JP2002242435A - Concrete compaction determining method and device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To know the timing optimal for stopping a vibrator used when compacting concrete without a determination of a person. SOLUTION: This device includes a process of prestoring data changing with the lapse of time of vibration up to finishing compaction from starting the compaction of the concrete as result data, a process of determining a coincidence rate of both by a neural network by successively comparing a change with the lapse of time of the detecting vibration with the result data while detecting the change with the lapse of time of the vibration by the vibrator from starting the compaction of the concrete in actual compaction work, and a process of determining it as the finish of the compaction when the coincidence rate is not less than a prescribed threshold value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique capable of stopping a vibrator at an appropriate timing when compacting cast concrete with a vibrator.

[0002]

2. Description of the Related Art Conventionally, when a large amount of concrete is cast, for example, concrete is transported by a transporting means such as a bucket by an amount that can be transported to a place where the concrete is to be poured, and is sequentially driven. ing. At this time, the concrete immediately after casting has a large amount of voids because the coarse and fine aggregates are not evenly mixed, and dense concrete cannot be obtained. Giving is done. By imparting vibration to the concrete, the mortar portion composed of fine aggregate, cement and water enters the gaps between the coarse aggregates and becomes uniform, so that dense concrete can be obtained.

However, inconvenience occurs even if the vibrator is applied for too long. That is, if the time for applying the vibrator is too long, there is a problem that breathing water is generated on the surface of the concrete. As described above, the coarse aggregate is settled by the vibrator for a long time, and breathing water containing extremely fine sand and cement floats on the surface, and latencies are generated. The layance generated in this way does not generate strength even when it is hardened, so if you want to add more concrete to the concrete that has been cast and hardened, the concrete before and after In order to improve the connection with the above, it has been performed to remove the latency in advance.

[0004] As described above, if the vibrator is applied for too long, latencies are generated, so that a great deal of labor is required for the work of removing the vibrator, and the amount of removed concrete is wasted, resulting in new waste. Extra concrete is needed for that. On the other hand, if the time for applying the vibrator is insufficient, a sufficiently dense concrete cannot be obtained, and the strength is insufficient. Therefore, the judgment of this compaction termination is based on the quality control of concrete.
In addition, although it is very important from the aspect of workability improvement, it has conventionally relied on expert experience and intuition.

[0005]

For this reason, it is necessary to always have skilled workers at the concrete pouring site. However, in the construction industry in recent years, this kind of skilled workers is in short supply. There were many situations where shallow people had to judge. As a result, there were many mistakes made by inexperienced judges. Furthermore, it is difficult for even an expert to judge the completion of compaction completely and correctly, and a judgment error occurs to some extent, which adversely affects the quality of the concrete and the workability of the casting operation.

In view of the above circumstances, the present invention has been made to provide a technique capable of knowing an optimal timing for stopping a vibrator without human judgment.

[0007]

According to a first aspect of the present invention, there is provided a concrete compaction judging method for judging completion of compaction when compacted concrete is compacted with a vibrator. In the process of accumulating the time-dependent change data of vibration from the start to the end of compaction as actual data, and in the actual compaction work, while detecting the time-dependent change of the vibration by the vibrator from the start of compaction of concrete, The method is characterized in that the method includes a step of successively comparing a change with time with the actual data to obtain a matching rate between the two, and a step of determining that compaction is completed when the matching rate is equal to or more than a predetermined threshold value.

According to a second aspect of the present invention, there is provided a concrete compaction judging method, comprising the steps of: constructing a neural network with input values as time-dependent change data of vibration by a vibrator and using output values as matching rates; In the process of completing the neural network and the actual compaction work, the concrete compaction work starts from the compaction start of the concrete by setting the coupling load coefficient using the time-dependent change data of the vibration by the vibrator as the teacher data with the maximum matching rate. A step of detecting a temporal change of the vibration by the vibrator and sequentially comparing the temporal change of the detected vibration with the teacher data using the neural network to obtain a matching rate between the two; A step of determining that compaction has been completed in the above cases. That. The concrete compaction judging method according to claim 3, further comprising: preparing time-dependent change data of the vibration by the vibrator when the compaction of the concrete is not completed; The method is characterized by including a step of completing a neural network by setting coefficients. The concrete compaction determination method according to claim 4 is further characterized in that a change in vibration caused by the vibrator is detected as a change in sound pressure.

