JP2019143388A - Shield machine control system and shield machine control method - Google Patents

Abstract

To provide a shield machine control system for controlling a shield machine without deteriorating the safety and quality of construction when executing control of a shield machine using a machine learning model, if there are abnormal values or missing values in the data input to the machine learning model.SOLUTION: The present invention is a system for controlling a shield machine with control data of shield machine output from a machine learning model using measurement value of monitoring item data for monitoring operation status of the shield machine as input data, and comprises a machine learning model unit that outputs the control data based on the monitoring item data for monitoring excavation status input from the shield machine, a rule base data generation unit for deriving the control data based on the measured value and the predetermined rule, a control unit that performs processing to output the control data to the machine learning model unit or the rule base data generation unit or to stop the shield machine based on the measurement value.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a shield excavator control system and a shield excavator control method.

Conventionally, when constructing a tunnel or the like by a shield construction method, the construction environment of the site where the shield excavator excavates (the state of the ground such as soil and water pressure) changes every moment depending on the position of the site. Therefore, it is necessary for the operator to control the traveling speed and traveling direction in excavation of the shield excavator in response to changes in the construction environment.
In other words, if the operator does not appropriately control the shield excavator corresponding to each of the sites having different construction environments, the accuracy and safety of the excavated tunnel design are lowered.

In addition, the operator cultivates the experience of controlling the shield excavator in response to changes in the construction environment and improves the skill level by performing tunnel construction at various sites.
When a skilled operator controls a shield excavator at a site during excavation, he / she performs control corresponding to the current construction environment of the site by applying knowledge of control in a similar construction environment in the past. . However, in the case of an operator with a small number of construction sites, in a construction environment that has never been experienced, it is not possible to control an appropriate shield excavator in the construction environment with poor experience and basic operational knowledge.

For this reason, the accuracy and safety with respect to the design of the tunnel to be excavated vary depending on the skill level of the operator in controlling each shield excavator.
In order to solve this problem, there is a configuration in which the shield excavator is automatically operated based on measurement data indicating the rotation state of the cutter of the shield excavator and the propulsion state of the propulsion jack during excavation (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 07-71189

  In Patent Document 1 described above, the operation of the shield excavator by a skilled operator cannot be sufficiently reproduced. In other words, since control is performed by comparing the measured measurement data with the set value, unlike the control based on the experience of a skilled operator, the control corresponding to the construction environment of the site that changes every moment is appropriately performed. However, the accuracy and safety of the excavated tunnel design cannot be improved.

In addition, using machine learning models that have been previously learned from teacher data, such as human emotion analysis and device operation, the same measurement values as teacher data are input to estimate analysis results and operation settings. Is generally done.
Therefore, it is conceivable to estimate a set value for performing an operation closer to a more skilled operator by using a machine learning model in which the operation of a skilled excavator is learned by machine learning using the operation of a skilled operator as teacher data. .

The machine learning model described above is configured to obtain a set value as an operation amount to be given next to operate the shield excavator based on a plurality of data including measurement values obtained from the shield excavator.
However, when the measured value is an abnormal numerical value or the measured value is missing, the machine learning model cannot obtain a set value for appropriately operating the shield excavator. In addition, since an abnormal set value is required, there is a risk of adversely affecting the safety and quality of construction using a shield excavator.

  The present invention has been made in consideration of the above circumstances, and when a shield excavator is controlled by a machine learning model corresponding to the construction environment of a skilled operator, input from the shield excavator to the machine learning model. It is an object of the present invention to provide a shield excavator control system and a shield excavator control method for controlling a shield excavator without deteriorating the safety and quality of construction when abnormal values or missing values exist in the data to be processed.

  In order to solve the above-described problems, a shield excavator control system according to the present invention controls the shield excavator output from a machine learning model using, as input data, measurement values of monitoring item data for monitoring the operation status of the shield excavator. A shield excavator control system for controlling the shield excavator according to the control data used for the monitoring item data to monitor the excavation state of the shield excavator input from a sensor provided in the shield excavator A machine learning model unit that outputs the control data based on a measurement value, a rule base data generation unit that derives the control data based on the measurement value and a predetermined rule, and the control data based on the measurement value Output to the machine learning model unit, or output to the rule base data generation unit, or the shield excavator And a control unit to perform processing of one of either stopping.

