A shield cutter disc mud cake pre-judgment method based on cutter multi-source monitoring information

Abstract

The application discloses a shield cutter disc mud cake prediction method based on cutter multi-source monitoring information, and belongs to the technical field of shield cutter disc detection, and comprises the following steps: collecting the temperature and vibration frequency of each cutter on the shield cutter disc in real time by using temperature and vibration sensors, synchronously recording shield tunneling parameters, and forming a multi-modal data set; based on an improved inverse distance weighted interpolation algorithm, predicting the temperature-vibration distribution of the region where no sensor is arranged, and generating a cutter dynamic thermal map; adopting an isolation forest algorithm to calculate the temperature change rate and vibration frequency change rate, and when the double-parameter change rate exceeds a preset threshold value, mud cake occurs and shutdown inspection is performed; otherwise, the shield machine continues to tunnel, the preset threshold value is dynamically corrected based on a Bayesian algorithm combined with newly collected data, and the preset threshold value is optimized through actual inspection result feedback. The method avoids misjudgment caused by single data monitoring and fixed threshold judgment, and improves the accuracy of shield cutter disc mud cake discrimination.

Description

Technical Field

[0001] This application relates to the field of tunnel boring machine cutterhead detection technology, and in particular to a method for predicting mud cake formation on tunnel boring machine cutterheads based on multi-source monitoring information of the cutterheads. Background Technology

[0002] In recent years, my country's urbanization has accelerated, leading to a significant increase in demand for transportation infrastructure and municipal public service facilities. Shield tunneling technology, due to its high safety, stability, energy efficiency, environmental friendliness, and high construction efficiency, is increasingly widely used in underground engineering construction. However, when shield tunneling is excavating in clay-rich strata such as strongly or moderately weathered mudstone or argillaceous sandstone, the fine clay particles cut off easily adhere to the cutterhead. Under earth pressure and frictional heat generated by the cutterhead, these particles can easily form hardened mud cakes. The formation of mud cakes reduces the penetration depth of the cutterhead, thereby decreasing the efficiency of shield tunneling. It can also cause the cutterhead to jam, wear unevenly, and damage the cutterhead structure, seriously threatening project safety.

[0003] During tunnel boring machine (TBM) construction, as mud cake accumulates on the cutterhead, the TBM's torque and thrust increase significantly, while its advance speed decreases. The TBM operator can determine that the cutterhead is in an abnormal state based on the slow advance speed, but this detection method has a significant lag problem. It is often only discovered when the mud cake has accumulated severely, and it cannot accurately pinpoint the exact location of the mud cake formation. On the other hand, the cutterhead temperature rises rapidly after the mud cake forms. Monitoring the cutterhead temperature can predict the mud cake situation. However, at present, relying solely on temperature data and ignoring multi-dimensional information such as geological data and TBM tunneling parameters, the detection accuracy is insufficient. Furthermore, the temperature threshold largely depends on human experience and lacks an adaptive optimization mechanism. Summary of the Invention

[0004] To address the aforementioned shortcomings in existing technologies, this application provides a method for predicting mud cake formation on the cutterhead of a tunnel boring machine (TBM) based on multi-source monitoring information of the cutting tools. By monitoring the temperature and vibration information of the cutting tools and coordinating with tunneling parameters to form multimodal data, and combining lightweight algorithms and adaptive optimization mechanisms, this method achieves efficient and accurate identification of mud cake formation on the TBM cutterhead.

[0005] To achieve the aforementioned objectives, the technical solution adopted in this application is as follows:

[0006] This application provides a method for predicting mud cake formation on the cutterhead of a tunnel boring machine based on multi-source monitoring information of the cutterhead, including:

[0007] S1: The temperature and vibration frequency of each cutter on the shield cutterhead are collected in real time using temperature and vibration sensors, and the shield tunneling parameters of the shield cutterhead are recorded synchronously to form a multimodal dataset.

[0008] S2: Based on the improved inverse distance weighted interpolation algorithm, the temperature and vibration of each point where the sensor is not installed are predicted according to the temperature and vibration frequency, and a dynamic thermal map of the cutter head is generated.

[0009] S3: The isolated forest algorithm is used to calculate the temperature change rate and vibration frequency change rate at each point. When the temperature change rate and vibration frequency change rate exceed the preset threshold, the cutter has formed mud cake. The machine is stopped for inspection and proceeds to S4. Otherwise, the tunnel boring machine continues to tunnel and the preset threshold is dynamically adjusted using the Bayesian algorithm. Proceed to S1.