According to a fifth aspect of the present invention, there is provided a concrete compaction judging device, comprising: output means for outputting time-dependent change data of vibration by a vibrator for compacting the poured concrete; and the predetermined time-dependent data using predetermined teacher data and a neural network. Determining means for comparing and calculating the matching rate, and comparing the calculated matching rate with a predetermined threshold value, and determining the compaction end when the matching rate is equal to or more than the threshold value. It is characterized by having.

The concrete compaction judging device according to claim 6 further includes a learning means for setting a coupling load coefficient of the neural network of the finding means based on the time-dependent change data of the vibration from the start to the end of the compaction as a reference. Have. The concrete compaction judging device according to claim 7 further includes a standby unit that outputs a vibrator stop command after a predetermined standby time elapses after the determination unit determines that compaction is completed. In the concrete compaction judging device according to the eighth aspect, the output means is configured to detect a change in vibration caused by the vibrator with a microphone and output data of a change with time of the signal.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A concrete compaction judging method and apparatus according to the present invention will be described below in detail with reference to the drawings showing an embodiment thereof.

In FIG. 1, reference numeral 10 denotes a concrete compaction device used in the present invention. Reference numeral 1 denotes a compaction work vehicle having a vibrator 2 at the end of a boom 11. The vibrator 2 vibrates the vibrating body 21 using an electric or hydraulic motor as a power source.
By operating the boom 11, it can be inserted into a desired place to give vibration. Reference numeral 3 denotes a microphone for detecting a sound caused by the vibration of the vibrator 2, which is installed in front of the operator's seat and facing the vibrator 2. In addition, it is attached via a buffer so that the vibration of the compaction work vehicle 1 is not directly transmitted. Reference numeral 4 denotes a control device installed at the rear of the driver's seat. As shown in FIG. 2, a noise meter 41 for outputting a sound pressure level due to the vibration of the vibrator 2 detected by the microphone 3, and a noise meter 41 An A / D converter 42 for converting the output sound pressure level signal into a digital signal; a judgment computer 43 for judging the timing to stop the vibrator from a change in sound pressure; And a buzzer 5 that sounds when it is pressed. In the above configuration, the microphone 3
And the control device 4 constitute a concrete compaction judging device of the present invention, the microphone 3 constitutes a part of output means, and the control device 4 constitutes learning means, finding means, judging means and standby means. ing.

In FIG. 1, A is concrete that has already been cast and hardened, and B is unhardened concrete that has been additionally cast by a bucket or the like (not shown). C is a temporary frame for holding concrete to be added. The concrete B has almost no slump, for example, and is slightly raised immediately after the casting as shown in the figure. The vibrator 2 is inserted into the concrete B in such a state, and vibration is generated to start compaction work. From this point, the microphone 3 and the control device 4 are operated. The control device 4 shown in FIG. 2 also includes a learning computer 44 described later. The learning computer 44 and the determination computer 43 provided in the control device 4 are programmed with a learning function using a neural network and a compaction termination determination function, as described later. The neural network programmed in the learning computer 44 and the determination computer 43 constitutes a hierarchical network including an input layer, an intermediate layer, and an output layer, as shown in FIG.

Next, the detailed configuration of the learning computer 44 of FIG. 2 will be described. The learning computer 44 as a learning unit stores a sound pressure level data based on the sound pressure level detected by the microphone 3.
1, a creating unit 442 for creating teacher data from the accumulated sound pressure level data, a second storage unit 443 for sequentially storing the created teacher data, and sequentially inputting the teacher data to obtain a connection weight coefficient of the neural network. Learning unit 44
4 is provided. The sound pressure level refers to the amplitude of a sound wave.

