JP2000008798A - Tunnel ventilation control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve ventilation control performance by improving a tunnel traffic density predicting method. SOLUTION: A traffic density pattern generating storage device 7 accumulates the measured value of traffic density, generates a daily traffic density pattern and updates it periodically. A Kalman filter inputs present traffic density and the daily traffic density pattern generated and stored in the traffic density pattern generating storage device, and predicts traffic density after the specified time. An air quantity and number determining means computes required air ventilation to make smoke density the target value using vehicle speed data, the predicted traffic density and the smoke density measured value in a tunnel, and determines the air quantity and number of ventilators.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a tunnel ventilation control device, and more particularly to an improvement thereof.

[0002]

2. Description of the Related Art In general, in order to perform ventilation in a tunnel, the amount of soot generated in the tunnel is measured by a soot concentration measuring device, and when the value exceeds an upper limit reference value, the air volume and the number of ventilators are increased. Then, when the value falls below the lower limit reference value, feedback control is performed to reduce the air volume and the number of ventilators. However, with such a control method, in the case of a tunnel with a long tunnel length, when the air volume and the number of ventilators are increased after the smoke concentration in the tunnel exceeds the upper limit reference value,
As an effect thereof, it takes a long time until the smoke concentration is reduced, and during that time, problems such as the high smoke concentration being continued occur.

For this reason, especially in a long tunnel, the traffic volume in the ascending and descending tunnels is measured, the traffic volume in the tunnel is predicted from the measured values, and the amount of soot generated from the predicted traffic volume is calculated. Feedforward control for determining the air volume and the number of ventilators for which the soot concentration is to be a target value has been adopted. As a known example, there is a "tunnel ventilation control system" disclosed in, for example, JP-A-57-4000. This is to check the traffic pattern showing the time change of the daily traffic volume on weekdays, Saturdays, holidays, seasons, etc., create a reference pattern, measure the traffic volume of the tunnel when controlling, A time change pattern of the traffic volume after the current time is predicted from the actually measured current traffic volume and the corresponding reference pattern. Then, the ventilation device is controlled based on the predicted traffic volume. In addition, basically the same method is used in "Ventilation equipment device for road tunnel and ventilation control method for road tunnel" disclosed in Japanese Patent Application Laid-Open No. 9-1558698, which is a well-known example of the prediction control method.

[0004] Other methods of estimating traffic volume include:
In some cases, neural network theory and fuzzy theory are applied to nonlinear traffic prediction models.

[0005]

The neural network theory and the fuzzy theory require a large amount of learning data, and when calculating the traffic volume in the ventilation control panel, the data volume becomes enormous. It takes some time to complete. In addition, using a reference pattern created from past data as a time-series model and predicting the future traffic volume from this and the measured traffic data, if the measured traffic volume is increasing, the predicted value will also increase. Become. However, the measured traffic volume includes a disturbance component, and even if the value increases, the traffic volume may decrease thereafter, and there is a problem that a prediction error is large.

SUMMARY OF THE INVENTION An object of the present invention is to provide a tunnel ventilation control device capable of determining an appropriate ventilation air volume / number calculation so as to more reliably predict a tunnel traffic volume.

[0007]

SUMMARY OF THE INVENTION To achieve the above object, the present invention provides a traffic / vehicle speed measuring means for measuring the traffic volume and vehicle speed of an up / down tunnel, and a soot concentration in the tunnel. A soot concentration measuring means for measuring the traffic volume, a traffic pattern generation storage means for storing a traffic volume measured by the traffic / vehicle speed measuring device and generating and storing a daily traffic pattern, Predicted traffic using a Kalman filter for calculating a predicted traffic volume after a predetermined time using the current traffic volume measured by the traffic volume / speed measurement unit and the traffic volume pattern stored in the traffic volume pattern generation storage unit Using the amount calculating means, the predicted traffic volume calculated by this means, the vehicle speed measured by the traffic volume / vehicle speed measuring means, and the smoke concentration measured by the soot concentration measuring means. It discloses a tunnel ventilation control apparatus being characterized in that and a air flow-volume determining means for determining a plurality of ventilation means operating any number.