  According to the shield excavator control method of the present invention, the measurement value of the monitoring item data for monitoring the operation status of the shield excavator is used as input data, and the shield excavator is controlled by the control data used for controlling the shield excavator output from a machine learning model. A shield excavator control method for controlling an excavator, wherein the machine learning model unit inputs the monitoring item data for monitoring the excavation state of the shield excavator, which is input from a sensor provided in the shield excavator. A machine learning model process for outputting the control data based on the value, a rule base data generation process for the rule base data generation unit to derive the control data based on the measurement value and a predetermined rule, and a control unit Based on the measurement value, the control data is output to the machine learning model unit or output to the rule base data generation unit. Whether to include a control process to perform the processing of one of either stopping the shield excavator.

  According to the present invention, when control of the shield excavator is executed by a machine learning model corresponding to a construction environment of a skilled operator, there is an abnormal value or a missing value in data input from the shield excavator to the machine learning model. In this case, it is possible to provide a shield excavator control system and a shield excavator control method for controlling the shield excavator without deteriorating the safety and quality of construction.

It is a figure explaining the shield excavator of the object which the shield excavator control system of this embodiment controls. It is a figure which shows the structural example of the shield excavator control system by this embodiment. 4 is a diagram illustrating a configuration example of a monitoring item data table written in a storage unit 20. FIG. It is a figure explaining the relationship between a learning data range and a limit value range. It is a figure of a flowchart which shows an example of control operation of the shield excavator by the shield excavator control system of this embodiment.

Hereinafter, an embodiment of a shield excavator control analysis system according to the present invention will be described with reference to FIG.
FIG. 1 is a diagram for explaining a shield excavator to be controlled by the shield excavator control system of the present embodiment.
Fig.1 (a) has shown the conceptual diagram of the shield excavator of the object which performs the control analysis of the shield excavator control analysis system of this embodiment.
The excavation mechanism 50 is a mechanism for excavating the shield excavator 30 while constructing the primary lining S by assembling the segments with an erector (not shown) at the rear of the cylindrical skin plate 2 in the direction of arrow D1. . In the excavation mechanism 50, the mud clay chamber 7 is provided behind the annular and face plate type cutter 10 having the face 23 in the direction of arrow D <b> 1. The mud clay chamber 7 is filled with a mud clay material injection pipe 8 from a mud clay material injection pipe 8 and converts the excavated soil into mud by powerful mixing with a mixing blade (not shown). The face 23 is stabilized by balancing with and excavation is performed. Here, the excavation residual soil accumulated in the mud clay chamber 7 of the excavation mechanism 50 is introduced into the screw conveyor 60 and is discharged to the outside of the excavating tunnel via the conveyors 62 and 63. The gantry M supports the screw conveyor 60 and the conveyors 62 and 63, respectively. Although not shown, the excavation mechanism 50 is provided with a propulsion jack (described later), and the excavation direction and the propulsion speed are controlled by the propulsion jack.

  In the present embodiment, the propulsion direction and propulsion speed of the skin plate 2 are controlled by hydraulic control of the propulsion jack (hereinafter, also simply referred to as direction control), and the rotation speed of the screw of the screw conveyor 60 (described later). The mud pressure is controlled by controlling the screw speed (hereinafter referred to simply as earth pressure control). A shield excavator control system, which will be described later, acquires each of the monitoring item data indicating the state of these controlled objects, and estimates control data for controlling the shield excavator using a machine learning model (described later).