The detailed configuration of the computer 43 for determination will be described. The determination computer 43 serving as a determination unit, a determination unit, and a standby unit includes a third storage unit 431 that stores sound pressure level data based on the sound pressure level detected by the microphone 3, and a stored sound pressure level data. A creation unit 432 for creating input data from the input data, a finding unit 433 for sequentially inputting the created input data and finding a matching rate between the performance data and the execution data by a neural network, and a determined matching rate set in advance. And a determination unit 434 that determines the end of compaction by comparing with a threshold value. The determining unit 433 corresponds to a determining unit, and the determining unit 434 corresponds to a determining unit.

The coupling weight coefficient between the layers of the hierarchical neural network obtained by learning in the learning section 444 of the learning computer 44 is a medium 45 such as a LAN.
Is sent to the computer 43 for determination through the computer, and is set as a coupling weight coefficient between the layers of the neural network of the neural network of the determiner 433 so that the determiner 433 of the computer 43 for determination makes an appropriate determination.

Therefore, the judgment computer 43 calculates the matching rate by processing the temporal change of the sound pressure level sequentially inputted by the compaction judging function by the neural network. If it is equal to or more than a threshold value (for example, 0.99), it is determined that compaction is completed, and after a predetermined delay time (for example, 4 seconds),
The buzzer 5 is sounded to notify that compaction has been completed.

Next, an example of creating teacher data will be described. First, in a plurality of cases at a concrete placing site, a time-dependent change in the sound pressure level from the start of compaction to the end of compaction when an expert determines the completion of compaction is measured. FIG. 4 (a) shows an actually measured broken line of the change with time and its envelope.
(B), (c) and (d) show. Observation of the envelope in FIG. 3 shows that there is a certain regularity common to each case.

Specifically, the sound pressure level (absolute value) detected by the microphone 3 immediately after the start of the compaction operation
Is a low level of about 0.1 Pa, as shown in (a), (b), (c), and (d) of FIG. Thereafter, as the compaction operation proceeds, the sound pressure level rises (S1) and rises to a high level peak. Then, when the compaction work further proceeds, the sound pressure level decreases (S2), and thereafter, the level slightly increases (S3). It is determined that compaction is completed 4 to 5 seconds later. This tends to be common although there are differences in sound pressure level and elapsed time in each case.

The reason why the sound pressure changes with the lapse of time since the start of the vibrator will be discussed with reference to FIG. In the first stage of the initial stage of compaction as shown in FIG. 5 (A), since the vibrator has a narrow vibration area due to the large number of voids, the sound pressure is set to the sound pressure in the air. Level (about 0.05Pa). That is,
Sound pressure level is low. Next, in the second stage of compaction as shown in FIG. 5 (B), as compaction proceeds, the voids in the aggregate become smaller due to vibration, and the vibration region expands. Thus, it is considered that the sound pressure level increases because the vibration propagates to a wide area and the wide area vibrates.

Further, in the third stage of compaction as shown in FIG. 5 (C), the voids in the aggregate become small and the mortar fills the voids, so that the fluidization of the concrete starts. Since the mechanical impedance of the vibrator and the concrete becomes smaller, the change of the vibration to the solid propagation sound becomes smaller, and the sound pressure level seems to start decreasing. The fourth compaction as shown in FIG.
At the stage, it is considered that the sound pressure is slightly increased by expanding the compaction area after the mortar fills the voids of the aggregate and the concrete becomes almost fluidized as a whole.
Breathing may occur simultaneously with the expansion of the compaction area. However, from the viewpoint of securing the strength, which is the maximum necessary condition for concrete, after a predetermined time has passed through the fourth stage, it is determined that compaction has been completed. It sounds like a thing.

