree based on the mapping relationship;

[0070] The warning sequence generation unit is used to take a tunnel as the warning linkage center, coordinate the tunnel warning linkage relationship between each tunnel except the tunnel corresponding to the warning linkage center and the tunnel corresponding to the warning linkage center, and arrange the tunnel warning linkage relationship in order from large to small to generate a warning linkage sequence.

[0071] See also Figure 2 In the second embodiment, a tunnel safety assessment and early warning method based on artificial intelligence is provided, and the method comprises the following steps:

[0072] S1: Establish a tunnel information database, arrange crack detection sensors, and initialize the time cycle sequence;

[0073] Specifically, a tunnel information database TU={TU i |i∈[1,m]}, where TU i represents the i-th tunnel, m is the total number of tunnels; in tunnel TU i Crack detection sensors are arranged at different positions in the crack detection system to generate a sensor registration database CDS. i =CDS in |n∈[1,N]}, where CDS in represents the nth crack detection sensor in the ith tunnel, and N is the total number of crack detection sensors;

[0074] Define the time period sequence t={t y |y∈[1,T]}, where t y is the yth periodic node, T is the total number of periodic nodes; at periodic node ty When the crack detection sensor is used, the crack width data is collected at the same frequency, and the crack data set C(t y ,TU i ,CDS in )={CW(CDS in |t y )|n∈[1,N]}, where CW(CDS in |t y ) indicates that at the period node t y When the crack detection sensor CDS in The measured crack width, usually in millimeters;

[0075] S2: quantify crack-related safety indices and generalize anomalies to tunnels;

[0076] Specifically, at cycle node t y When the crack width CW (CDS in |t y ) is calculated using the following formula:

[0077]

[0078] Among them, CSI[CW(CDS in |t y )] represents the crack width CW (CDS in |t y ), μ is the expected safety value, σ 2 is the variance, and is the mean crack width;

[0079] Preset safety index threshold, if the safety index value CSI[CW(CDS in |t y )] is less than or equal to the safety index threshold, the safety index value CSI[CW(CDS in |t y )] recorded in the abnormal feature set AC(TU i , t y ), otherwise the safety index value CSI[CW(CDS in |t y )] removed;

[0080] Based on the abnormal feature set AC[TU i , t y ], at period node t y When evaluating and providing feedback on tunnel TU i The abnormal generalization degree is as follows:

[0081]

[0082] Among them, AC(TU i , t y ) indicates that at the period node t y Time Tunnel TU i The abnormal generalization degree, max{CSI[CW(CDS in |t y )]∈AC[TU i , t y ]} means in the abnormal feature set AC[TU i , t y ], select the maximum safety index value, min{CSI[CW(CDS in |t y )]∈AC[TU i , t y ]} means in the abnormal feature set AC[TU i , t y ] Select the minimum safety index value;

[0083] S3: Draw the abnormal generalization curve of the tunnel and perform curve intersection processing in the unified coordinate system;

[0084] Specifically, a two-dimensional coordinate system is constructed with periodic nodes as the horizontal coordinate value and abnormal generalization degree as the vertical coordinate value to draw the abnormal generalization curve of the tunnel; i The corresponding abnormal generalization curve is denoted as GC(TU i ), the rth tunnel TU r The corresponding abnormal generalization curve is denoted as GC(TU r ), and i≠r;

[0085] In the two-dimensional coordinate system, mark the abnormal generalization curve GC (TU i ) and abnormal generalization curve GC(TU r ) between them; the intersections are numbered in sequence, and each intersection is mapped to a unique minimum continuous period fragment in the time period sequence t, and the fth intersection is recorded as PI f , then the intersection point PI f The unique minimum continuous period segment corresponding to the mapping is denoted as [t y , t y+1 ], then the mapping relationship PI f :[t y , t y+1 ];

[0086] S4: Based on the curve intersection relationship, analyze the early warning linkage relationship between tunnels, generate an early warning linkage sequence, and send it to the working port;

