Asphalt plant operation support device and asphalt plant

Abstract

To provide a technique to appropriately and simply hand down the manual operation techniques of experienced operators of asphalt plants.SOLUTION: An operation support device 100 of an asphalt plant 1 includes a recommended opening output part 84 that outputs a recommended opening degree Do of a burner with a dryer that heats and dries aggregates, and a display part 72 that displays the recommended opening degree Do. The recommended opening degree output part 84 has trained machine learning models M1, M2 that output the recommended opening degree Do when multiple input values are input. The input values input to the trained machine learning models M1, M2 include at least a target temperature T31 that is a target value of the post-heating temperature of the aggregates, a post-heating aggregate temperature D11 that detects the post-heating temperature of aggregates, and an aggregate supply amount T11 that is the supply amount of aggregates supplied to the dryer. By letting the trained machine learning models M1, M2 to learn by using driving data from experienced operators, the manual driving techniques of experienced operators can be appropriately and easily handed down.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an operation assistance device for assisting the operation of an asphalt plant, and to an asphalt plant. [Background technology]

[0002] An asphalt plant has a dryer for heat-treating aggregate. A conventional asphalt plant is described in, for example, Patent Document 1. The asphalt plant described in Patent Document 1 has a new material dryer that heat-dries aggregate (new material). The aggregate heated and dried by these dryers is sieved by a vibrating sieve provided at the top of the plant body and stored according to particle size in an aggregate storage bin provided at the bottom of the vibrating sieve. The aggregate is then weighed out in the required amount from the aggregate storage bin and supplied to a mixer at the bottom of the plant body, where it is stirred and mixed to produce an asphalt mixture.

[0003] Normally, the operation of the dryer in an asphalt plant can be switched between manual and automatic operation modes. In the automatic operation mode, the burner opening in the dryer is automatically controlled by the operator setting the target aggregate heating temperature on the operation panel. In the automatic operation mode, the burner opening is automatically adjusted using various control technologies such as PID control based on sensor detection values ​​such as the aggregate temperature at the dryer outlet. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-26647 A Summary of the Invention [Problem to be solved by the invention]

[0005] In this automatic operation mode, a relatively stable operation is possible when the temperature inside the drum of the dryer is maintained at a predetermined temperature or higher. However, when the operation is stopped at night or restarted after a long period of operation, the temperature inside the drum of the dryer is low. When the operation is started in this state where the temperature inside the drum is low, the operation is easily affected by conditions such as the weather, the moisture content of the input aggregate, and the supply amount of the input aggregate, and therefore the operation is not necessarily stable.

[0006] When starting operation of the dryer when the temperature inside the drum is low, the burner is first ignited to preheat the dryer drum, and after the temperature of the exhaust gas discharged from the dryer reaches a predetermined temperature, aggregate is supplied to the dryer and heating of the aggregate begins.

[0007] In this case, when operating in automatic operation mode, depending on the moisture content of the aggregate fed into the drum, the amount of aggregate supplied, and weather conditions, the aggregate discharged from the dryer outlet may not reach an aggregate temperature (160°C) that can be used to manufacture asphalt mixture. In that case, the unusable aggregate is transported to the aggregate storage bin in the plant body downstream of the dryer. Therefore, in order to recover the unusable aggregate, it is necessary to supply the aggregate from the aggregate storage bin to the mixer at the bottom of the plant body and discharge it from the mixer. In other words, the work of recovering the unusable aggregate takes time and effort.

[0008] For this reason, when starting the dryer operation when the temperature inside the drum is low, the operator often adjusts the burner opening in manual operation mode. A skilled operator with ample plant operation experience can appropriately operate the burner opening based on the moisture content of the aggregate to be put into the drum, the amount of aggregate supplied, and weather conditions, and can heat the aggregate put in initially to the target aggregate temperature (160°C) before discharging it. This eliminates the need to recover aggregate that is too cold to be used, and makes it possible to shorten the time from starting the plant operation to starting the production of asphalt mixture, compared to automatic operation mode.

[0009] However, when an unskilled operator with relatively little experience in operating a plant performs manual operation, it is difficult for him or her to perform the same appropriate operation as an experienced operator. As a result, the burner opening may be opened too much, causing the exhaust gas temperature to rise and burning out the bag filter, which is a dust-collecting filter installed downstream of the exhaust gas flue. In addition, the burner opening may not be adjusted properly, and it may take time to raise the temperature of the aggregate discharged from the dryer outlet, resulting in the time from the start of plant operation to the start of asphalt mixture production being longer than in automatic operation mode.

[0010] In order for unskilled operators to learn proper operation like skilled operators, skilled operators need to provide driving guidance to unskilled operators when they are operating manually, and the skilled operators' operating know-how needs to be compiled into a manual and shared with unskilled operators. However, these efforts are time-consuming and labor-intensive for skilled operators, so there is a demand for technology that allows manual operation skills to be passed on more appropriately and simply.