First, FIGS. 4 (a), (b), (c),
The teacher data created based on the time-dependent data of the sound pressure level indicated by the envelope shown in (d) is input to the learning unit 444 of the learning computer 44. As shown in FIG. 6, one piece of teacher data created at this time is set every two seconds during a period from the start of compaction to 60 seconds after the compaction, and is composed of a total of 31 data. Note that the period and the set interval of the teacher data are not limited to these. FIG. 6A shows an example of teacher data in a pattern in which the sound pressure level gradually increases from the start of compaction, decreases after passing the peak, and becomes 0 after 22 seconds. In the teacher data, as in the case of each envelope in FIG. 4, the sound pressure level rises as the compaction work progresses (S1), and rises to a high level peak. Then, when the compacting operation further proceeds, the sound pressure level decreases (S2), and thereafter, the sound pressure level slightly increases (S2).
3) It has a feature common to each case of FIG. 4 in that compaction ends after 4 to 5 seconds. Then, the output from the neural network, that is, the teacher signal at this time is set to the maximum value (for example, 0.99). Then, FIG.
Shows an example of teacher data in a pattern in which the sound pressure level gradually increases from the start of compaction, decreases after passing the peak, and becomes 0 within 20 seconds. This teacher data indicates a pattern in which compaction cannot be determined. Then, the output from the neural network at this time, that is, the teacher signal is set to the minimum value (for example, 0.01). In this way, two pieces of teacher data are generated from one result data, the maximum value and the minimum value of the teacher signal. These teacher signals express the matching rate between the teacher data and the actual data.

A case where teacher data as shown in FIG. 6A is input to the neural network of the learning unit 444 will be described with reference to FIG. The teacher data indicated by the envelope in FIG. 6A is composed of 31 sound pressure level data (0 to 1) from D (0) to D (30). These teacher data are input to the input units N (0) to N (i) in the input layer of the neural network in FIG. That is,
The input unit N receives the teacher data D (0) at the start of compaction.
(0), and input the teacher data D (1) two seconds later into the input unit N (1), and sequentially input them.

In units M (0) to M (j) of the intermediate layer of the neural network shown in FIG. 3, output data I (0) to I (i) from the input unit of the input layer and the intermediate layer to the intermediate layer Input data to the units M (0) to M (j) of each intermediate layer using the coupling weight coefficient W (ji) to the unit and the offset values of the units M (0) to M (j) of the intermediate layer. U (0) to U (j) are obtained. Then, output data H (0) from units M (0) to M (j) of each intermediate layer
ＨH (j) is obtained by the sigmoid function shown in FIG. The input / output characteristics of this sigmoid function are, as shown in FIG.
When j) is 0, the output data is 0.5, and when the input data exceeds 0, the output data gradually approaches 1. That is, according to the sigmoid function having such characteristics, output data of 0 to 1 is obtained instead of output data of 0 or 1 as in a general threshold function. Output data of each intermediate layer unit is calculated as input data of the output layer unit P. That is, using the output data H (0) to H (j) from the units of each intermediate layer, the coupling load coefficient V (kj) from the intermediate layer to the output layer, and the offset value of the unit of the output layer, The input data U (k) to the unit P in the output layer is obtained. Then, in the output layer unit P, output data O (k) is obtained using the input data S (0) to U (k) and the sigmoid function.

From the difference between the input teacher signal and the output data O (k) of the output layer, an error δk with respect to the coupling weight coefficient connected to the unit of the output layer and the offset value of the unit of the intermediate layer is obtained. Further, based on the error δk, the coupling load coefficient Vkj from the intermediate layer to the output layer, and the output data of the intermediate layer, a coupling load coefficient connected to the unit of the intermediate layer and an error δj with respect to the offset of the unit of the intermediate layer are obtained.

By adding the error δk, the output data of the unit of the intermediate layer, and the cough of the constant α, the coupling weight coefficient Vkj from the unit of the intermediate layer to the unit of the output layer is added.
To correct. Further, the offset of the unit in the output layer is corrected by adding the product of the error δk and the constant β. Next, by adding the error δj in the unit of the intermediate layer and the product of the output data of the unit of the input layer and the constant α, the coupling weight coefficient Wji from the unit of the input layer to the unit of the intermediate layer is corrected. . The offset of the unit in the intermediate layer is corrected by adding the product of the error δj and the constant β. In this way, a neural network is constructed in which the coupling weight coefficients are sequentially corrected to obtain output data suitable for the teacher data.

Next, the connection weight coefficient is similarly corrected using the teacher data of another pattern. By applying such a back-propagation algorithm to teacher data of a plurality of patterns, it is possible to make a proper determination with respect to a temporal change of the sound pressure level of various patterns. You can get it. Further, by learning a pattern in which the compaction end cannot be determined as shown in FIG. 6B, for example, when the sound pressure changes at the same sound pressure as in FIG. 6 during actual compaction,
End of compaction in the time zone between FIGS. (A) and (b) /
The accuracy of unfinished determination is increased. The obtained coupling weight coefficient is sent to the determination computer 43 by a medium 45 such as a LAN,
The connection weight coefficient between the layers of the neural network forming the finding unit 433 is set.