Further, according to the present invention, the traffic pattern generating / storing means stores the traffic pattern on weekdays, Saturdays, Sundays,
Disclosed is a ventilation control device that is separately generated and stored for a holiday and a special day including at least a tray and a new year.

Further, the present invention is characterized in that the traffic pattern generation and storage means creates the traffic pattern based on the actual traffic in the past five weeks, and automatically updates the traffic pattern every week. A ventilation control device is disclosed.

[0010]

Embodiments of the present invention will be described below in detail with reference to the drawings. 2 and 3 are schematic diagrams showing an example of the equipment configuration of the tunnel ventilation control device according to the present invention. FIG. 2 is a sectional view and FIG. 3 is a plan view. In FIG. 1, a tunnel has a traffic / vehicle speed measuring device 1 for distinguishing between a large vehicle and a small vehicle on an up / down line and measuring the traffic volume (number) and the speed (vehicle speed) thereof, and a soot concentration measuring device 2. ,
It has a carbon monoxide concentration measuring device 3 and an anemometer 4, and the measured values are transmitted to a ventilation control device 5. The soot concentration measuring device 2 measures the soot concentration based on the light transmittance (VI value). The ventilation control device 5 determines the air volume and the number of the ventilators 6 based on these measured values to realize ventilation in the tunnel.

FIG. 1 shows the configuration of the ventilation control device 5, which comprises a traffic pattern generation / storage device 7, a Kalman filter 8, and an air flow / number of vehicles determining means 9. With this configuration,
The traffic volume pattern generation and storage device 7 captures the traffic volume measured by the traffic volume / vehicle speed measurement device 1 and divides each traffic volume into weekdays, Saturdays, Sundays, holidays, special days (Bon, New Year), and the like. On Saturday and Sunday, each traffic volume pattern is created by averaging and smoothing from the traffic volume data of the past 5 weeks. Similarly, on holidays and special days, the past traffic volume data is averaged and smoothed to create a traffic volume pattern. This traffic pattern is automatically updated by calculating every week, going back to the past 5 weeks, averaging and smoothing. As a result, a daily traffic pattern as shown in FIG. 4 is created. Such a traffic pattern is created conventionally, and it goes without saying that there are various methods other than the method described here.

Next, a prediction method using the Kalman filter 8 will be described. The input to this filter is a traffic volume pattern that differs from the actual traffic volume obtained from the traffic volume / vehicle speed measurement device 1 depending on the day of the week, a special day, and the like. This traffic pattern is automatically selected according to the day of the week or the like. Now, let YK be the actual traffic volume at time K within a day and XK be the traffic volume of the traffic volume pattern.

(Equation 1) far. Here, BK is a proportional parameter at each time K. Assuming that the two unknown variables YK and BK in (Equation 1) are state variables, the system equation is as follows, where the system error is WK.

(Equation 2) In addition, if we consider that the traffic volume is the amount to which noise (irregular fluctuation component) is added, the observation equation is

(Equation 3) Here, ZK: observation value, EK: observation error. By applying the Kalman filter algorithm to the system equation (Equation 2) and the observation equation (Equation 3), the time K
, The state variables YK and BK can be sequentially estimated. Then, assuming the estimated traffic volume YK + 6 six hours ahead,
It is calculated from the trend information YK estimated at time K and the proportional parameter BK.

(Equation 4) Is required. That is, first, the traffic volume trend information is estimated based on the Kalman filter theory, and the proportional coefficient with the traffic volume pattern is estimated, and the traffic volume trend information is extrapolated along the normalized traffic volume pattern to obtain the predicted traffic volume. I will ask.