[0010] S4: If the results of the shutdown inspection are consistent with the threshold prediction results, then the mud cake is cleaned, the shield cutterhead continues shield tunneling, and the preset threshold is dynamically adjusted using the Bayesian algorithm, then proceed to S1; otherwise, the preset threshold is optimized, then proceed to S1.

[0011] Furthermore, S2 specifically includes:

[0012] S201: Based on the cutter temperature and vibration frequency of the shield cutterhead during operation collected in S1, the improved inverse distance weighted interpolation algorithm is used to calculate the temperature and vibration frequency of the cutterhead at all positions.

[0013] S202: Standardize the calculated temperature and vibration frequency of the cutter head at all positions. The data standardization formula is as follows:

[0014]

[0015] Where u is the mean, σ is the standard deviation, and X is the original sample data. norm For standardized sample data;

[0016] S203: Normalize the standardized sample data and map the normalized temperature-vibration data through the HSL color space;

[0017] S204: Convert the HSL value of each data point to RGB color, map the area of ​​the cutter head to the corresponding color according to the coordinates, and generate a dynamic heat map of the cutter head.

[0018] Furthermore, the improved inverse distance weight interpolation algorithm:

[0019]

[0020] Among them, T i V represents temperature. i Let d be the vibration frequency. i Where is the distance, p is the attenuation coefficient, x and y represent the positions of the cutter head, i is the monitoring point, and n is the total number of monitoring points.

[0021] Furthermore, S3 specifically includes:

[0022] S301: Extract subsamples from the normalized data and construct an isolation tree;

[0023] S302: Based on the constructed isolated trees, form an isolated forest; for each subsample, traverse each tree and record the path length from the root node to the leaf node;

[0024] S303: Calculate the average path length across all trees and standardize it to anomaly score s.

[0025]

[0026] Where a is the sample point, m is the total number of samples, c(.) is the path length correction term, h(.) is the path length from the root node to the leaf node, and E(.) is the average path length.

[0027] S304: Select the data location where the abnormal score s>0.5, and calculate the temperature change rate and vibration frequency change rate of the corresponding cutter head position;

[0028] S305: Compare the temperature change rate and vibration frequency change rate with the preset threshold. If the temperature change rate and vibration frequency change rate of a certain cutter are greater than the temperature threshold and vibration threshold, it is determined that the cutter has formed mud cake. The machine is stopped for inspection and proceeds to S4. Otherwise, the tunnel boring machine continues to tunnel and the preset threshold is dynamically adjusted using a Bayesian algorithm. Proceed to S1.

[0029] Furthermore, the step of dynamically adjusting the preset threshold using a Bayesian algorithm specifically includes:

[0030] A1: Obtain historical data and real-time data collected from the shield cutterhead;

[0031] A2: Construct a Bayesian linear regression model based on the data:

[0032] ΔT=β0+β1·X 地层类型 +β2·X 掘进参数 +β3·X 历史误报率 +ε T

[0033] ΔV=γ0+γ1·X 地层类型 +γ2·X 掘进参数 +γ3·X 历史误报率 +ε V

[0034] Where ΔT and ΔV are the temperature threshold adjustment values ​​and vibration threshold adjustment values, β0 and γ0 are the initial parameter weighting coefficients, β1 and γ1 are the geological type weighting coefficients, β2 and γ2 are the tunneling parameter weighting coefficients, β3 and γ3 are the historical false alarm rate weighting coefficients, and ε T ε V For noise, X 地质类型 X 掘进参数 X 历史误报率 For the type parameters in the shield tunneling process;

[0035] A3: The weighting coefficients are determined by estimating the posterior distribution using Markov chain Monte Carlo estimation.

[0036] A4: Based on the weighting coefficients, tunneling parameters, and Bayesian linear regression model, obtain the current temperature threshold adjustment value and vibration threshold adjustment value;

[0037] A5: Generate the current temperature threshold and vibration threshold based on the base threshold and the threshold adjustment value.

[0038] Furthermore, optimizing the preset threshold includes:

[0039] When the results of the shutdown inspection do not match the results of the threshold judgment, the weighting coefficient of the historical false alarm rate is modified.

