TRC1801 Evaluation of WIM AutoCalibration Practices and Parameters - - PowerPoint PPT Presentation

TRC1801 Evaluation of WIM Auto‐Calibration Practices and Parameters

1

Sarah Hernandez Assistant Professor Department of Civil Engineering University of Arkansas

1. Evaluate the Weigh‐in‐Motion (WIM) auto‐calibration practices used by ARDOT. 2. Recommend effective auto‐calibration practices.

Key Objectives

2

Fayetteville WIM Site on I‐49

Task Overview

Task 1 Review WIM Auto‐ Calibration Methods Task 2 Field Data Collection Task 3 Calibration Analyses Task 4 Implementation Activities

3

We are here…

1. Weigh‐in‐Motion (WIM) Data and Uses 2. WIM Hardware 3. Calibration and Auto‐ Calibration Practices 4. Data Collection and Preliminary Findings

Presentation Outline

4

Truck traversing load cell WIM sensor

(Ref: LTBP Review of WIM, 2016)

FHWA Classification Scheme (Ref: TxDOT, 2017)

1.Long term pavement performance (LTPP) 2.Pavement design and maintenance (MEPDG) 3.Bridge design 4.Monitoring highway usage and funding allocation 5.Truck size and weight enforcement

Weigh‐in‐Motion (WIM) Data and Uses

5

Quartz piezoelectric sensor with inductive loops

WIM Hardware: Configuration

6

Staggered bending plate installation with inductive loops

(Ref: WIM Data Analysists Manual)

WIM Hardware: Types

7

Sensor Type Initial Cost Performance Sensitivity Life Span

Bending Plate $20,000 +/‐ 5% Minor 6 years Piezoelectric Sensor $9,000 +/‐ 10% Temperature 4 years

(Ref: Zhang, L., Evaluation of the Technical and Economic Performance of WIM Sensing Technologies, 2015; FHWA State Practices in WIM, 1997)

WIM Site Locations in Arkansas

8

Functional Class Number of WIM Stations Interstates 11 Freeways 4 Principal Arterials 11 Minor Collector 3 Major Collector 5 Total 34

‐60 ‐40 ‐20 20 40 60 80 100 ‐60 ‐40 ‐20 20 40 60 Percent Error in ESAL Value (%) Calibration Drift (Percent Error, %)

Effects of Calibration Drift on ESAL Values (Ref: FHWA, 1998)

Effects of Temperature Effects of Wear and Tear

Effects of Poor Calibration

9

Calibration drift shown as shifts in Gross Vehicle Weight (GVW) shifts

Name Issues Method

Test Trucks

• Most accurate
• Expensive
• Time consuming and labor intensive

Traffic Trucks

• Must have high volume of trucks

crossing WIM and static scales

• Time consuming

Auto‐Calibration

• Works well for sites with heavy

tractor‐trailer volumes

• Low cost and efficient

WIM Calibration and Auto‐Calibration

10

WIM WIM

Static Scale Traffic Test Trucks Traffic

WIM

Results from NCHRP Synthesis study “High Speed Weigh in Motion System Calibration” (2008)

State of the Practice of Calibration Methods

11

Which calibration method is used? What type of sensors is autocalibration used for?

Test Trucks Traffic Trucks Data QC Other Traffic Data 67% 33% 57% 18% Enforcement 6% 59% 29% 6% Both 33% 22% 39% 5% Type I Type II Traffic Data 14% 75% Enforcement 50% 50% Both 0% 100%

ARDOT Auto‐Calibration Procedure

12

Lane Type of Vehicle Reference Axle Target Value

• No. of

Vehicles

1 (outer lane) Five‐axle semi‐tractor trailers Front axle 10.2 kips 50 2 (inner lane) Passenger Cars Front axle 2.2 kips 50 3 (inner lane) Passenger Cars Front axle 2.3 kips 75 4 (outer lane) Five‐axle semi‐tractor trailers Front axle 10.8 kips 25

(Ref: Correspondence with Peek Traffic)

MnDOT Auto‐Calibration Algorithm

13

(Ref: NCHRP Synthesis study “High Speed Weigh in Motion System Calibration”, 2008)

Front Axle Weight Reference values stratified by GVW

Adjustment Factors Based

• n Size of Data Sample

Gross Vehicle Weight (lbs) Desired Front Axle Weight

< 32,000 8,500 32,000 – 70,000 9,300 > 70,000 10,400 Percent Deviation Allowed 3.5%

• Min. Hours Monitored

48

• Min. Class 9 Trucks Weighed

250

• Truck GPS data

available from national “vendors”.

• Partnered with
• GPS sample used to

track same truck across multiple WIM sites.

Auto‐Calibration with Truck GPS Data

14

US Coverage Map of Weigh Stations and Agricultural Inspection Stations with Drivewyze coverage

GPS Truck Tracking at WIM Sites

15

January February

Recommended sites based Drivewyze truck GPS: 1.Lamar EB (360009) 2.Lonoke EB (430037) 3.Bald Knob NB (730068) 4.Dardanelle SB (750006) 5.Glen Rose SB (301769)

Field Data Collection

16

Lamar Dardanelle Glen Rose Bald Knob Lonoke Alma

Field Data Collection: Procedures

17

Photo and Static Scale Measurements at Alma Video and WIM Measurements at Lonoke

Record ID Date Time Array Vehicle No Class Speed‐ mph #‐ Axles 7213 03‐15‐2018 15:42:52 Lane 1 107 6 71 3 7214 03‐15‐2018 15:42:59 Lane 2 198 8 75 4 7215 03‐15‐2018 15:43:40 Lane 1 113 9 68 5

Field Data Collection: Preliminary Findings

18

Field Data Collection: WIM Weight Distributions

19

Lonoke Lamar

Field Data Collection: Static Weight Differences

20

Lonoke Lamar

Front Axle Weight Front Axle Weight

• Continue to process field data  Alma WIM scale data
• Development of traffic stream auto‐calibration parameters
• Development of GPS based auto‐calibration

Next Steps

21

Task 1 Review WIM Auto‐ Calibration Methods Task 2 Field Data Collection Task 3 Calibration Analyses Task 4 Implementation Activities

Sarah Hernandez Assistant Professor Department of Civil Engineering University of Arkansas sarahvh@uark.edu

Questions?

22