[0011] The present invention has been made in consideration of the above-mentioned points, and has an object to provide an operation assistance device and an operation assistance method for an asphalt plant that enable the manual operation techniques of an experienced operator of an asphalt plant to be appropriately and simply transferred. [Means for solving the problem]

[0012] In order to solve the above problems, the first invention of the present application is an operation support device for an asphalt plant comprising: a dryer having a burner at one end of a drum for heating and drying aggregate; an exhaust flue for discharging exhaust gas from the dryer to the outside; a dust collecting filter inserted in the exhaust flue; and an input unit that allows an operator to operate the burner opening, which is the opening of the burner. The device is equipped with a recommended opening output unit that outputs a recommended opening of the burner; and a display unit that displays the recommended opening to the operator, and the recommended opening output unit has a trained machine learning model that outputs the recommended opening when multiple input values ​​are input, and the input values ​​input to the trained machine learning model include at least a target temperature that is a target value for the temperature of the aggregate at the outlet of the dryer, a heated aggregate temperature detected by detecting the temperature of the aggregate at the outlet of the dryer, and an aggregate supply amount that is the supply amount of the aggregate supplied to the dryer.

[0013] A second invention of the present application is an operating assistance device of the first invention, wherein the input values ​​further include at least one of a filter inlet temperature detected by detecting the temperature of the exhaust gas in the exhaust flue at the inlet of the dust collecting filter, a pre-heated aggregate temperature which is the temperature of the aggregate supplied to the dryer, an aggregate temperature rise rate which is the difference between the post-heated aggregate temperature and the pre-heated aggregate temperature, a flue temperature detected by detecting the temperature in the exhaust flue, an outside air temperature of the asphalt plant, and a humidity outside the asphalt plant.

[0014] A third aspect of the present invention is a driving assistance device according to the first aspect of the present invention, comprising: a learning unit that performs machine learning on a machine learning model using operation data of the asphalt plant by the operator. The learning unit performs supervised learning on the machine learning model using the input values ​​in the operation data as input variables and the burner opening degree in the operation data as teacher data.

[0015] A fourth invention of the present application is a driving assistance device of the third invention, wherein the learning unit generates a plurality of trained machine learning models corresponding to each of the plurality of operators by performing machine learning on the machine learning model using a plurality of the driving data for each of the plurality of operators.

[0016] The fifth invention of the present application is an asphalt plant comprising a dryer having a burner at one end of a drum for heating and drying aggregate, an exhaust flue for discharging exhaust gas from the dryer to the outside, a dust collecting filter inserted in the exhaust flue, an input unit that allows an operator to operate the burner opening, which is the opening degree of the burner, and an operation support device according to any one of the first to fourth inventions. Effect of the Invention

[0017] According to the first to fifth inventions of the present application, when performing manual operation, the operator can adjust the burner opening of the dryer by referring to the recommended opening. This allows the operation to be performed at an appropriate burner opening even when an unskilled operator with relatively little experience in operating an asphalt plant performs manual operation. As a result, the aggregate temperature can be efficiently heated to a desired temperature while suppressing burnout of the dust collection filter. In other words, manual operation techniques can be appropriately and simply transferred without guidance by an experienced operator.

[0018] In particular, according to the fourth invention of the present application, a trained machine learning model can be generated for each skilled operator. Therefore, other operators can learn the characteristics of each skilled operator. In addition, trained machine learning models can be generated for each driving perspective, such as fuel-efficient driving and driving that starts asphalt mixture production in a short time, not just for each skilled operator, so that other operators can learn fuel-efficient driving operations, driving operations that start asphalt mixture production in a short time, etc. [Brief description of the drawings]

[0019] [Figure 1]FIG. 1 is a diagram showing the configuration of an asphalt plant. [Diagram 2] FIG. 2 is a diagram conceptually illustrating the configuration of an operation unit and a control unit of an asphalt plant. [Diagram 3] FIG. 1 is a diagram showing an example of the configuration of a machine learning model M used for operational support of an asphalt plant. [Figure 4] FIG. 1 is a diagram showing an example of the configuration of a machine learning model M used for operational support of an asphalt plant. [Diagram 5] 11 is a diagram showing an example of a display mode when a recommended opening output unit displays a recommended opening on a display unit. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

[0021] <1. Embodiment> <1-1. Asphalt plant configuration> Fig. 1 is a diagram showing the configuration of an asphalt plant 1. This asphalt plant 1 produces an asphalt mixture by mixing new material, recycled material, molten asphalt, and filler (stone powder). As shown in Fig. 1, the asphalt plant 1 includes a new material hopper 10, a recycled material hopper 20, a new material dryer 30, a recycled dryer 40, an exhaust path 50, and a plant main body 90.

[0022] The new material hopper 10 temporarily stores new aggregate for asphalt mixture (hereinafter referred to as "new material") that is brought into the asphalt plant 1. The new material hopper 10 supplies the new material to the drum 31 of the new material dryer 30. As the new material, various natural coarse aggregates (natural gravel), crushed stone, natural fine aggregates (natural sand), crushed sand, etc. are used.