Next, using the judgment computer 43 in which the coupling weight coefficient obtained by the above processing is set,
The procedure for determining the timing of the end of compaction in the actual placing work will be described with reference to the flowchart of FIG. Computer for judging output of A / D converter 42
Set to import to. Then, the operation of the vibrator is started in step S1, and the vibration sound detected by the microphone in step S2 is transmitted to step S2.
In step S4, a filtering process is performed, and in step S4, an envelope of a change in sound pressure level with time is created. Step S5
In, the created envelope is approximated by a cubic curve.
In step S6, the sound pressure level data is stored in the third storage unit 431 every predetermined time, for example, every two seconds. For example, as shown in FIG. 7, after six seconds, four sound pressure level data E (0) to E (3) are accumulated. Then, in step S7, the creation unit 432 sets the values of the remaining sound pressure level data E (4) to E (30) to 0, and sets 31 sound pressure level data E (0) to E (30). Create

In step S8, the 31 sound pressure level data E (0) to E (3) created by the creating section 432 are set.
0) is input to the calculating unit 433 in which the coupling weight coefficient sent from the learning computer 44 is set. That is, the sound pressure level data E (0) is input to the input unit N (0) of the input layer, and the other sound pressure level data E (1) to E (30).
Are sequentially input to the input units N (1) to N (30). Then, in step S9, the matching rate output from the output layer unit P is input to the determination unit 434. In step S10, the determination unit 434 compares the match rate with a predetermined threshold value, for example, 0.95, and when the match rate is equal to or greater than the threshold value, determines that compaction is to be completed.
Proceed to. If it is less than the threshold value, the process returns to step S2, further waits until data E (4) two seconds later is accumulated, and thereafter, the same applies to the five sound pressure level data E detected by the microphone. The same processing is performed on (0) to E (4) and the sound pressure level data E (5) to E (30) created by the creating unit, and the matching rate is determined and determined. If the matching rate based on the five sound pressure level data from the start of compaction to 8 seconds after is also less than the threshold value, the sound pressure level data for every two seconds is sequentially accumulated, and the matching rate is similarly determined and determined. . Further, as shown in FIG. 7B, nine sound pressure level data up to 16 seconds later are input, and the matching rate is determined to determine. Also in this case, the compaction is not determined to be completed.

After the compacting unit 434 determines that compaction is completed, in step S11, after waiting for a predetermined time, for example, four seconds later, the buzzer 5 is sounded in step S12 to notify the user. Step S11 corresponds to a standby unit. The worker who hears the sound of the buzzer 5 stops the vibrator 2 in step S13 to stop the compaction work.

As described above, when the compaction work by the vibrator is stopped by the judgment of the judgment computer by the neural network, the large amount of breathing water is generated before the generation of a large amount of breathing water and the density is sufficiently high regardless of the judgment of the person. Since concrete can be obtained, there is no occurrence of laittance and sufficient strength can be obtained.

In the above embodiment, the envelope data is approximated by a curve. However, the present invention is not limited to this, and the envelope data may be directly used as an input value. In the above embodiment, a sigmoid function is used as an output function to each unit. However, the present invention is not limited to this, and a threshold function may be used. Further, the number of the intermediate layers is only one, but is not limited thereto, and a plurality of layers may be provided. Further, the data recognition section (the section from the start of compaction to 4 seconds before the end of compaction) in the input value is not limited to the present embodiment, but can be expanded or contracted as appropriate. Along with this, the standby time can be changed as appropriate, including 0 seconds. Note that the determination computer may also be used as the learning computer to obtain the coupling weight coefficient. Further, the microphone is not limited to the place shown in FIG. 1 and may be provided at a position closer to the vibrator as long as the microphone is provided so that vibration of the vibrator is not directly transmitted. Further, an alarm lamp may be turned on or blinked instead of the buzzer, or the vibrator may be automatically stopped. Further, a microphone, a control device, and a buzzer may be built in a unit separated and independent from the compaction work vehicle 1. Furthermore, instead of detecting a change in sound pressure by a microphone, a change in current of an electric motor as a drive source of a vibrator and a change in load of a hydraulic motor are electrically detected, and based on the detected change, a compaction state is determined. May be determined to stop the vibrator.