[0040] The beneficial effects of this application are as follows: By monitoring multimodal data such as temperature and vibration, the phenomenon of cutterhead mud cake formation is predicted, reducing the bias of single data points and improving detection reliability. Furthermore, the use of lightweight algorithms such as isolated forests reduces the computational burden on temperature and vibration data, improving computational efficiency. Simultaneously, the use of a Bayesian algorithm to dynamically adjust the preset threshold reduces manual intervention, thereby improving the accuracy of cutterhead mud cake identification. In addition, if the system misidentifies or misses a case, it can adaptively adjust the preset threshold, further enhancing the accuracy of cutterhead mud cake identification. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0042] Figure 1 This is a flowchart illustrating a method for predicting mud cake formation on the cutterhead of a tunnel boring machine based on multi-source monitoring information of the cutterhead, as provided in an embodiment of this application.

[0043] Figure 2 This is a schematic diagram of a cutter head information monitoring system provided in an embodiment of this application.

[0044] Figure 3 A dynamic thermal diagram of a cutter head is provided for an embodiment of this application.

[0045] Among them: 1-Shield machine, 2-Cutoff head, 3-Cutoff blade, 4-Temperature sensor, 5-Vibration sensor, 6-Wireless signal receiving module, 7-Shield main control room, 8-Shield host control system. Detailed Implementation

[0046] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0047] The temperature-based method for monitoring cutterhead mud cake relies solely on temperature data, neglecting multi-dimensional information such as geological data and tunneling parameters, resulting in insufficient detection accuracy. To improve the accuracy of cutterhead mud cake detection, it is possible to determine cutterhead mud cake based on the temperature and vibration data of the cutterhead.

[0048] It is understandable that friction increases and temperature rises in a stepwise manner when mud cake forms. Mud cake accumulation leads to uneven stress on the cutterhead, and the vibration spectrum energy concentrates in the low frequency band. There is a time lag between temperature rise and vibration anomaly, and vibration anomaly usually precedes temperature change. By jointly detecting anomalies in temperature and vibration signals, the early characteristics of mud cake formation on the cutterhead can be identified, improving monitoring accuracy, ensuring the normal operation of the tunnel boring machine cutterhead, and reducing operation and maintenance costs.

[0049] Based on this, this application provides a method for predicting mud cake formation on the cutterhead of a tunnel boring machine (TBM) based on multi-source monitoring information of the cutterhead. This method can be found in [reference needed]. Figure 1 , Figure 1 The diagram shown is a flowchart of a method for predicting mud cake formation on a tunnel boring machine cutterhead based on multi-source monitoring information of the cutterhead, as provided in this application embodiment. The method includes:

[0050] S1: The temperature and vibration frequency of each cutter on the shield cutterhead are collected in real time using temperature and vibration sensors, and the shield tunneling parameters of the shield cutterhead are recorded synchronously to form a multimodal dataset.

[0051] In one possible implementation, the cutter head information monitoring system can refer to... Figure 2 , Figure 2The schematic diagram of a cutterhead information monitoring system provided in this application embodiment is as follows: a temperature sensor 4 and a vibration sensor 5 are installed on the inner wall of each cutter 3 of the cutterhead 2. The temperature sensor 4 and the vibration sensor 5 have built-in wireless transmission modules, which can wirelessly transmit real-time monitored temperature and vibration data; a wireless signal receiving module 6 is installed on the inner side of the shield cutterhead 2 to receive temperature and vibration data; the wireless signal receiving module 6 transmits the acquired monitoring data to the host control system 8 of the shield tunnel main control room 7 via wired transmission.

[0052] In one possible implementation, the shield tunneling parameters, including shield thrust, torque, and tunneling speed, are recorded synchronously at the moment of sensor acquisition, forming a multimodal dataset.

[0053] S2: Based on the improved inverse distance weighted interpolation algorithm, the temperature and vibration of each point where the sensor is not installed are predicted according to the temperature and vibration frequency, and a dynamic thermal map of the cutter head is generated.

[0054] Furthermore, S2 specifically includes:

[0055] S201: Based on the cutter temperature and vibration frequency of the shield cutterhead during operation collected in S1, the improved inverse distance weighted interpolation algorithm is used to calculate the temperature and vibration frequency of the cutterhead at all positions.

[0056] S202: Standardize the calculated temperature and vibration frequency of the cutter head at all positions. The data standardization formula is as follows:

[0057]

[0058] Where u is the mean, σ is the standard deviation, and X is the original sample data. norm For standardized sample data;

[0059] S203: Normalize the standardized sample data and map the normalized temperature-vibration data through the HSL color space;

[0060] S204: Convert the HSL value of each data point to RGB color, map the area of ​​the cutter head to the corresponding color according to the coordinates, and generate a dynamic heatmap of the cutter head. The dynamic heatmap is as follows: Figure 3 As shown.