  In this embodiment, the propulsion direction and the propulsion speed of the skin plate 2 are controlled (direction control) by hydraulic control as propulsion jack control data, and the screw rotation speed (screw rotation) is controlled as control data for the screw conveyor 60. The control data for controlling the mud pressure (soil pressure control) is acquired by controlling the speed). In the present embodiment, the monitoring item data includes, for example, cutter torque, cutter speed, propulsion pressure, propulsion as data (monitoring item data) for monitoring the construction environment being excavated and the operating state of the shield excavator 30. Speed, propulsion speed instruction deviation value, control earth pressure, face earth pressure average value instruction outside dummy, agitator torque, screw speed, primary screw pressure, secondary screw pressure, NO. 1 copy stroke, NO. 1 Copy stroke command value difference, NO. 1 copy position, NO. 1 copy position instruction deviation value, pitching, pitching instruction value difference, rolling, rolling instruction value difference, vertical folding angle, vertical folding angle instruction value difference, left and right folding angle, left and right folding angle instruction value difference, S / M front trunk direction, S / M front trunk direction instruction value difference, S / M rear trunk direction instruction value, S / M rear trunk direction instruction value difference, planned route horizontal deviation (control point), planned route vertical deviation (control point), There are azimuth (management point), azimuth indication value difference (management point), planned route azimuth (management point), pitch (management point), planned route pitch (management point), and the like.

  FIG. 1B shows a conceptual diagram illustrating a propulsion jack that propels the excavation mechanism 50. As shown in FIG. 1B, a force point for propelling the surface of the skin plate 2 is set depending on which position of the propulsion jack is driven. In FIG. 1 (b), 22 propulsion jacks are shown, but this number is not limited. By distributing the jack pressure of each propulsion jack, the propulsion jack force point position Fx on the x axis and the propulsion jack force point position Fy on the y axis are set, and the mark P becomes the position of the force point. When the propulsion pressure is applied to a position corresponding to the force point on the surface of the skin plate 2, the direction in which the excavation mechanism 50 propels is set. The control in this direction is performed by a propulsion jack pattern indicating which of the propulsion jacks is driven. The propulsion speed is set by the amount of oil (oil amount) supplied per unit time to increase the jack pressure.

  FIG. 2 is a diagram illustrating a configuration example of the shield excavator control system according to the present embodiment. In FIG. 2, the shield excavator control system 1 includes a monitoring item data input unit 11, a data determination unit 12, a correction complement unit 13, a data range determination unit 14, an abnormal stop determination unit 15, a control unit 16, and a machine learning model unit 17. , A rule base data generation unit 18, an output check unit 19, and a storage unit 20.

  The monitoring item data input unit 11 receives the data of each of the monitoring items described above at a timing when a timing signal indicating the passage of a predetermined period of time (for example, 1 second) is supplied from a timing unit (not shown). Is used as a measurement value, and the monitoring item data is acquired from each of the detection means (sensor, measuring instrument, etc.) provided in each part. Further, the monitoring item data input unit 11 sequentially writes and stores the measured values of the acquired monitoring item data in the monitoring item data table in the storage unit 20. At this time, the monitoring item data input unit 11 stores a plurality of measurement values for a predetermined period for each type of monitoring item data, and overwrites the oldest measurement value when writing a new measurement value. To remember.

  Also, the monitoring item data input unit 11 supplies the machine learning model unit 17 and the rule base data generation unit 18 with the measurement values normalized (normalized) or averaged by the monitoring item data. If necessary, perform standardization and averaging calculations. For example, in the case of a measurement value such as a control earth pressure in monitoring item data, the averaging process may be supplied as an abnormally high or abnormally low numerical value due to noise superposition or the like. As a result, repeated abnormal stops due to noise affect the progress of construction by the shield excavator. Therefore, the monitoring item data input unit 11 obtains a measurement value and a moving average in a predetermined period in the past for the measurement value of the monitoring item data set in advance, and uses this moving average as the measurement value of the monitoring item data. Used as

  FIG. 3 is a diagram illustrating a configuration example of the monitoring item data table written in the storage unit 20. For each record, columns of monitoring item data type, acquired data, missing period, corresponding sensor measurement range, learning data range, and limit value range are provided. The monitoring item data type indicates the type of monitoring item data. The acquired data is the latest measured value of the acquired monitoring item data. The missing period is an elapsed time after the measured value to be measured is supplied as the missing value. The corresponding sensor measurement range indicates each of the upper limit value and the lower limit value of the measurement range of the sensor that acquires the monitoring item data. The learning data range is a range of measurement values of monitoring item data (range consisting of an upper limit value X2 and a lower limit value X1 of the measurement value) used as teacher data for learning a machine learning model (described later) in the machine learning model unit 17. Show. The limit value range is set for each type of monitoring item data, and is a limit value range (range consisting of an upper limit value Y2 and a lower limit value Y1 of the limit value) indicating danger as a measurement value.