[0023] New materials of each particle size are stored in new material hoppers 10 according to their particle size, and after the required amount of new materials according to the mix design of the asphalt mixture to be produced is cut out from each new material hopper 10, all of the cut new materials are supplied together to the drum 31 of the new material dryer 30.

[0024] The new wood dryer 30 is a device that heats and dries new wood supplied from the new wood hopper 10. The new wood dryer 30 comprises a drum 31, a machine base 32, and a burner 33. The drum 31 has a new wood dryer inlet 301 at one end, through which new wood is fed from the new wood hopper 10, and a new wood dryer outlet 302 at the other end, through which new wood that has been subjected to a heat drying process is discharged. The machine base 32 is provided to support the drum 31 from below so as to be freely rotatable. The burner 33 is located at the other end of the new wood dryer 30.

[0025] The new material dryer 30 is connected to the plant body 90 by a vertical conveying device 101. Specifically, the new material dryer outlet 302 is connected to the upstream end of the vertical conveying device 101. In addition, an upper portion of one end of the drum 31 is connected to a first exhaust flue 51 of the exhaust path 50, which will be described later.

[0026] The new material dryer 30 has a scraping blade (not shown) on the inner periphery of the drum 31. The scraping blade rotates at a predetermined speed together with the drum 31 by a drive device (not shown), and agitates the new material supplied to the drum 31. The burner 33 heats the new material supplied to the drum 31 to a predetermined temperature and heats and dries it.

[0027] The vertical conveying device 101 conveys the new wood that has been subjected to the heating and drying process in the new wood dryer 30 to the plant main body 90. Exhaust gas generated by heating the new wood is discharged to the first exhaust flue 51.

[0028] The recycled material hopper 20 temporarily stores the recycled aggregate (recycled material) brought into the asphalt plant 1. The recycled material hopper 20 supplies the recycled material to a drum 41 of the recycle dryer 40.

[0029] The regenerative dryer 40 is a device that heat-treats the regenerated material supplied from the regenerated material hopper 20 in the asphalt plant 1. The regenerative dryer 40 includes a drum 41, a machine base 42, a burner 43, a regenerated material storage bin 44, and a regenerated material measuring tank 45.

[0030] The drum 41 has a regenerative dryer inlet 401 at one end, through which the regenerative material is fed from the regenerative material hopper 20, and a regenerative dryer outlet 402 at the other end, through which the regenerative material that has been subjected to the heat drying process is discharged. The machine base 42 is provided to rotatably support the drum 41 from below. The burner 43 is located at one end of the regenerative dryer 40.

[0031] The regenerant material storage bin 44 is located at the other end of the regenerant dryer 40. The regenerant material storage bin 44 is connected to the regenerant dryer outlet 402 of the drum 41. Therefore, the regenerant material that has been subjected to the heat drying process is supplied from the drum 41 to the regenerant material storage bin 44.

[0032] The recycled material metering tank 45 is located at the lower end of the recycled material storage bin 44. The recycled material metering tank 45 measures the recycled material stored inside the recycled material storage bin 44 and supplies it downward.

[0033] The regenerative dryer 40 is connected to the plant body 90 by an input chute 102. Specifically, the upstream end of the input chute 102 is connected to an outlet at the bottom of the regenerative material metering tank 45. In addition, the top of the regenerative material storage bin 44 is connected to a second exhaust flue 52 of the exhaust path 50, which will be described later.

[0034] The regenerative dryer 40 has a scraping blade (not shown) on the inner periphery of the drum 41. The scraping blade rotates at a predetermined speed together with the drum 41 by a drive device (not shown), and stirs the regenerative material supplied to the drum 41. The burner 43 heats the regenerative material supplied to the drum 41 to a predetermined temperature and heats and dries it.

[0035] The heated regenerated material is temporarily stored in the regenerated material storage bin 44. The regenerated material measuring tank 45 measures a predetermined amount of the regenerated material stored in the regenerated material storage bin 44 and supplies it to the input chute 102. The input chute 102 transports the predetermined amount of regenerated material to the plant main body 90. Exhaust gas generated by heating the regenerated material is discharged from the regenerated material storage bin 44 to the second exhaust flue 52.

[0036] The exhaust path 50 is a gas exhaust path composed of piping and the like, which discharges exhaust gas from the new material dryer 30 and the regenerative dryer 40, which are the sources of exhaust gas, to the outside. The exhaust path 50 has a first exhaust flue 51, a second exhaust flue 52, and a common flue 53. The upstream end of the first exhaust flue 51 is connected to the top of the drum 31 of the new material dryer 30. The upstream end of the second exhaust flue 52 is connected to the top of the regenerative material storage bin 44 of the regenerative dryer 40. The downstream end of the first exhaust flue 51 and the downstream end of the second exhaust flue 52 are connected to the upstream end of the common flue 53. As a result, the exhaust gas discharged from the new material dryer 30 and the exhaust gas discharged from the regenerative dryer 40 join together and flow through the common flue 53.

