This is a self-archived – parallel published version of this article in the publication archive of the University of Vaasa. It might differ from the original. Design and Implementation of a Wireless Automation Module for Diesel Engines Author(s): Siemuri, Akpojoto; Glocker, Tobias; Mekkanen, Mike; Kauhaniemi, Kimmo; Mantere, Timo; Rösgren, Jonatan; Kuusisto , Jari; Elmusrati, Mohammed Title: Design and Implementation of a Wireless Automation Module for Diesel Engines Year: 2019 Version: Accepted manuscript Copyright ©2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Please cite the original version: Siemuri, A., Glocker, T., Mekkanen, M., Kauhaniemi, K., Mantere, T., Rösgren, J., Kuusisto, J. & Elmusrati, M. (2019). Design and Implementation of a Wireless Automation Module for Diesel Engines. In 2019 27th Telecommunications Forum (TELFOR), 1-4. https://doi.org/10.1109/TELFOR48224.2019.8971093 Abstract — This paper describes the design of wireless CANprotocol with the aim to replace existing wired CAN protocolcommunication between the Smart NOx sensor on dieselengines and the Engine Control Unit (ECU). Wirelessindustrial automation becomes essential in modern industrywith the emerge of 5G networks and the IoT. Besides theadvantages of wireless nodes in automation systems, there aremany real challenges. Several practical design challenges havebeen successfully addressed in this paper.Keywords — CAN Protocol, Engine Control Module(ECM), Smart NOx sensor, Speedgoat, WirelessCommunication, ZigBee. I. INTRODUCTION odern industry has developed rapidly, fueled by anincreasing growth in global economy. However, datacollection, analysis, and integration has become an essentialpillar for the new industrial structure. It is extremelyimportant to have real-time information in automationsystems on all levels. One of the first steps in wirelessautomation, is to decide the required wirelesscommunication protocol for certain automation system. Thedecision should be based on the automation requirementssuch as: latency, data rate, coverage distance, reliability (interms of outage and packet losses), costs, security, etc.Therefore, studying of some well-known wirelesscommunication solutions is crucial in achieving reliable andflexible data transfer [1].The research presented in this paper investigates thefeasibility of implementing a reliable, secure and faulttolerant wireless communication between the Smart NOxsensor on diesel engines and the Speedgoat (real-time rapidprototyping tool) or Engine Control Module (ECM). TheSmart NOx sensor is connected to the Engine Control Unit(ECU) with a wired CAN bus connection. Data istransmitted using SAE J1939 protocol which is built on topof CAN Networks. SAE J1939 is developed specifically foruse in heavy duty environments, with an emphasize onachieving reliable and fault tolerant communication. Theapproach taken in this research, is based on a case study ofWärtsilä’s Smart NOx sensor on a W4L20 diesel enginewith the objective, to replace the wired CAN bus withwireless node. The design steps, implementation andperformance analysis are discussed in next sections. A. Smart NOx sensor The smart NOx is a sensor that measures the oxygen (O2) percentage and nitrogen oxides (NOx) ppm in the exhaustof combustion engines. Oxygen is measured as apercentage, while the NOx concentration is measured inparts per million (ppm) [2].Nitrogen Oxides (NOx) are a group of poisonous, highlyreactive gases of which two occur naturally namely nitricoxide (NO) and nitrogen dioxide (NO2). The combustion offossil fuels is the most common source of NOx emissions.The amount of emission depends on the air-fuel mix ratioas well as the amount of nitrogen in the fuel. At hightemperatures and conditions that encourage oxidation NOxformation in combustion is favored. NO2 has adverseeffects on human health and at high concentrations it canlead to the inflammation of the airways [3]. B. Speedgoat and Engine Control Module (ECM) The speedgoat applies Real-time systems with SimulinkReal-Time™ from MathWorks to various applicationsacross many industries, in the lab, field, classroom, orembedded in machinery. Speedgoat solutions and Simulinkare seamlessly integrated and allows for fast test run ofSimulink software designs with hardware [4].The Engine Control Module (ECM), also called EngineControl Unit (ECU), is a kind of electronic control unit thatmanages the control of series of actuators on an internalcombustion engine to ensure that the engine’s performanceis