FIG. 4 is a diagram for explaining the relationship between the learning data range and the limit value range.
The learning data range of the monitoring item data is a range P1 that is not more than the upper limit value X2 of the management value and not less than the lower limit value X1. The allowable range of the monitoring item data is a range P22 that exceeds the upper limit value X2 of the management value and is not more than the upper limit value Y2 of the limit value, a range that is less than the lower limit value X1 of the management value, and is not less than the lower limit value Y11. P11. The abnormal value is a value exceeding the upper limit value Y2 of the limit value and a value less than the lower limit value Y1 of the limit value.

  Returning to FIG. 2, the data determination unit 12 determines whether the newly measured measurement value is missing (for example, “0” data) or abnormal. At this time, the data determination unit 12 refers to the monitoring item data table in the storage unit 20 and determines that it is abnormal when it exceeds the corresponding sensor measurement range. Then, the data determination unit 12 attaches additional information such as missing or abnormal to the measurement value of the monitoring item data determined to be missing or abnormal, and outputs it to the correction complementing unit 13.

  When the measurement value of the monitoring item data is supplied from the data determining unit 12, the correction complementing unit 13 determines whether the additional information is missing or abnormal. Here, when the additional information is missing, the correction complementing unit 13 refers to the monitoring item data table in the storage unit 20 and reads the history of the measurement values of the corresponding monitoring item data (measurement values acquired in a predetermined period). Then, the measurement values within a predetermined period are averaged to generate a complementary measurement value, and the complemented measurement value is written and stored in the missing measurement value portion. At this time, the correction complementing unit 13 counts the period during which complementation continues from the time when the measurement value is first complemented, and writes the count value in the missing period part of the monitoring item data table in the storage unit 20. Remember. In the present embodiment, when a measurement value is not obtained due to loss, it is indicated as complementing that the measurement value created in the past history is used as a missing measurement value, and the abnormal measurement value is within the normal range. Correcting is referred to as correction.

  Further, the correction complementing unit 13 refers to the monitoring item data table in the storage unit 20 when the additional information is abnormal, and the corresponding sensor measurement of the sensor that measures the measurement value of the corresponding monitoring item data. Based on the range, the measurement value that is considered abnormal is corrected. At this time, when the measured value exceeds the upper limit value of the corresponding sensor measurement range, the correction complement unit 13 replaces the measured value with this upper limit value, while the measured value is less than the lower limit value of the corresponding sensor measurement range. The measured value is replaced with this lower limit value. Then, the correction complementing unit 13 stores the corrected measurement value in the monitoring item data table in the storage unit 20 by overwriting the original measurement value. In addition, the correction complementing unit 13 counts a period during which the correction continues from the time when the measurement value is first corrected, and writes the count value in the abnormal period portion of the monitoring item data table in the storage unit 20. Remember.