[0037] In the common flue 53, an inertial dust collector 61, a filter dust collector 62, and an exhaust fan 63 are inserted in order from the upstream side. The downstream end of the common flue 53 is connected to a chimney 64. The exhaust fan 63 generates a flow of exhaust gas from the upstream side to the downstream side in the common flue 53. As a result, the exhaust gas that flows into the common flue 53 from the first exhaust flue 51 and the second exhaust flue 52 is discharged from the chimney 64 after dust is removed in the inertial dust collector 61 and the filter dust collector 62. The inertial dust collector 61 and the filter dust collector 62 are collectively referred to as a "dust collection filter."

[0038] The inertial dust collector 61 separates dust particles having a relatively large particle size from the dust particles contained in the exhaust gas. For example, a cyclone type dust collector is used as the inertial dust collector 61. The dust particles having a relatively large particle size separated in the inertial dust collector 61 are transported to the vicinity of the new material dryer outlet 302 by a transport device (not shown) and transported to the plant main body 90 by the vertical transport device 101 together with the new material that has been heat-treated in the new material dryer 30. The exhaust gas discharged from the inertial dust collector 61 is introduced into the downstream filter dust collector 62.

[0039] The filter dust collector 62 separates fine dust (particles) contained in the exhaust gas. The filter dust collector 62 captures and removes the fine dust using a filter cloth. The fine dust removed by the filter cloth is transported to a filler storage bin 94 (described later) of the plant main body 90 via a pressure feed pipe 103. The filter dust collector 62 discharges the exhaust gas from which the dust has been removed to the downstream side of the common flue 53. The exhaust gas from which the dust has been removed by the inertial dust collector 61 and the filter dust collector 62 is discharged from a chimney 64.

[0040] The plant body 90 is a facility for producing an asphalt mixture by mixing new materials that have been heated and dried, recycled materials that have been heated and dried, filler, and molten asphalt. The plant body 90 includes a vibrating sieve 91, an aggregate storage bin 92, an aggregate measuring tank 93, a filler storage bin 94, a filler measuring tank 95, a molten asphalt tank 96, a supply pump 97, an asphalt measuring tank 98, and a mixer 99.

[0041] The vibrating sieve 91 receives new material supplied from the vertical conveying device 101. The vibrating sieve 91 sieves the received new material and supplies it to the aggregate storage bin 92 by particle size. In other words, the aggregate storage bin 92 is an aggregate storage section that stores the aggregate (new material) dried by the new material dryer 30. The aggregate measuring tank 93 weighs each of the new materials stored in the aggregate storage bin 92 by particle size and supplies it to the mixer 99.

[0042] The filler storage bin 94 is a filler storage section that temporarily stores the stone powder brought into the asphalt plant 1 and the powder transported from the filter dust collector 62 via the pressure feed pipe 103. In the following description, the stone powder brought into the asphalt plant 1 and the powder transported from the pressure feed pipe 103 are collectively referred to as "filler". The filler measuring tank 95 measures a predetermined amount of filler stored in the filler storage bin 94 and supplies it to the mixer 99.

[0043] The molten asphalt tank 96 is a molten asphalt storage unit that temporarily stores the molten asphalt, which is a material for the asphalt mixture. The supply pump 97 is an asphalt transport unit that transports the molten asphalt from the molten asphalt tank 96 to the asphalt metering tank 98. The asphalt metering tank 98 measures the asphalt transported by the supply pump 97 and supplies it to the mixer 99.

[0044] The mixer 99 produces an asphalt mixture by mixing new material supplied from the aggregate metering tank 93, recycled material supplied from the recycled material metering tank 45, filler supplied from the filler metering tank 95, and molten asphalt supplied from the asphalt metering tank 98 for a predetermined period of time.

[0045] <1-2. About the control unit of the asphalt plant> The asphalt plant 1 has an operation unit 70 and a control unit 80 for controlling the above-mentioned respective units. Fig. 2 is a diagram conceptually showing the configuration of the operation unit 70 and the control unit 80 of the asphalt plant 1.

[0046] The operation unit 70 is a device for an operator to operate. As shown in FIG.

[0047] The input unit 71 is provided with a large number of operation buttons, operation levers, etc. for operating each part of the asphalt plant 1. In the input unit 71, operation command values ​​and target values ​​for each part of the asphalt plant 1 can be input to the control unit 80.

[0048] The operation command values ​​that can be input to the input unit 71 are, for example, as follows: A first supply amount T11, which is the supply amount of each grain size of new wood supplied from the new wood hopper 10 to the new wood dryer 30 A second supply amount T12, which is the amount of regenerated material supplied from the regenerated material hopper 20 to the regenerated dryer 40 First burner opening T21, which is the opening of the burner 33 of the new wood dryer 30 The second burner opening T22, which is the opening of the burner 43 of the regenerative dryer 40

[0049] Moreover, the target values ​​that can be input in the input unit 71 are, for example, as follows: A first target temperature T31 which is a target temperature of the new wood discharged from the new wood dryer 30 A second target temperature T32 which is the target temperature of the regenerated material discharged from the regenerated dryer 40

[0050] The display unit 72 can display the operating status of each part of the asphalt plant 1, detection values ​​from sensors provided in each part of the asphalt plant 1, etc. In addition, the display unit 72 can display a recommended opening Do received from a recommended opening output unit 84 described later.