optimal. This is done by reading the values from all thesensors within the engine bay and interpreting the datausing multidimensional performance maps (referred to aslookup tables) and adjusting the engine actuatorsaccordingly [5]. The diesel engine used in this research is amedium speed W4L20 diesel engine. The engine producesapproximately 1 MW of power and it is paired with ABBgenerator [6]. This combustion engine is in the VaasaEnergy Business Innovation Center (VEBIC) laboratory.VEBIC is a new research and innovation platform hostedby the University of Vaasa. The VEBIC environment hastwo laboratories namely, the internal combustion enginelaboratory and a separate but related fuel developmentlaboratory. It also has a program for energy and sustainabledevelopment research projects [7]. II. CAN PROTOCOL AND THE WIRELESS MODULE In this section, the required wireless module is discussed.The selection is based on the automation requirements andthe available wireless standards. Design and Implementation of a WirelessAutomation Module for Diesel Engines Akpojoto Siemuri, Student Member, IEEE, Tobias Glocker, Mike Mekkanen, Kimmo Kauhaniemi,Member, IEEE, Timo Mantere, Jonatan Rösgren, Jari Kuusisto and Mohammed Elmusrati, SeniorMember, IEEE M A. Controller Area Network (CAN) CAN is a solution for automation industries and the CANprotocol is used in systems that need to transmit and receivea small amount of data with real-time requirements. CANbus was originally developed for the car industry to replacepoint to point connections in automotive systems. CANprotocol has been stipulated as international standard by150 International Standard Organizations [8]. CANtransmits signals on the CAN network using two wires,CAN-High and CAN-Low. These two wires operate indifferential mode carrying inverted voltages whichdecreases noise interference. The standard being useddetermines the voltage level and other characteristics of thephysical layer. The two standards are the ISO11898 (CANHigh Speed) standard and the ISO11519 (CAN Low Speed)standard [9]. B. Wireless Communication Protocols Wireless applications typically require bursttransmission, reduced overhead, and they use a very smallamount of data per node, therefore, the bandwidth is not themain requirement. Some applications require coverage oflarge areas; reliability, availability, bounded latency forreal-time behavior and energy efficiency as some keyperformance indicators [10].The XBee module used during implementation uses theZigBee (IEEE 802.15.4) technology. ZigBee is a short-range wireless protocol which is a standard for personal-area networks developed by ZigBee Alliance aiming atproviding a low cost, low power consumption, reliable andtwo-way wireless communication standard for short rangeapplications. It allows the nodes to find new routesthroughout the network when one route fails. Thus, Zigbeeis a robust wireless solution [11].The choice of using the ZigBee protocol was made fromcomparing four wireless solutions based on the analysis ofexperiment results. The wireless protocols compared werenamely WiFi, LoRa, BLE and Zigbee. Performanceanalysis for these four wireless protocols was conductedbased on key considerations that should influence thechoice of wireless protocols for a specific application. Theexperiments and measurements were performed inTechnobothnia laboratory. Technobothnia is a wide rangedadvanced and modern laboratory unit that occupies 8000m2 and which is within the campus of University of Vaasa [12].Furthermore, the experiments related to the engine has beencarried out in the Engine test room located in VEBIC.From the experiments, the Xbee had a better RSSI valuesand security feature over the other wireless modules used.These features with some other factors like, goodperformance in the packet loss test; ability to enhance thebattery life; better penetrating capability and range whencompared to the BLE leads to the choice of implementingthe ZigBee wireless protocol in designing the wireless-CAN protocol called the Xbee-CAN module. III. SYSTEM ARCHITECTURE The system consists of a 24V power supply for the smartNOx sensor, which is connected to the CAN Bus of thewireless-CAN module (transmitter). Furthermore, thewireless-CAN modules (receiver) is connected to thespeedgoat, the speedgoat device that contains a MATLAB Simulink model in order to calculate, monitor, and displaythe O2 % and NOx ppm. A. The Xbee-CAN bridge Hardware The setup (see in Fig. 1 and 2) provides a proof-of-concept that