The data range determination unit 14 determines the maximum value and the minimum value of the measurement values in the teacher data used when each measurement value of the monitoring item data (including the measurement values that have been supplemented and corrected) is used to learn the machine learning model. It is determined whether or not it is included in a learning data range (which is greater than or equal to the minimum value of the measured value of teacher data (lower limit value of management value) and less than or equal to the maximum value (upper limit value of management value)). At this time, the data range determination unit 14 determines whether the measurement value of each monitoring item data is included in the learning data range or the allowable range in the determination result. Further, when the measured value exceeds the upper limit value of the limit value range or when the measured value is less than the lower limit value of the limit value range, the data range determination unit 14 sends a shield excavator (shield excavator) to the abnormal stop determination unit 15. 30 and so on), an abnormal stop signal for abnormally stopping is transmitted. The learning data range described above is actually measured with the numerical value obtained by subtracting the first set margin value from the minimum value of the measured value as the lower limit value and the numerical value obtained by adding the second set margin value to the maximum value of the measured value as the upper limit value. It is good also as the range comprised with the allowance with respect to the range of the measured value of the upper limit and lower limit which were made. The first set margin value and the second set margin value are positive values, and either the same numerical value or different numerical values may be used. Each of the first set margin value and the second set margin value may be generated by multiplying an upper limit value and a lower limit value of the measured value by a predetermined coefficient. However, when each of the upper limit value and the lower limit value is a negative numerical value, it is converted into an absolute value and the learning data range is configured as a positive numerical value by the above-described processing, or is added in the case of the lower limit value. In this case, a process for constructing the learning data range is performed by the subtracted process.
In addition, when the measurement value of the monitoring item data that is neither missing nor abnormal is supplied, the data determination unit 12 resets the count value of the missing period and the abnormal period of the monitoring item data table in the storage unit 20 to “0”. To do.

  The abnormal stop determination unit 15 refers to the monitoring item data table in the storage unit 20 for each period for estimating the set value of the control data, and each of the missing period and the abnormal period of the monitoring item data is set in advance. It is determined whether or not the set limit period has been exceeded. At this time, the abnormal stop determination unit 15 performs the abnormal stop of the shield excavator when either the missing period or the abnormal period of the monitoring item data exceeds the respective limit period. Here, when the deficit period exceeds the limit period, there may be a case where a sensor provided in the shield excavator or a sensor provided around the shield excavator has failed. And, if the excavation state corresponding to the excavation including the operation state of the shield excavator and the surrounding environmental condition of excavation is unknown and the operation of the shield excavator is continued, the safety and quality of construction may be reduced. Therefore, the shield excavator control system 1 in the present embodiment abnormally stops the shield excavator when the measurement value continues to be deficient.

Also, when the abnormal period exceeds the limit period, when the sensor is broken, the shield excavator is in an abnormal operating state (low propulsion speed is continuing), or around the excavating area It is conceivable that the environmental state is abnormal (a state in which a high control earth pressure continues).
Continuing the operation of the shield excavator while the sensor is malfunctioning or the shield excavator is operating abnormally and the surrounding environmental conditions are excavating may reduce the safety and quality of construction. Therefore, the shield excavator control system 1 in the present embodiment abnormally stops the shield excavator when the abnormality of the measurement value continues. The abnormal stop determination unit 15 also abnormally stops the shield excavator even when an abnormal stop signal is supplied from the data range determination unit 14.

  The control unit 16 estimates the set value of the control data of the operation of the shield excavator based on the determination result of whether or not the measurement value of the monitoring item data from the data range determination unit 14 is included in the learning data range. Control of which of the machine learning model unit 17 and the rule base data generation unit 18 is performed is performed. At this time, when the measurement result of all the monitoring item data is included in the learning data range, the control unit 16 estimates the set value of the control data for the operation of the shield excavator. 17 is performed. On the other hand, if the measurement value of any of the monitoring item data is outside the learning data range and included in the allowable range, the control unit 16 rules the estimation of the set value of the control data for the operation of the shield excavator The base data generation unit 18 is caused to perform this.

  The machine learning model unit 17 generates, using teacher data, a machine learning model for estimating a set value of control data for operating the shield excavator according to an input measurement value using a machine learning algorithm. The set value of the control data of the operation of the shield excavator corresponding to the measurement value is estimated using the generated machine learning model. For example, in the machine learning model according to the present embodiment, set values of monitoring item data (feature vector) are input as input data to a machine learning algorithm, and control data (propulsion jack hydraulic control data, screw rotation of the screw conveyor 60) Estimate and output the setting value (speed, etc.).