[0051] 1, the control unit 80 is configured by a computer including a processor 801 such as a CPU or GPU, a memory 802 such as a RAM, and a storage medium 803 such as a hard disk drive or SSD. The control unit 80 controls the operation of each part in the asphalt plant 1 by the processor 801 performing arithmetic processing based on a computer program P and data D stored in the storage medium 803.

[0052] The control unit 80 has an operation control unit 81, an automatic driving control unit 82, a learning unit 83, and a recommended opening output unit 84, which are functionally realized on software by such a computer. The control unit 80 also has a memory unit 89.

[0053] The control unit 80 receives operation command values ​​and target values ​​input from the input unit 71. In addition, various detection values ​​detected by sensors provided in each part of the asphalt plant 1 are input to the control unit 80. The control unit 80 controls the operation of each part of the asphalt plant 1 based on these values.

[0054] The detection values ​​input to the control unit 80 are, for example, as follows: A first aggregate temperature D11, which is the temperature of the new material discharged from the new material dryer 30, detected by a temperature sensor S11 provided at the new material dryer outlet 302. A second aggregate temperature D12, which is the temperature of the regenerated material discharged from the regenerated dryer 40, detected by a temperature sensor S12 provided at the regenerated dryer outlet 402. A first flue temperature D21, which is the temperature of the exhaust gas discharged from the new material dryer 30 and detected by a temperature sensor S21 provided at the connection point of the first exhaust flue 51 with the new material dryer 30 A second flue temperature D22, which is the temperature of the flue gas discharged from the regenerative dryer 40 and detected by a temperature sensor S22 provided at the connection point of the second exhaust flue 52 with the regenerative dryer 40 Filter inlet temperature D3 detected by a temperature sensor S3 provided at the inlet of the inertial dust collector 61 of the common flue 53 An external temperature D4 detected by a temperature sensor S4 installed outside the asphalt plant 1 External humidity D5 detected by a humidity sensor S5 installed outside the asphalt plant 1

[0055] The operation control unit 81 controls the operation of each part of the asphalt plant 1. The operation control unit 81 can be switched between an automatic operation mode and a manual operation mode. In the manual operation mode, the operation control unit 81 controls the operation of each part of the asphalt plant 1 based on an operation command input by the operator from the input unit 71. In the manual operation mode, the operation control unit 81 controls the burner opening of the burner 33 of the new material dryer 30 according to the first burner opening T21 input by the operator from the input unit 71, and controls the burner opening of the burner 43 of the regenerative dryer 40 according to the second burner opening T22 input by the operator from the input unit 71.

[0056] In the automatic operation mode, the automatic operation control unit 82 calculates the first burner opening degree T21 and the second burner opening degree T22 based on the first target temperature T31 and the second target temperature T32 input from the input unit 71 and the temperatures D11, D12, D21, D22, and D3 input from the respective temperature sensors S11, S12, S21, S22, and S3 in the asphalt plant 1. Then, the operation control unit 81 controls the burner opening degrees of the burner 33 of the new material dryer 30 and the burner 43 of the regenerative dryer 40 according to the first burner opening degree T21 and the second burner opening degree T22 calculated by the automatic operation control unit 82.

[0057] The automatic operation control unit 82 calculates the first burner opening degree T21 and the second burner opening degree T22 by feedback control such as PID control and feedforward control. Therefore, when the asphalt plant 1 is started up or shut down, or during non-steady operation such as a sudden increase or decrease in the amount of aggregate supplied, if the operation is performed in the automatic operation mode, the aggregate discharged from the outlet of the dryers 30, 40 may not reach an aggregate temperature (160°C) that can be used for producing an asphalt mixture, depending on conditions such as weather, aggregate moisture content, and aggregate supply amount. Therefore, during non-steady operation, it is preferable that the operator manually operates in the manual operation mode with reference to the recommended opening degree Do output by the recommended opening degree output unit 84.

[0058] The learning unit 83 performs machine learning on the machine learning model M. Specifically, the learning unit 83 performs machine learning on the unlearned machine learning model M0 to generate a trained machine learning model, and performs re-learning on the trained machine learning models M1, M2, etc. Note that the unlearned and trained machine learning models M0, M1, M2, etc. are collectively referred to as the machine learning model M.

[0059] The machine learning model M trained in the learning unit 83 is a model for outputting recommended openings Do1, Do2 that estimate appropriate burner openings T21, T22 during unsteady operation such as start-up and shut-down of the operation of the asphalt plant 1 or a sudden increase or decrease in the aggregate supply amount. The machine learning model M trained in the learning unit 83 outputs recommended openings Do1, Do2 when multiple input values ​​are input.