When the predicted traffic volume is calculated by the Kalman filter 8 as described above, the air volume / number of vehicles determining means 9
Determines the number of operating ventilators according to the procedure shown in FIG. The first step is a process of determining the total number of vehicles in the tunnel in the control cycle (step 50).
1). FIG. 6 is an explanatory diagram of this processing. It is assumed that the traffic / vehicle speed measuring means 1 is located at a distance L0 from the left exit of the tunnel 10 having a length L. The time when a car traveling in the up lane enters the tunnel 10 (the right end) is tUI
N, the time when the position of the measuring device 1 is reached is t, and the vehicle speed is v,

(Equation 5) It is. Similarly, the car running in the down lane is tunnel 1
Assuming that the time at which the vehicle enters 0 (the left end) is tDIN, the time at which the position of the measuring device 1 is reached is t, and the vehicle speed is v,

(Equation 6) Then, assuming the measured traffic volume for 5 minutes ascending and descending to NU and ND respectively, the control cycle is set to 1 minute and the number of vehicles n in the tunnel during that time is

(Equation 7) Given by Where tUIN and tDIN are (Equation 5) and (Equation 6)
And t10 is the time required to pass through a tunnel of length L at vehicle speed v;

(Equation 8) It is assumed that the owned number n in (Equation 7) is used separately for large vehicles and small vehicles, and the corresponding owned numbers are calculated as nL large vehicles and ns small vehicles, respectively.

Next, the amount of soot generated in the tunnel is determined from the determined number n of items held in the tunnel (step 502). Now μ ': average soot concentration in exhaust gas, ki: slope correction coefficient,
kh: altitude correction coefficient, VG: average exhaust gas amount, T: control cycle (5 minutes), Q: soot generation amount,
If small cars with different values are added with subscripts L and S respectively,

(Equation 9) It is represented by However

(Equation 10) It is.

When the amount Q of soot generated during one control cycle T in the tunnel is obtained, the target ventilation amount q and the target tunnel wind speed vr are obtained using the obtained amount (step 50).
3, 504). The soot concentration measuring device obtains the concentration from the light transmittance VI in the tunnel, and the relationship between them is given by a logarithmic function. This relationship is expressed by the respective target values, that is, the relationship between the target soot concentration dd0 and the target VI value VI0.

[Equation 11] It is. Then, from the target soot concentration dd given by this equation and the soot generation amount Q obtained by (Equation 9), the target ventilation amount q per unit time is

(Equation 12) Is required. The target tunnel wind velocity vr required to realize this ventilation q is given by

(Equation 13) Becomes

Next, the natural ventilation power at the present time is obtained (steps 505 to 508). Assuming that the mass of air in the tunnel is M, the ventilation force Ft in the tunnel is obtained from the equation of motion.

[Equation 14] In the steady state, the left side is 0, so that Ft = 0
Becomes On the other hand, if the ventilation force Ft in the tunnel is Fn: natural ventilation force, Ff: pipe friction force, Fp: traffic ventilation force (due to vehicle running), and Fj: mechanical ventilation force

(Equation 15) Therefore, by setting Ft = 0 in the steady state, the natural ventilation force Fn becomes

(Equation 16) Given by On this right side, the traffic ventilation force Fp is obtained as follows. That is, ρ: air density, Ae: traffic cross section, Vav: average vehicle speed during one control cycle T, A
Since e and Vav are both given different values in the upward and downward directions, they are distinguished by adding subscripts U and D to them. Then

[Equation 17] This is determined in step 505.

The mechanical ventilation force Fj is given by: ηj: Fan boosting efficiency,
Vj: Fan outlet wind speed, Aj: Fan outlet area, n
j: Number of fans operated

(Equation 18) The pipe friction force Ff is λ: pipe friction coefficient, and D: tunnel representative diameter.

[Equation 19] Becomes The calculations of (Equation 18) and (Equation 19) are obtained in steps 506 and 507, and subsequently the natural ventilation force Fn at the present time is calculated in step 508.