[0061] Among them, the improved inverse distance weight interpolation algorithm in S2 is as follows:

[0062]

[0063] Among them, T i V represents temperature. i Let d be the vibration frequency. iWhere is the distance, p is the attenuation coefficient, x and y represent the positions of the cutter head, i is the monitoring point, and n is the total number of monitoring points.

[0064] In one possible embodiment, an RGB color system is used, and the values ​​of R, G, and B are adjusted according to the temperature. For example, if an rg color scheme is used, the part with higher temperature is redder and the part with lower temperature is greener. The temperature gradually changes from red to yellow and then to green from high to low. Under visual observation, it is possible to determine whether the tool is abnormal based on the dynamic thermal map of the tool head.

[0065] S3: The isolated forest algorithm is used to calculate the temperature change rate and vibration frequency change rate at each point. When the temperature change rate and vibration frequency change rate exceed the preset threshold, the cutter has formed mud cake. The machine is stopped for inspection and proceeds to S4. Otherwise, the tunnel boring machine continues to tunnel and the preset threshold is dynamically adjusted using the Bayesian algorithm. Proceed to S1.

[0066] Furthermore, S3 specifically includes:

[0067] S301: Extract subsamples from the normalized data and construct an isolation tree;

[0068] S302: Based on the constructed isolated trees, form an isolated forest; for each subsample, traverse each tree and record the path length from the root node to the leaf node;

[0069] S303: Calculate the average path length across all trees and standardize it to anomaly score s.

[0070]

[0071] Where a is the sample point, m is the total number of samples, c(.) is the path length correction term, h(.) is the path length from the root node to the leaf node, and E(.) is the average path length.

[0072] S304: Select the data location where the abnormal score s>0.5, and calculate the temperature change rate and vibration frequency change rate of the corresponding cutter head position;

[0073] S305: Compare the temperature change rate and vibration frequency change rate with the preset threshold. If the temperature change rate and vibration frequency change rate of a certain cutter are greater than the temperature threshold and vibration threshold, it is determined that the cutter has formed mud cake. The machine is stopped for inspection and proceeds to S4. Otherwise, the tunnel boring machine continues to tunnel and the preset threshold is dynamically adjusted using a Bayesian algorithm. Proceed to S1.

[0074] In one possible implementation, based on the cutter temperature and vibration frequency of the shield cutterhead during operation collected in step S1, the IDW interpolation method is used to calculate the temperature and vibration frequency of the cutterhead at all positions; (2) the temperature and vibration frequency are standardized and normalized; (3) sub-samples are extracted from the normalized data to construct an isolated tree. Each isolated tree is recursively divided until the termination condition is met. The termination condition is generally: the node has only one sample left, or the height of the tree reaches the preset limit, usually log2(α), where α is the total number of sub-samples. (4) Construct 100 isolated trees to form an isolated forest; for each data point α, traverse each tree and record the path length h(α) from the root node to the leaf node; (5) Calculate the average path length E(h(α)) in all trees and standardize it to anomaly score s, s(α,n)=2-E(h(α)) / c(n), c(n) is the path length correction term; s≈1: highly likely anomaly point, s≤0.5, normal sample; (6) Select the data position s>0.5 and calculate the temperature change rate and vibration frequency change rate at that position, ΔT=(T t -T0) / t, ΔV=(V t -V0) / t, where T t V is the temperature during time interval t, where T0 is the initial temperature. t t is the vibration frequency during the time period, and V0 is the initial vibration frequency; (7) the position of the cutterhead cake is determined at the position where ΔT>ΔT0, ΔV>ΔV0, where ΔT0 and ΔV0 are the temperature threshold and vibration threshold, respectively.