[0033]

As described above, according to the present invention, it is possible to know an appropriate timing for stopping the vibrator irrespective of human judgment, so that the time required for applying the vibrator is too short and sufficient strength can be obtained. And the time required to apply the ibrator is too long to generate breathing water, and the burden of monitoring the timing of the compaction judgment by the operator can be reduced. Can also respond to the current situation where there is a shortage.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a configuration of an embodiment of a concrete compaction device according to the present invention.

FIG. 2 is a block diagram of a main part.

FIG. 3 is a configuration diagram of a neural network.

FIG. 4 is a view showing a change in surface height of concrete.

FIG. 5 is a schematic diagram for explaining the situation of concrete compaction work.

FIG. 6 is an example of teacher data.

FIG. 7 is an example of actual input data.

FIG. 8 is a flowchart illustrating a procedure for determining the end of compaction.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Concrete compaction apparatus 1 Compaction work vehicle 2 Vibrator 3 Microphone 4 Control device 5 Buzzer 43 Computer for judgment 431 Third accumulation part 432 Creation part 433 Determining part 434 Determination part 44 Computer for learning 441 First accumulation part 442 Creation part 443 second storage unit 444 learning unit 45 medium such as LAN

Claims (8)

[Claims] In a concrete compaction judging method for judging the end of compaction when compacting cast concrete with a vibrator, data on changes with time of vibration from the start of compaction of concrete to the end of compaction are previously recorded in actual data. In the process of accumulating as and actual compaction work, while detecting the temporal change of the vibration by the vibrator from the start of compaction of the concrete, the temporal change of the detected vibration is sequentially compared with the actual data, and the matching rate of the two is compared. , And a step of determining that compaction has been completed when the match rate is equal to or greater than a predetermined threshold value. 2. A step of constructing a neural network using input values as time-dependent vibration data of a vibrator, and using output values as a matching rate, and converting time-dependent data of vibration from a vibrator from the start of compaction to the end of compaction of concrete. In the process of completing the neural network and the actual compaction work by detecting the change over time of the vibration from the vibrator from the start of compaction of concrete, by setting the coupling weight coefficient as the teacher data with the maximum matching rate, A process of sequentially comparing the temporal change of the detected vibration with the teacher data using the neural network to obtain a matching rate between the two, and determining that compaction is completed when the matching rate is equal to or more than a predetermined threshold value. And determining the compaction of the concrete. 3. Neural data is created by preparing time-dependent change data of vibration by a vibrator when concrete compaction has not been completed, and using the time-dependent change data as teacher data with a minimum matching rate as a coupling load coefficient. The method of claim 2, further comprising the step of completing a network. 4. The method according to claim 1, wherein a change in vibration caused by the vibrator is detected as a change in sound pressure.
The method for determining concrete compaction according to any one of the above. 5. An output means for outputting time-dependent change data of vibration by a vibrator for compacting the poured concrete, and calculating a matching rate by comparing said time-dependent change data with predetermined teacher data using a neural network. Output means, and determining means for comparing the calculated matching rate with a predetermined threshold value, and determining that compaction is completed when the matching rate is equal to or greater than the threshold value. Concrete compaction judgment device. 6. A learning means for setting a coupling weight coefficient of a neural network of a deriving means on the basis of time-dependent change data of vibration from the start to the end of compaction as a reference. 3. The concrete compaction judging device according to 1. 7. A stand-by means for outputting a vibrator stop command after a predetermined stand-by time has elapsed after the judgment means judges that compaction has been completed. The concrete compaction judging device according to claim 1. 8. The concrete fastening device according to claim 5, wherein the output means detects a change in vibration caused by the vibrator with a microphone and outputs data indicating a change with time of the signal. Hardening judgment device.