can be further developed from a prototype intoa product.At the transmitter side, a Multiprotocol Radio Shield isconnected over the Arduino board and the CAN Busmodule is placed in socket 0 of the Multiprotocol RadioShield while the Xbee module is placed in socket 1. TheCAN Bus module is used to interface the transmitter Xbeemodule with the smart NOx sensor using twisted pair cables(CAN High and CAN Low). Fig. 1. Block diagram for the hardware setup of Xbee-CAN bridge Transmitter. Fig. 2. Block diagram for the hardware setup of Xbee-CAN bridge Receiver. At the receiver side, a Multiprotocol Radio Shield isconnected over the Arduino board and the CAN Busmodule is placed in socket 0 of the Multiprotocol RadioShield while the Xbee module is placed in socket 1. TheCAN Bus module is used to interface the receiver Xbeemodule with the speedgoat device using twisted pair cables(CAN High and CAN Low). B. The Xbee-CAN bridge Software The programming of the hardware devices was doneusing the Arduino IDE environment.At the transmitter side, the smart NOx sensor has a 29-bit CAN ID and the transmitter hardware is programmed tosend an initialization heating signal “00000004h” (8 byteshexadecimal) through the CAN Bus to this CAN ID to startheating and collecting the smart NOx data. The data isreceived through the CAN Bus and then transferred throughSPI to the Xbee for wireless transmission. While at thereceiver side, the hardware is programmed to receive thesmart NOx sensor data and transfers the data through SPI tothe CAN Bus of the receiver module. The CAN Bus isconnected to the speedgoat using two twisted pair cable(CAN High and CAN Low) for analyzing the received data.The receiver module also performs packet loss and RSSImeasurements. The flowcharts for the programming of the transmitterand receiver modules are illustrated in Fig. 3 and 4respectively. Fig. 3. Flowchart for the Xbee transmitter codes. Fig. 4. Flowchart for the Xbee receiver codes. C. Details of Transmitted Payload The transmitted payload is a 10-byte hexadecimal datacomprising of 1-byte preamble, 8-byte smart NOx data and1-byte checksum data as illustrated in Fig. 5. Fig. 5. Transmitted Smart NOx Payload. Equation (1) is the checksum and it is the sum of the 8-byte smart NOx data only. + 0ݔܱܰ + 1ݔܱܰ + 2ݔܱܰ … + 7ݔܱܰ = ܿℎ݁ܿ݇0ݔܱܰ݉ݑݏ + + 1ݔܱܰ + 2ݔܱܰ … + = 7ݔܱܰ ܿℎ݁ܿ݇(1) ݉ݑݏ The preamble is computed as illustrated in Equation (2).The preamble is computed from the sum of the 8-byte smartNOx data plus the checksum value, that is, smartNOx[8] +checksum[1] = preamble[1]. 0ݔܱܰ + 1ݔܱܰ + 2ݔܱܰ + … + 7ݔܱܰ + ܿℎ݁ܿ݇݉ݑݏ = (2)݈ܾ݁݉ܽ݁ݎ݌ Both, the checksum and preamble are used to verify dataintegrity, that ensures an error free data transmission andprevents the alteration of data. IV. PERFORMANCE TEST OF THE DESIGNED XBEE-CAN MODULE ON WÄRTSILÄ’S W4L20 DIESEL ENGINE The smart NOx sensor was installed on the Wärtsilä 4L20Diesel Engine and the O2% and NOx ppm values weremeasured and compared with the readings from theSICK|MCS100E. The MCS100E HW is an analyzer systemused for extractive measurement of up to eight (8) activegas components from an engine [13]. Table 1 illustrates thecomparison between the values from MCS100E andwireless Xbee-CAN prototype seen on the speedgoat devicefor the W4L20 diesel engine in different operation modes(engine is idle, running without load and running with load).It is the percentage error of the Xbee-CAN moduleprototype values when compared to the values from theMCS100E. The measurements from MCS100E and theXbee-CAN module were performed at the same time. Thedifference between their readings in Table 1 comes from thefact that both had its own separate sensor (installed ondifferent location on the exhaust of the same engine),therefore, their measurement time could not be properlysynchronized. However, it was concluded from observationthat the values obtained were close to what was expectedfrom the engine test for the Xbee-CAN module and theMCS100E (as a reference). TABLE 1: COMPARISON OF THE VALUES FROM SICK|MCS 100E AND THE XBEE-CAN MODULE CONNECTED TO A SMART NOX SENSOR. MeasurementDevice Engine Operation ModesEngine isIdle Enginerunningwithout load Enginerunning withloadNOxppm O2 % NOxppm O2 % NOxppm O2 % Xbee-CANmodulepercentageerror (%) 0.0 0.0 3.09 7.22 3.83 2.06 In the present application (the engine test phase), thedelay noticed in the wireless CAN (Xbee-CAN module) ascompared to the wired CAN network, is still within theacceptable delay range.For the Xbee-CAN module, Fig. 6 gives