[0087] Specifically, based on the mapping relationship PI f :[t y , t y+1 ], calculate the tunnel early warning linkage relationship degree, the formula is as follows:

[0088]

[0089] Among them, RV(TU i →TU r ) indicates tunnel TU i and Tunnel TU r The tunnel warning linkage relationship between i →TU r )|PI f ] indicates that in the mapping relationship PI f :[t y , t y+1 ], tunnel TU i and Tunnel TU r The generalization closeness between them, F represents the abnormal generalization curve GC (TU i ) and abnormal generalization curve GC(TU r ) between the total number of intersections;

[0090] Tunnel TU i For the early warning linkage center, calculate the tunnel TU i Each tunnel outside the tunnel TU i The tunnel warning linkage relationship degrees between them are calculated, and the tunnel warning linkage relationship degrees are arranged in descending order to generate a warning linkage sequence, which is sent to the working port.

[0091] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0092] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A tunnel safety assessment and early warning method based on artificial intelligence, characterized in that: The method comprises the following steps: S1: Establish a tunnel information database, arrange crack detection sensors, and initialize the time cycle sequence; S2: quantify crack-related safety indices and generalize anomalies to tunnels; S3: Draw the abnormal generalization curve of the tunnel and perform curve intersection processing in the unified coordinate system; S4: Based on the curve intersection relationship, analyze the early warning linkage relationship between tunnels, generate an early warning linkage sequence, and send it to the working port; The S1 comprises the following steps: Establish tunnel information database TU = {TU i |i∈[1,m]}, where TU i represents the i-th tunnel, m is the total number of tunnels; in tunnel TU i Crack detection sensors are arranged at different positions in the crack detection system to generate a sensor registration database CDS. i ={CDS in |n∈[1,N]}, where CDS in represents the nth crack detection sensor in the ith tunnel, and N is the total number of crack detection sensors; Define the time period sequence t={t y |y∈[1,T]}, where t y is the yth periodic node, T is the total number of periodic nodes; at periodic node t y When the crack detection sensor is used, the crack width data is collected at the same frequency, and the crack data set C(t y ,TU i ,CDS in )={CW(CDS in |t y )|n∈[1,N]}, where CW(CDS in |t y ) indicates that at the period node t y When the crack detection sensor CDS in Measured crack width; The S2 comprises the following steps: At cycle node t y When the crack width CW (CDS in |t y ) is calculated using the following formula: Among them, CSI[CW(CDS in |t y )] represents the crack width CW (CDS in |t y ), μ is the expected safety value, σ 2 is the variance, and is the mean crack width; Preset safety index threshold, if the safety index value CSI[CW(CDS in |t y )] is less than or equal to the safety index threshold, the safety index value CSI[CW(CDS in |t y )] recorded in the abnormal feature set AC(TU i , t y ), otherwise the safety index value CSI[CW(CDS in |t y )] removed; Based on the abnormal feature set AC[TU i , t y ], at period node t y When evaluating and providing feedback on tunnel TU i The abnormal generalization degree is as follows: Among them, AC(TU i , t y ) indicates that at the period node t y Time Tunnel TU i The abnormal generalization degree, max{CSI[CW(CDS in |t y )]∈AC[TU i , t y ]} means in the abnormal feature set AC[TU i , t y ], select the maximum safety index value, min{CSI[CW(CDS in |t y )]∈AC[TU i , t y ]} means in the abnormal feature set AC[TU i , t y ] Select the minimum safety index value; The S3 comprises the following steps: With the periodic node as the horizontal coordinate value and the abnormal generalization degree as the vertical coordinate value, a two-dimensional coordinate system is constructed to draw the abnormal generalization curve of the tunnel; i The corresponding abnormal generalization curve is denoted as GC(TU i ), the rth tunnel TU r The corresponding abnormal generalization curve is denoted as GC(TU r ), and i≠r; In the two-dimensional coordinate system, the abnormal generalization curve GC (TU i ) and abnormal generalization curve GC(TU r ) between them; the intersections are numbered in sequence, and each intersection is mapped to a unique minimum continuous period fragment in the time period sequence t, and the fth intersection is recorded as PI f , then the intersection point PI f The unique minimum continuous period segment corresponding to the mapping is denoted as [t y , t y+1 ], then the mapping relationship PI f :[t y , t y+1 ]; Described S4 comprises the following steps: Based on the mapping relationship PI f :[t y , t y+1 ], calculate the tunnel early warning linkage relationship degree, the formula is as follows: Among them, RV(TU i →TU r ) indicates tunnel TU i and Tunnel TU r The tunnel warning linkage relationship between i →TU r )|PI f ] indicates that in the mapping relationship PI f :[t y , t y+1 ], tunnel TU i and Tunnel TU r The generalization closeness between them, F represents the abnormal generalization curve GC (TU i ) and abnormal generalization curve GC(TU r ) between the total number of intersections; Tunnel TU i For the early warning linkage center, calculate the tunnel TU i Each tunnel outside the tunnel TU i The tunnel warning linkage relationship degrees between them are calculated, and the tunnel warning linkage relationship degrees are arranged in descending order to generate a warning linkage sequence, which is sent to the working port.