[0075] Furthermore, the step of dynamically adjusting the preset threshold using a Bayesian algorithm specifically includes:

[0076] A1: Obtain historical data and real-time data collected from the shield cutterhead;

[0077] A2: Construct a Bayesian linear regression model based on the data:

[0078] ΔT=β0+β1·X 地层类型 +β2·X 掘进参数 +β3·X 历史误报率 +ε T

[0079] ΔV=γ0+γ1·X 地层类型 +γ2·X 掘进参数 +γ3·X 历史误报率 +ε V

[0080] Where ΔT and ΔV are the temperature threshold adjustment values ​​and vibration threshold adjustment values, β0 and γ0 are the initial parameter weighting coefficients, β1 and γ1 are the geological type weighting coefficients, β2 and γ2 are the tunneling parameter weighting coefficients, β3 and γ3 are the historical false alarm rate weighting coefficients, and ε T ε V For noise, X 地质类型 X 掘进参数 X 历史误报率 These are the tunneling parameters during the shield tunneling process;

[0081] A3: The weighting coefficients are determined by estimating the posterior distribution using Markov chain Monte Carlo estimation.

[0082] A4: Based on the weighting coefficients, tunneling parameters, and Bayesian linear regression model, obtain the current temperature threshold adjustment value and vibration threshold adjustment value;

[0083] A5: Generate the current temperature threshold and vibration threshold based on the base threshold and the threshold adjustment value.

[0084] In one possible implementation, (1) the formation type data is thermally encoded, and the tunneling parameters and historical false alarm rates are annotated; (2) a Bayesian linear regression model is used to obtain the temperature threshold adjustment value ΔT and the vibration threshold adjustment value ΔV.

[0085] ΔT=β0+β1·X 地层类型 +β2·X 掘进参数 +β3·X 历史误报率 +ε T

[0086] ΔV=γ0+γ1·X 地层类型 +γ2·X 掘进参数 +γ3·X 历史误报率 +ε V

[0087] Set the weighting coefficient β i γ i It conforms to a Gaussian distribution;

[0088] (3) Use Markov chain Monte Carlo to estimate the posterior distribution; (4) Extract the mean of each weight coefficient from the posterior distribution, combine it with the tunneling parameters in step S1, substitute it into the regression model to obtain the current temperature threshold and vibration threshold adjustment values, and then superimpose them onto the basic threshold to generate the current temperature threshold and vibration threshold as ΔT0+ΔT and ΔV0+ΔV, respectively.

[0089] S4: If the results of the shutdown inspection are consistent with the threshold prediction results, then the mud cake is cleaned, the shield cutterhead continues shield tunneling, and the preset threshold is dynamically adjusted using the Bayesian algorithm, then proceed to S1; otherwise, the preset threshold is optimized, then proceed to S1.

[0090] Furthermore, optimizing the preset threshold includes:

[0091] When the results of the shutdown inspection do not match the results of the threshold judgment, the weighting coefficient of the historical false alarm rate is modified.

[0092] In one possible implementation, if the shutdown inspection result is consistent with the threshold prediction result, it indicates that the prediction is correct and the cutter has indeed experienced mud cake formation. In this case, the mud cake is cleaned, the shield cutterhead continues shield tunneling, the shield tunneling parameters of the shield cutterhead are recorded, new multimodal data is formed, and the preset threshold is dynamically adjusted using a Bayesian algorithm based on the new multimodal data. Otherwise, it indicates that the prediction is inaccurate, and the preset threshold is optimized.

[0093] In one possible implementation, when the system detects a false alarm, i.e., the threshold judgment result indicates that mud cake phenomenon has occurred, but the shutdown inspection result does not show mud cake phenomenon, it is considered a false alarm, and the historical false alarm rate weighting coefficients β3 and γ3 are reduced to decrease the threshold sensitivity; when the system detects a missed alarm, i.e., the threshold judgment result does not indicate that mud cake phenomenon has occurred, but the shutdown inspection result shows that mud cake phenomenon has occurred, the historical false alarm rate weighting coefficient is increased to increase the threshold sensitivity.

[0094] This application predicts cutterhead cake formation by monitoring multimodal data such as temperature and vibration, reducing the bias of single data points and improving detection reliability. Furthermore, it utilizes lightweight algorithms such as isolated forests to reduce computational calculations related to temperature and vibration data, thus improving computational efficiency. Simultaneously, it employs a Bayesian algorithm to dynamically adjust preset thresholds, reducing manual intervention and improving the accuracy of cutterhead cake formation detection. Moreover, in the event of a misjudgment or missed judgment, the system can adaptively adjust the preset thresholds, further enhancing the accuracy of cutterhead cake formation detection.