The above is the calculation of the various ventilation powers at the present time, but the relationship of (Equation 16) is always satisfied as long as the steady state (Ft = 0) is considered. Therefore, it can be considered that (Equation 16) holds for the predicted values of the respective amounts several cycles ahead of the control cycle, for example, six cycles ahead. Therefore, the predicted value Fjf of the mechanical ventilation required several cycles ahead
Can be expressed by adding a subscript f to the predicted value of each ventilation

(Equation 20) Becomes As the predicted natural ventilation force Fnf on the right side, the Fn value obtained in step 508 is used assuming that the value at the current time continues as it is. Further, the predicted traffic ventilation force Fpf is obtained using the predicted traffic volume predicted by the Kalman filter 8 in place of the ascending and descending current traffic volumes nU and nD in (Equation 17) (step 509). The predicted pipe friction force Fff is (Equation 19)
May be used as is. Thus, from (Equation 20), the predicted mechanical ventilation force Fjf several cycles ahead can be calculated from (Equation 20) (Step 510).

The predicted mechanical ventilation force Fjf thus required
Is obtained, the number of operating units njf of the ventilator 6 required to realize this mechanical ventilation force is obtained from (Equation 18).

(Equation 21) (Step 511). Therefore, how many of the ventilators are to be operated is determined, and the operation control of the ventilators is actually performed according to this value (step 512). In this way, by controlling the operation of the ventilator by predicting the state in the tunnel several cycles ahead of the control cycle, the problem of control delay is solved, and the disturbance component generated in the actual traffic volume by using the Kalman filter , It is possible to realize an accurate prediction and, therefore, a more accurate exhaust control. In the operation control of the ventilator in step 512, it is preferable to select a unit performing the ventilation operation so that the operation time of each ventilator is averaged.

[0020]

According to the present invention, the traffic volume can be predicted without being disturbed by the fluctuation of the primary measurement value, and the ventilation operation with higher performance can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration example of a tunnel ventilation control device according to the present invention.

FIG. 2 is a cross-sectional view schematically illustrating an example of a facility arrangement of a tunnel ventilation device according to the present invention.

FIG. 3 is a plan view schematically showing an example of a facility arrangement of a tunnel ventilation device according to the present invention.

FIG. 4 is a diagram illustrating an example of a traffic pattern;

FIG. 5 is a flowchart showing the operation of the apparatus of FIG. 1;

FIG. 6 is an explanatory diagram of detecting the number of vehicles owned in a tunnel.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Traffic volume and vehicle speed measuring device 2 Smoke concentration measuring device 3 Carbon monoxide concentration measuring device 4 Wind direction anemometer 5 Ventilation control device 6 Ventilator 7 Traffic volume pattern generation storage device 8 Kalman filter 9 Air volume / number determining means

Claims (3)

[Claims] 1. A traffic volume / vehicle speed measuring means for measuring traffic volume and vehicle speed going up and down a tunnel, a smoke concentration measuring device for measuring smoke concentration in a tunnel, and the traffic volume / vehicle speed measurement Means for generating and storing a daily traffic pattern by storing the traffic measured by the means; and a current traffic measured by the traffic / vehicle speed measuring means and the traffic. A predicted traffic volume calculation unit using a Kalman filter for calculating a predicted traffic volume after a predetermined time using the traffic volume pattern stored in the pattern generation storage unit, and a predicted traffic volume calculated by the unit; Traffic volume·
A vehicle speed measured by a vehicle speed measuring means, and an air volume / number determining means for determining how many ventilation units are to be operated by using the soot concentration measured by the soot concentration measuring means; A tunnel ventilation control device. 2. The traffic pattern generating and storing means separately generates and stores the traffic pattern for weekdays, Saturdays, Sundays, holidays, and at least special days including at least Bon and New Year. The ventilation control device according to claim 1. 3. The traffic volume pattern generation and storage means creates the traffic volume pattern based on actual traffic volume in the past five weeks, and automatically updates the traffic volume pattern on a weekly basis. The ventilation control device according to item 1.