  For this reason, the machine learning model unit 17 first inputs the measured value of the monitoring item data, which is a feature quantity vector for which the correct control data is known, to the machine learning algorithm as teacher data, and estimates the set value of the control data. Generate a machine learning model. And the machine learning model part 17 inputs the measured value of the monitoring item data supplied from a sensor etc. with respect to the learned machine learning model, and sets the control data which is estimated to match the inputted measured value Estimate the value.

Further, the machine learning model unit 17 reconstructs the machine learning model by including the feedback in the teacher data when correct / incorrect feedback is obtained with respect to the set value of the control data estimated using the generated machine learning model. I do. Then, after the reconstruction, the machine learning model unit 17 uses the reconstructed machine learning model to estimate the set value of the control data that matches the measurement value of the input monitoring item data.
Here, the machine learning creates a model in which the measured value of the monitoring item data is an explanatory variable and the set value of the control data is an explanatory variable. In other words, a model between input explanatory variables and output explained variables is created by machine learning, and by using this model, highly relevant explained variables corresponding to combinations of explanatory variables. To be obtained. By using this model, it is possible to clearly obtain which control data that is the explained variable needs to be controlled in response to the change in the monitoring item data that is the explaining variable. Here, as a machine learning technique for creating a model, any of commonly used techniques such as decision tree learning, neural network, genetic programming, support vector machine, and deep learning may be used.

The rule base data generation unit 18 has, for example, the function of an expert system, stores a set of production rules including a condition part and a conclusion part, and measures the monitoring item data supplied and the production of the condition part. Determine the conformity with the rule.
Then, the rule base data generation unit 18 uses the contents described in the conclusion part corresponding to the condition part having the highest conformance as an inference conclusion, that is, the set value of the control data indicated in the conclusion part is an estimated value. Output as.

The output check unit 19 is a set value range in which each set value of control data as an estimated value supplied from each of the machine learning model unit 17 and the rule base data generation unit 18 is set in advance for each control data. It is determined whether it is included.
At this time, if the set value of the control data exceeds the upper limit value of the set range, the output check unit 19 corrects the set value to the upper limit value and outputs the corrected value to the shield excavator. Moreover, when the set value of the control data is less than the lower limit value of the set range, the output check unit 19 corrects this set value to the lower limit value and outputs the corrected value to the shield excavator.

FIG. 5 is a diagram illustrating an example of a flowchart of the control operation of the shield excavator by the shield excavator control system of the present embodiment.
Step S1: The monitoring item data input unit 11 inputs the measured values of the monitoring item data from each of the sensors that measure the measured values of the monitoring item data provided in the shield excavator, and stores the measured values in the storage unit The data is written and stored in the 20 monitoring item data tables. Then, the monitoring item data input unit 11 stores the measurement value of the monitoring item data that is set to have the moving average as the measurement value in advance, that is, the past measurement value in the predetermined range in which the moving average is calculated, as the storage unit 20. Are read from the monitoring item data table, and the moving average of the measured values is calculated. Then, the monitoring item data input unit 11 uses the obtained moving average calculation result as a measurement value of the monitoring item data, and writes and stores the measurement value in the monitoring item data table of the storage unit 20.

  Step S2: The data determination unit 12 refers to the monitoring item data table in the storage unit 20, extracts the latest measurement value, and determines whether this measurement value is abnormal or missing. At this time, if the measurement value is abnormal or missing, the data determination unit 12 proceeds to step S3. If the measurement value is not abnormal or missing, the process proceeds to step S4.

  Step S3: The correction complementing unit 13 confirms the additional information added to the measurement value of the monitoring item data supplied from the data determining unit 12. Then, when the measured value is indicated as abnormal in the additional information, the correction complementing unit 13 corrects the measured value using the upper limit value or the lower limit value of the corresponding sensor measurement range in accordance with the numerical value of the measured value. Do. The correction complementing unit 13 sets the corrected measurement value as a new measurement value in the monitoring item data table of the storage unit 20 and overwrites the measurement value before correction. At this time, the correction complementing unit 13 performs a process of incrementing the abnormal period of the corrected monitoring item data in the monitoring item data table of the storage unit 20.