[0060] In the following, a case will be described where only new materials are used at the start-up of the asphalt plant 1. In other words, since no regenerated materials are used, the regenerated dryer 40 is not used. In this case, the machine learning model M outputs only the recommended opening degree Do1 of the burner 33 of the new material dryer 30.

[0061] In this case, the input values ​​for the machine learning model M are at least A first target temperature T31 which is a target temperature of the new wood discharged from the new wood dryer 30 A first aggregate temperature D11 (a temperature of aggregate after first heating) is the temperature of the new material discharged from the new material dryer 30 and is detected by a temperature sensor S11 provided at the new material dryer outlet 302. A first aggregate supply amount, which is the total supply amount of new material supplied to the new material dryer 30, calculated from the first supply amount T11 Includes.

[0062] In addition, the input value for the machine learning model M is Filter inlet temperature D3 detected by a temperature sensor S3 provided at the inlet of the inertial dust collector 61 of the common flue 53 The first pre-heating aggregate temperature, which is the temperature of the new material supplied to the new material dryer 30 The first aggregate temperature rise rate, which is the difference between the aggregate temperature after the first heating and the aggregate temperature before the first heating Flue temperature detected in the common flue 53 downstream of the inertial dust collector 61 and the filter dust collector 62 and upstream of the exhaust fan 63 An external temperature D4 detected by a temperature sensor S4 installed outside the asphalt plant 1 External humidity D5 detected by a humidity sensor S5 installed outside the asphalt plant 1 These input values ​​may be arbitrarily selected by the user by checking or unchecking a check box displayed on the display unit 72.

[0063] 3 and 4 are diagrams showing an example of the configuration of a machine learning model M used for operational support of the asphalt plant 1. In the example of Fig. 3, this machine learning model M has a two-layer LSTM (Long Short Term Memory). LSTM is a type of RNN (Recurrent Neural Network), which is suitable for learning and predicting time-series data and is capable of learning long-term and short-term features.

[0064] The machine learning model M created by the present inventor was evaluated by the MAE (Mean Absolute Error) between the output recommended opening and the actual burner opening in the teacher data. Table 1 below shows the MAE when learning was performed using operation data Ds collected for a period of 2 weeks, 1 month, and 1 year for each machine learning model M created by stacking the LSTM in one, two, and three layers. As shown in Table 1, when the collection period of the operation data Ds to be learned was 2 weeks, the machine learning model M with two layers of LSTM had a lower MAE value than the machine learning model M with one layer of LSTM, and more accurately estimated human driving. Furthermore, even when the collection period of the operation data Ds to be learned was 1 month or 1 year, the MAE value of the machine learning model M created by stacking two layers of LSTM was smaller than that of the other machine learning models M. From this, it was found that appropriate burner openings T21 and T22 can be predicted by stacking two layers of LSTM. [Table 1]

[0065] In the example of Figure 4, in addition to the example of Figure 3, a convolutional layer is added before the two-layer LSTM. As above, under the conditions that the driving data Ds to be learned is collected for one month and the LSTM has two layers, machine learning model M without a convolutional layer as shown in Figure 3 and machine learning model M with a convolutional layer as shown in Figure 4 were created and the MAE values ​​were compared. As a result, the MAE of machine learning model M without a convolutional layer was 6.02, while the MAE of machine learning model M with a convolutional layer was 5.30, a smaller value. This shows that the estimation accuracy of LSTM is improved by performing feature extraction in the convolutional layer.

[0066] The memory unit 89 stores a large number of sets of operation data Ds, which is time-series data including each sensing data and the burner opening degree T21 during past operations of the asphalt plant 1. The learning unit 83 selects, from among these multiple operation data Ds, multiple pieces of operation data Ds at the start-up of operation in the manual operation mode by a skilled operator.

[0067] Thereafter, the learning unit 83 creates teacher data from the selected multiple operation data Ds, with each of the above input values ​​as input variables and the first burner opening T21 as an output variable, and learns the machine learning model M using the teacher data. The learning end condition in the learning unit 83 is, for example, that the difference between the recommended opening output by the machine learning model M and the first burner opening T21 of the teacher data is equal to or less than a predetermined threshold. The trained machine learning model M that has completed learning is handed over to the recommended opening output unit 84.

[0068] When performing manual operation, the operator can adjust the burner opening T21 of the dryer 30 with reference to the recommended opening output by the trained machine learning model M. By using the operation data Ds that is deemed to have been an appropriate operation as the teacher data of the machine learning model M, even when an unskilled operator with relatively little experience in operating the asphalt plant 1 performs manual operation, the operation can be performed at an appropriate burner opening T21. As a result, the aggregate temperature can be efficiently heated to a desired temperature while suppressing burnout of the dust collection filter (the inertial dust collector 61 and the filtration dust collector 62). In other words, manual operation techniques can be appropriately and simply transferred without the need for guidance by an experienced operator.