2. A tunnel safety assessment and early warning system based on artificial intelligence, executing the tunnel safety assessment and early warning method based on artificial intelligence as claimed in claim 1, characterized in that: The system includes: a data initialization module, a data quantification module, a curve characterization analysis module and an early warning linkage analysis module; The data initialization module is used to establish a tunnel information database, arrange crack detection sensors, and initialize a time period sequence; The data quantification module is used to quantify the safety index related to the cracks and perform abnormal generalization processing on the tunnel; The curve characterization and analysis module is used to draw the abnormal generalization curve of the tunnel and perform curve intersection processing in a unified coordinate system; The early warning linkage analysis module analyzes the early warning linkage relationship between tunnels based on the curve intersection relationship, generates an early warning linkage sequence, and sends it to the working port.

3. The tunnel safety assessment and early warning system based on artificial intelligence according to claim 2 is characterized in that: The data initialization module includes a tunnel information base unit and a cycle statistics unit; The tunnel information database unit is used to establish a tunnel information database and compile the serial numbers of crack detection sensors; The period statistics unit is used to initialize the time period sequence so that the crack detection sensor collects crack width data at the same frequency according to the time period sequence.

4. The tunnel safety assessment and early warning system based on artificial intelligence according to claim 3 is characterized in that: The data quantification module includes a safety index quantification unit and an abnormal generalization quantification unit; The safety index quantification unit is used to quantify the safety index value of the crack width and generate an abnormal feature set; The abnormal generalization quantification unit evaluates and feeds back the abnormal generalization degree of the tunnel based on the abnormal feature set.

5. The tunnel safety assessment and early warning system based on artificial intelligence according to claim 4 is characterized in that: The curve characterization and analysis module includes a coordinate system construction unit and a curve intersection processing unit; The coordinate system construction unit is used to construct a two-dimensional coordinate system with periodic nodes as abscissa values ​​and anomaly generalization degrees as ordinate values, and draw anomaly generalization curves of different tunnels; The curve intersection processing unit marks the intersection points between the abnormal generalization curves based on the two-dimensional coordinate system, encodes the intersection points, maps the intersection points in the minimum continuous cycle segment based on the time period sequence, and generates a mapping relationship. The minimum continuous cycle segment is composed of two consecutive adjacent cycle nodes.

6. The tunnel safety assessment and early warning system based on artificial intelligence according to claim 5 is characterized by: The early warning linkage analysis module includes a linkage relationship processing unit and an early warning sequence generation unit; The linkage relationship processing unit calculates the generalization proximity between tunnels and the tunnel warning linkage relationship degree based on the mapping relationship; The warning sequence generation unit is used to take a tunnel as the warning linkage center, coordinate the tunnel warning linkage relationship degrees between each tunnel except the tunnel corresponding to the warning linkage center and the tunnel corresponding to the warning linkage center, and arrange the tunnel warning linkage relationship degrees in descending order to generate a warning linkage sequence.

Images