[0095] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. The above descriptions are merely preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A method for predicting mud cake formation on the cutterhead of a tunnel boring machine based on multi-source monitoring information of the cutterhead, characterized in that, include: S1: The temperature and vibration frequency of each cutter on the shield cutterhead are collected in real time using temperature and vibration sensors, and the shield tunneling parameters of the shield cutterhead are recorded synchronously to form a multimodal dataset. S2: Based on the improved inverse distance weighted interpolation algorithm, the temperature and vibration of each point where the sensor is not installed are predicted according to the temperature and vibration frequency, and a dynamic thermal map of the cutter head is generated. S3: The isolated forest algorithm is used to calculate the temperature change rate and vibration frequency change rate at each point. When the temperature change rate and vibration frequency change rate exceed the preset threshold, the cutter has formed mud cake. The machine is stopped for inspection and proceeds to S4. Otherwise, the tunnel boring machine continues to tunnel and the preset threshold is dynamically adjusted using the Bayesian algorithm. Proceed to S1. S4: If the results of the shutdown inspection are consistent with the threshold prediction results, then the mud cake is cleaned, the shield cutterhead continues shield tunneling, and the preset threshold is dynamically adjusted using the Bayesian algorithm, then proceed to S1; Otherwise, optimize the preset threshold and proceed to S1; The step of dynamically adjusting the preset threshold using a Bayesian algorithm specifically includes: A1: Obtain historical data and real-time data collected from the shield cutterhead; A2: Construct a Bayesian linear regression model based on the data: in, , These are the temperature threshold adjustment value and the vibration threshold adjustment value. , These are the initial parameter weighting coefficients. , This represents the weighting coefficient for geological types. , These are the weighting coefficients for the tunneling parameters. , This is the weighting coefficient for historical false alarm rates. , For noise items, , , For the type parameters in the shield tunneling process; A3: The weighting coefficients are determined by estimating the posterior distribution using Markov chain Monte Carlo estimation. A4: Based on the weighting coefficients, tunneling parameters, and Bayesian linear regression model, obtain the current temperature threshold adjustment value and vibration threshold adjustment value; A5: Generate the current temperature threshold and vibration threshold based on the base threshold and the threshold adjustment value; The optimization of the preset threshold includes: When the results of the shutdown inspection do not match the results of the threshold judgment, the weighting coefficient of the historical false alarm rate is modified.

2. The method for predicting mud cake formation on the cutterhead of a tunnel boring machine based on multi-source monitoring information of the cutterhead as described in claim 1, characterized in that, S2 specifically includes: S201: Based on the cutter temperature and vibration frequency of the shield cutterhead during operation collected in S1, the improved inverse distance weighted interpolation algorithm is used to calculate the temperature and vibration frequency of the cutterhead at all positions. S202: Standardize the calculated temperature and vibration frequency of the cutter head at all positions. The data standardization formula is as follows: in, The mean, Standard deviation As the original sample data, For standardized sample data; S203: Normalize the standardized sample data and map the normalized temperature-vibration data through the HSL color space; S204: Convert the HSL value of each data point to RGB color, map the area of ​​the cutter head to the corresponding color according to the coordinates, and generate a dynamic heat map of the cutter head.

3. The method for predicting mud cake formation on the shield cutterhead based on multi-source monitoring information of the cutterhead according to claim 2, characterized in that, The improved inverse distance weighted interpolation algorithm: in, For temperature, The vibration frequency, For distance, The attenuation coefficient is... , Represents the position of the cutter head. As a monitoring point, This is the main monitoring point.

4. The method for predicting mud cake formation on the cutterhead of a tunnel boring machine based on multi-source monitoring information of the cutterhead as described in claim 1, characterized in that, S3 specifically includes: S301: Extract subsamples from the normalized data and construct an isolation tree; S302: Based on the constructed isolated trees, form an isolated forest; for each subsample, traverse each tree and record the path length from the root node to the leaf node; S303: Calculate the average path length across all trees and standardize it into anomaly scores. : in, For sample points, The total number of samples, This is a path length correction term. This represents the path length from the root node to the leaf node. This represents the average path length. S304: Select the abnormal score The data location is used to calculate the rate of temperature change and the rate of vibration frequency change corresponding to the cutter head position; S305: Compare the temperature change rate and vibration frequency change rate with the preset threshold. If the temperature change rate and vibration frequency change rate of a certain cutter are greater than the temperature threshold and vibration threshold, it is determined that the cutter has formed mud cake. The machine is stopped for inspection and proceeds to S4. Otherwise, the tunnel boring machine continues to tunnel and the preset threshold is dynamically adjusted using a Bayesian algorithm. Proceed to S1.

Images