[0069] The selection criteria for the driving data Ds to be used as teacher data from among the multiple driving data Ds may be changed according to needs. For example, a trained machine learning model M1 trained using the driving data Ds of an experienced operator A as teacher data and a trained machine learning model M2 trained using the driving data Ds of an experienced operator B as teacher data may be created. In other words, machine learning may be performed on the machine learning model M using multiple driving data Ds for each of multiple operators.

[0070] In this way, multiple trained machine learning models M corresponding to multiple skilled operators can be generated. As a result, other operators can learn the characteristics of each skilled operator. In other words, an unskilled operator can realize and learn the operation of the asphalt plant 1 with the same know-how as skilled operators A and B.

[0071] In addition, by grouping the driving data Ds to be used as teacher data according to seasons and outside temperature ranges, appropriate driving can be achieved for the seasons and outside temperature ranges. In addition, the selection criteria for the driving data Ds to be used as teacher data are not limited to the operator, and driving data Ds may be selected according to various viewpoints, such as driving data Ds with particularly good fuel efficiency or driving data Ds with a short time until the start of asphalt mixture production.

[0072] The recommended opening output unit 84 outputs the recommended opening Do1 using the trained machine learning models M1, M2 delivered from the learning unit 83. Specifically, from various values ​​input from the input unit 71 or detection values ​​input from various sensors, input values ​​to each model M1, M2 are input to each model M1, M2.

[0073] For example, the trained machine learning model M1 outputs a recommended opening Do1 at the start-up of operation of the asphalt plant 1, and its input variables are three values: a first target temperature T31 input from the input unit 71, a first aggregate temperature D11 input from the temperature sensor S11, and a first aggregate supply amount calculated from the first supply amount T11.

[0074] The recommended opening output unit 84 inputs the above input variables into the trained machine learning model M1, and displays the recommended opening Do1 output from the trained machine learning model M1 on the display unit 72. This allows the operator to perform manual driving on the input unit 71 while visually checking the recommended opening Do1 displayed on the display unit 72. In this embodiment, the learning unit 83, the recommended opening output unit 84, the input unit 71, and the display unit 72 configure a driving assistance device.

[0075] In this way, the driving assistance device outputs recommended opening degrees Do1, Do2 using trained machine learning models M1, M2 that have been trained using driving data Ds of an experienced operator and driving data Ds with good fuel efficiency, and displays them on the display unit 72.

[0076] This allows the operation to be performed at appropriate burner openings T21 and T22 even when an unskilled operator with relatively little experience in operating the asphalt plant 1 performs manual operation. As a result, the aggregate temperature can be efficiently increased to a desired temperature while suppressing burnout of the dust collection filters (the inertial dust collector 61 and the filtration dust collector 62). In other words, manual operation techniques can be appropriately and simply transferred without guidance by an experienced operator.

[0077] 5 is a diagram showing an example of a display mode when the recommended opening output unit 84 displays the recommended opening Do on the display unit 72. In the example of FIG. 5, the recommended opening Do is not only displayed numerically, but also the change in the recommended opening Do according to the elapsed time is shown in a graph G. In this way, the operator can visually confirm not only the recommended opening Do at the present time, but also the change over time of the recommended opening Do at a time beyond the present time. Therefore, the operator can perform manual operation on the input unit 71 while anticipating the change over time of the recommended opening Do.

[0078] <2. Modification> Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment.

[0079] In the above embodiment, the control unit 80 of the asphalt plant 1 has the learning unit 83, but the present invention is not limited to this. The learning unit 83 may be realized by a computer separate from the control unit 80. In this case, the control unit 80 transfers a plurality of pieces of operating data Ds to the separate computer, and the learning unit 83 performs a learning process for the machine learning model M. Thereafter, the trained machine learning model M is transferred from the separate computer to the recommended opening degree output unit 84 of the control unit 80.

[0080] In the above embodiment, the machine learning model M outputs only the recommended opening degree Do1 of the burner 33 of the new material dryer 30. However, the machine learning model M may output the recommended opening degree Do1 of the burner 33 of the new material dryer 30 and the recommended opening degree Do2 of the burner 43 of the regenerative dryer 40.

[0081] In that case, the input values ​​for the machine learning model M are at least A first target temperature T31 which is a target temperature of the new wood discharged from the new wood dryer 30 A second target temperature T32 which is the target temperature of the regenerated material discharged from the regenerated dryer 40 A first aggregate temperature D11 (a temperature of aggregate after first heating) is the temperature of the new material discharged from the new material dryer 30 and is detected by a temperature sensor S11 provided at the new material dryer outlet 302. A second aggregate temperature D12 (a temperature of the aggregate after the second heating) which is the temperature of the new material discharged from the new material dryer 30 and detected by a temperature sensor S12 provided at the regeneration dryer outlet 402 A first aggregate supply amount, which is the total supply amount of new material supplied to the new material dryer 30, calculated from the first supply amount T11 The second aggregate supply amount, which is the total supply amount of new material supplied to the regenerator 40, calculated from the second supply amount T12 Includes.