  In addition, when the measurement value is indicated as missing in the additional information, the correction complementing unit 13 refers to the monitoring item data table of the storage unit 20 and averages (or intermediate value) of past measurement values in a predetermined range. Etc., and the missing measurement values are complemented. The correction complementing unit 13 sets the supplemented measurement value as a new measurement value in the monitoring item data table of the storage unit 20, and writes and stores it in the location of the missing measurement value before the complement. At this time, the correction complementing unit 13 performs a process of incrementing the missing period of the supplemented monitoring item data in the monitoring item data table of the storage unit 20.

  Step S4: The data range determination unit 14 refers to the measurement value of each monitoring item data in the monitoring item data table of the storage unit 20, and determines whether or not the limit value range is exceeded. At this time, if the measured value exceeds the limit value range, the data range determination unit 14 transmits an abnormal stop signal to the abnormal stop determination unit 15 and proceeds to step S8, while the measured value is within the limit value range. If included, the process proceeds to step S5.

  Step S5: The abnormal stop determination unit 15 refers to the monitoring item data table in the storage unit 20, and determines whether there is monitoring item data having a missing period or an abnormal period exceeding the limit period. Then, when there is monitoring item data having a missing period or an abnormal period that exceeds the limit period, the abnormal stop determination unit 15 advances the process to step S8. On the other hand, if there is no monitoring item data, the process advances to step S6. Proceed.

  Step S6: The data range determination unit 14 determines whether or not all the measurement values of the monitoring item data are included in the respective learning data ranges. At this time, if all the measurement values are included in the learning data range, the data range determination unit 14 proceeds with the process to step S7. On the other hand, if any measurement value is not included in the learning data range, The process proceeds to step S9.

  Step S7: When it is determined that all measurement values are included in the learning data range, the control unit 16 causes the machine learning model unit 17 to estimate the set value of the control data. Thereby, the machine learning model unit 17 refers to the monitoring item data table of the storage unit 20, reads the latest measurement value, and outputs the estimated value that is input to the machine learning model and output as the set value of the control data. The data is output to the check unit 19, and the process proceeds to step S1.

  Step S8: When the abnormal stop signal is supplied from the data range determination unit 14 or when there is monitoring item data having a missing period or an abnormal period exceeding the limit period, the abnormal stop determination unit 15 forcibly shields the excavation. Stop the machine abnormally.

  Step S9: The control unit 16 determines that the measurement values in all the monitoring item data are not included in the learning data range corresponding to the monitoring item data, and the measurement values not included in the learning data range are limited. When it is not included in the value range, that is, when it is determined that the measured value not included in the learning data range of the monitoring item data is within the allowable range (exceeding the learning data range and less than the limit value) Then, the rule base data generation unit 18 is caused to estimate the set value of the control data. Thereby, the rule base data generation unit 18 refers to the monitoring item data table of the storage unit 20, reads the latest measurement value, inputs the estimated value output from the conclusion unit to the condition unit, and sets the control data The value is output to the output check unit 19 and the process proceeds to step S1.

  With the configuration described above, according to the present embodiment, the measurement values of all the monitoring item data supplied from the sensors provided in the shield excavator are included in the learning data range, or the learning data range and the limit value range Whether the control data set value is estimated by a machine learning model or rule base, or the shield excavator is abnormally stopped depending on whether it is included in the allowable range between Therefore, if there is an abnormal value or missing value in the measured value, the shield excavator can be controlled without deteriorating the safety and quality of the construction.

  In addition, according to the present embodiment, when the measurement value of the monitoring item data does not exceed the limit value range, each of the missing period and the abnormal period is provided even if there is a missing measurement value or an abnormal measurement value. Therefore, even if a measurement value such as noise is supplied from the sensor, the measurement value is observed during a predetermined period. Therefore, the shield excavator is not stopped abnormally (emergency stop) transiently, and shield excavation is performed. The operation of the machine can be continued and the progress of the construction will not be affected.