the sliding graphof the O2% and NOx ppm values from the speedgoatmonitoring tool. Fig. 6. Speedgoat sliding graph of the O2% and NOx ppmvalues. V. APPLYING ADDITIVE MANUFACTURING TO THE DESIGNED PROTOTYPE Additive Manufacturing is applied in this project todesign the protective casing for the designed Xbee-CANReceiver/Transmitter. Two protective casings weredesigned, and 3D printed to encapsulate the receiver and transmitter XBee-CAN Modules prototypes. The installedXbee-CAN Receiver/Transmitter in the 3D printedprotective casings are illustrated in Fig. 7. Fig. 7. Wireless Xbee-CAN prototype in 3D printedcasing. The Material used in printing the protective casings isPLA filament. The 3D printers used for the printing of theprototype casing are the PRUSA and MINIFactory. Bothused the following settings during printing: printingtemperature of 200℃ for the extruder and 60℃ for the bedplate and an infill of 20%. VI. CONCLUSION Research on the feasibility of replacing the existing wiredCAN bus connection between the smart NOx sensor andSpeedgoat and possibly in the future the Engine ControlUnit (ECU) with a wireless communication solution wasperformed. The devices have been designed to meet therequired application in line with the performanceexpectation for communication in an industrialenvironment. The designed wireless Xbee-CAN prototypeis currently installed and running on a W4L20 diesel enginein VEBIC. Possible future work involves integrating FieldProgrammable Gate Arrays (FPGAs) to the device. Themost important motivation is to optimize the software toimprove the battery life of the current prototype byoptimizing all parameters related to data rates, power andenergy consumption. ACKNOWLEDGMENT Special thanks to Reino Virrankoski for the opportunityto work on this project and thanks to Rayko Toshev forgranting access to the digital manufacturing laboratory inTechnobothnia, University of Vaasa to achieve the 3Dprinting aspect of this project. REFERENCES [1] G. Xiang, D. Huang, Y. Chen, W. Jin and Y. Luo,"The Design of a Distributed Control System Basedon CAN bus," in Proceedings of 2013 IEEE,International Conference on Mechatonic andAutomation, August 4 – 7, Takamatsu, Japan., 2013. [2] I. Senft and L. Bertrand, "Specification Smart NOxSensor - Uninox24V," Wärtsilä Oyj Abp, 2010. [3] "Nitrogen oxides (NOx) emissions," 2018. [Online].Available: https://www.eea.europa.eu/data-and-maps/indicators/eea-32-nitrogen-oxides-nox-emissions-1. [4] "Applications and Industries Overview," 2007-18.[Online]. Available:https://www.speedgoat.com/applications-industries. [5] "Wikipedia - Electronic Diesel Control," 2018.[Online]. Available: https://en.wikipedia.org/wiki/Electronic_Diesel_Control. [6] K. Antti, J. Marko, K. Teemu and F. Tero, "W4L20VEBIC Genset dynamics—baseframe design,"Journal of Structural Mechanics, vol. 50, p. 292,2017. [7] "VEBIC - Vaasa Energy Business InnovationCentre," 23 10 2018. [Online]. Available:https://www.univaasa.fi/fi/sites/vebic/. [8] W. Xiao-feng, X. Yi-si and C. Li-xiang, "Applicationand Implementation of CAN Bus Technology inIndustry Real-time Data Communication.," inProceedings of 2009 IEEE, International Conferenceon Industrial Mechatronics and Automation(ICIMA), 2009. [9] S. Nilsson, "Controller Area Network - CANInformation.," 2018. [Online]. Available:http://hem.bredband.net/stafni/developer/CAN.htm. [10] A. Khan and K. Turowski, "A survey of currentchallenges in manufacturing industry and preparationfor industry 4.0," 2016. [Online]. Available:https://www.springer.com/cda/content/document/cda_downloaddocument/9783319336084-c2.pdf?SGWID=0-0-45-1564143-p179960859. [11] "ZigBee – Technical documents," 2013. [Online].Available: http://www.ti.com/wireless-connectivity/simplelink-solutions/zigbee/technical-documents.html. [12] "Technobothnia," 2019. [Online]. Available:https://www.technobothnia.fi/. [13] "CEMS solutions MCS100E HW," 2019. [Online].Available: https://www.sick.com/fi/en/analyzer-solutions/cems-solutions/mcs100e-hw/c/g285463. I. Introduction A. Smart NOx sensor B. Speedgoat and Engine Control Module (ECM) II. Can Protocol and the Wireless Module A. Controller Area Network (CAN) B. Wireless Communication Protocols III. System Architecture A. The Xbee-CAN bridge Hardware B. The Xbee-CAN bridge Software C. Details of Transmitted Payload IV. Performance test of the designed xbee-can module on wärtsilä’s w4l20 diesel engine V. Applying additive manufacturing to the designed prototype VI. Conclusion Acknowledgment References