[0082] In the above embodiment, the asphalt plant 1 uses new materials and recycled materials as aggregates, and uses sludge powder and crushed recycled building materials as part of the filler to produce an asphalt mixture. However, the asphalt plant 1 may use only either new materials or recycled materials as aggregates to produce an asphalt mixture. For example, when only new materials are used as aggregates, the recycled materials are not heat-treated, and the regenerative dryer 40 is not operated.

[0083] In the above embodiment, the dust collectors 61, 62 are provided in the common flue 53 connected to both the first exhaust flue 51 and the second exhaust flue 52. However, in the asphalt plant 1, a dust collector may be provided in each of the first exhaust flue 51 and the second exhaust flue 52.

[0084] Furthermore, in the above embodiment, no reference is made to the condition for terminating learning in the learning unit 83, but the condition for terminating learning may be set as appropriate. For example, the condition for terminating learning in the learning unit 83 may be that the MAE value is 5 or less, and in other words, various methods and values ​​may be adopted as long as the learned machine learning model M is at a level where a non-expert can actually imitate the driving of a target expert.

[0085] Furthermore, the elements appearing in the above-described embodiments and modifications may be combined as appropriate to the extent that no contradiction arises. [Industrial Applicability]

[0086] INDUSTRIAL APPLICABILITY The present invention can be used in an operation assistance device and an operation assistance method for an asphalt plant, which can appropriately and simply transfer the manual operation techniques of an experienced operator of an asphalt plant. [Explanation of symbols]

[0087] 1: Asphalt plant 30: New wood dryer 31: Drums 33: Burna 40: Regenerative dryer 41: Drums 43: Burna 50: Exhaust route 51: First exhaust flue 52: Second exhaust flue 53: Common flue 61:Inertial dust collector 62:Filtration dust collector 70:Operation unit 71: Input section 72: Display section 80: Control section 81: Motion control section 82: Automatic driving control unit 83: Learning Department 84: Recommended opening output section 89: Storage section 302: New wood dryer outlet 402: Regenerator dryer outlet D: Data D11: First aggregate temperature D12: Second aggregate temperature D21: 1st flue temperature D22: 2nd flue temperature D3: Filter inlet temperature D4:External temperature D5:External humidity Do1, Do2: Recommended opening Ds: Operation data M, M0, M1, M2: Machine learning models T11: 1st supply amount T12: 2nd supply amount T21: First burner opening T22: Second burner opening T31: 1st target temperature T32: 2nd target temperature

Claims

1. a dryer having a burner at one end of a drum for heating and drying the aggregate; an exhaust flue for discharging exhaust gas from the dryer to the outside; A dust collecting filter inserted in the exhaust flue; An input unit that allows an operator to operate a burner opening degree, which is the opening degree of the burner; An operation assistance device for an asphalt plant comprising: A recommended opening output unit that outputs a recommended opening of the burner; A display unit that displays the recommended opening degree to the operator; and Equipped with The recommended opening output unit is A trained machine learning model that outputs the recommended opening degree when a plurality of input values ​​are input, The input value input to the trained machine learning model is at least A target temperature is a target value of the temperature of the aggregate at an outlet of the dryer; a post-heating aggregate temperature obtained by detecting the temperature of the aggregate at an outlet of the dryer; an aggregate supply amount, which is the amount of the aggregate supplied to the dryer; A driving assistance device comprising:

2. The driving assistance device according to claim 1, The input value is a filter inlet temperature obtained by detecting the temperature of the exhaust gas in the exhaust flue at the inlet of the dust collecting filter; a pre-heating aggregate temperature, which is the temperature of the aggregate supplied to the dryer; an aggregate temperature rise rate which is a difference between the aggregate temperature after heating and the aggregate temperature before heating; A flue temperature detected by detecting the temperature in the exhaust flue; The outside air temperature of the asphalt plant; Humidity outside the asphalt plant; The driving assistance device further includes at least one of the following:

3. The driving assistance device according to claim 1, A learning unit that performs machine learning on a machine learning model using operation data of the asphalt plant by the operator. and The learning unit performs supervised learning for the machine learning model using the input values ​​in the operating data as input variables and the burner opening degree in the operating data as teacher data.

4. The driving assistance device according to claim 3, The learning unit generates a plurality of trained machine learning models corresponding to each of the plurality of operators by performing machine learning on the machine learning model using a plurality of the driving data for each of the plurality of operators.

5. a dryer having a burner at one end of a drum for heating and drying the aggregate; an exhaust flue for discharging exhaust gas from the dryer to the outside; A dust collecting filter inserted in the exhaust flue; An input unit that allows an operator to operate a burner opening degree, which is the opening degree of the burner; A driving assistance device according to any one of claims 1 to 4, Asphalt plant equipped with

Images