  In addition, according to the present embodiment, when the measurement value of the monitoring item data is likely to cause noise, the moving average of a predetermined period is calculated and used as the measured value. Even if it exceeds momentarily, that is, even if a superimposed measurement value such as noise is supplied from the sensor, an abnormality in the measurement value is observed not only at that time but also in a predetermined period. The shield excavator is not abnormally stopped in a transient state, and the shield excavator can be operated continuously without affecting the progress of the construction.

  The program for realizing the function of the shield excavator control system of FIG. 1 in the present invention is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read by the computer system and executed. Thus, the shield excavator may be controlled. Note that the “computer system” herein includes an OS (Operating System) and hardware such as peripheral devices. The “computer system” includes a WWW (World Wide Web) system having a homepage providing environment (or display environment). The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM (Read Only Memory), a CD-ROM (Compact Disc-Read Only Memory), or a computer system. A storage device such as a hard disk. Further, the “computer-readable recording medium” refers to a volatile memory (RAM (Random Access) in a computer system that becomes a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. Memory)) as well as those that hold programs for a certain period of time.

  The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

DESCRIPTION OF SYMBOLS 1 ... Shield excavator control system 11 ... Monitoring item data input part 12 ... Data determination part 13 ... Correction complement part 14 ... Data range determination part 15 ... Abnormal stop determination part 16 ... Control part 17 ... Machine learning model part 18 ... Rule base Data generation unit 19 ... output check unit 20 ... storage unit

Claims (6)

Shield excavator control for controlling the shield excavator based on control data used for controlling the shield excavator output from a machine learning model using the measurement value of monitoring item data for monitoring the operation status of the shield excavator as input data A system,
A machine learning model unit that outputs the control data based on the measured value of the monitoring item data for monitoring the excavation state of the shield excavator, which is input from a detector provided in the shield excavator;
A rule base data generation unit for deriving the control data based on the measurement value and a predetermined rule;
Based on the measured value, a control unit that causes the machine learning model unit to output the rule-based data generation unit, or to stop the shield excavator A shield excavator control system comprising: A data determination unit for determining whether the measurement value of the monitoring item data is missing or an abnormal value;
A data range determination unit that determines whether or not the measurement value of the monitoring item data is within the data range of the teacher data used when learning the machine learning model;
When the measurement value of the monitoring item data is missing or abnormal value, the control unit stops the control of the shield excavator, and when the measurement value of the monitoring item data is out of the data range, The control data is generated by a rule base processing unit, and when the measured value of the monitoring item data is within the data range, the control data is generated by the machine learning model unit. Item 2. A shield excavator control system according to item 1. Complement when the measurement value of the monitoring item data is missing, further comprising a complementary correction unit for correcting when the measurement value is an abnormal value,
The shield excavator control system according to claim 2, wherein the data range determination unit determines whether or not the measured value of the monitored item data that has been supplemented and corrected is within the data range. An abnormality determination unit that determines whether or not the missing or abnormal value of the measurement value of the monitoring item data continues for a predetermined period;
The control unit is
If the abnormal value of the monitoring item data is outside the data range and within a predetermined allowable value, the monitoring item data is supplied to the rule base data generation unit,
4. The shield according to claim 2, wherein the shield excavator is stopped when the deficit continues for a predetermined period or when the defect value is an abnormal value that continuously exceeds an allowable value for a predetermined period. Excavator control system. The monitoring item data input unit performs a moving average of measured values of a predetermined type of monitoring item data in the monitoring item data within a predetermined range, and uses the result of the moving average as a measured value. The shield excavator control system according to claim 4. Shield excavator control for controlling the shield excavator based on control data used for controlling the shield excavator output from a machine learning model using the measurement value of monitoring item data for monitoring the operation status of the shield excavator as input data A method,
A machine learning model process in which a machine learning model unit outputs the control data based on the measured value of the monitoring item data for monitoring the excavation state of the shield excavator, which is input from a sensor provided in the shield excavator; ,
A rule base data generation process in which a rule base data generation unit derives the control data based on the measurement value and a predetermined rule;
Based on the measurement value, the control unit performs either processing of outputting the control data to the machine learning model unit, outputting the control data to the rule base data generation unit, or stopping the shield excavator. A control method for shield excavator characterized by comprising:

Images