sept2024_thumbnail_cover

Beyond lighting installation...IoT structure

By Ir Peter Y WONG

If you choose to listen to this article, you are welcome to download the PDF version of the Journal (September 2024 issue) and activate the “Read Out Loud” function in Adobe Reader. For more details, please read the user's note.

 

“With sensors of different natures connected to the wireless modules installed in lighting fittings for onward communication with the master controlling programme, an intelligent lighting installation becomes the backbone for other intelligent provisions of performance.”

 

From Best design practice: Lighting installation and beyond - IoT structure (Cover Story of Hong Kong Engineer’s June 2023 issue1)

 

 

In 1989, a toaster connected to internet, and capable of remote operations, was considered as the first Internet of Things device.2 In 1999, the computer scientist Kevin Ashton “invented” the term IoT with Radio-frequency identification (RFID) chips in tracking products along the supply chain.3

 

Self Photos / Files - c1

The first IoT device was created in 1989 and it is not what you’d expect; John Romkey connected his Sunbeam Deluxe Automatic Radiant Control toaster to the internet via a Internet protocol suite (TCP/IP), making a toaster the world’s first ever IoT device

 

The Internet and the Things

 

The Internet

 

What we think and what we do change the world. With the Internet, we can amass huge amounts of information for thinking. With the Internet, we can replicate what we wish to do in distance. For the past decades, the Internet has been changing the world insofar as it changes the way we think and exchange ideas with others. Or, the computers?

 

The Thing

 

It is a device—a sensor in most cases—with self-processing capability. The logic of the device controlling the automatic doors at the shopping arcades is one of “The Things”. Powered by electricity, the automatic doors that open for people to pass through and stop when the built-in logic so decides was invented in 1931. But the concept goes all the way back to the 1st century, when the Greek-Egyptian mathematician and engineer Heron of Alexandria invented a rope-connected hydraulic system powered by thermal energy and gravity to open and close temple doors.4

 

Self Photos / Files - c2

Heron of Alexandria – Automated temple doors (© artfacts-berlin.de)

 

Building Information Performance (BIP)

 

The Internet is a global network. But we only need a local communication network within the building to control the performance of building services provisions, equipment and estates management systems to pursue our goals in efficacy and well-being. It is not Building Information Modelling (BIM) in telling you where they actually were; it is BIP in telling you how they are actually performing now.

 

A software is only as smart as what we tell it to do. The software needs abundant information to process and execute what we had told the software of what we wish it to perform under the circumstances, in the distance.

 

Nowadays, the controlling mechanism of many equipment systems is a variant model of the Programmable Logic Controller developed in the late 1960s to replace the hard-wired relay-logic along the same ladder-logic practice. The sequential performance works fine for individual system with “output results” in response to “input conditions”. We do not need to interfere with the manufacturers’ expertise in equipment design and operation; we only need to provide more data as “input conditions” for the system to promptly execute “output results” as what we wish the software to perform, in the distance.

 

With the exploding developments of wireless personal area network amidst competing and conflicting factors of reliability, connectivity, security, power consumption, speed of operation and cost since, the bandwagon of wireless mesh network cum IoT is the future solution for building automation systems.

 

The Building Services Branch of the Architectural Services Department (ArchSD) had taken a harbinger role in specifying the ZigBee protocol in one of its projects and the Electrical and Mechanical Services Department (EMSD) is examining the interoperability of building services systems by software programme.

 

In line with ArchSD specification, the characteristics and properties of the ZigBee protocol were quoted below as exemplars. Notwithstanding this, the ultimate template should be one based on Low-power Wireless Personal Area Networks for a more universal and perpetual specification. The analysis below supports why wireless mesh network cum IoT is the future solution for building automation systems. To EMSD’s quest for coordinated cross systems performance, the illustrations below show how cluster control and firmware enabled by wireless mesh network cum IoT, interoperability of routine and systems-wise can be realised on the spot and in real time, without a master software programme.

 

 

Wireless mesh network cum IoT is the future solution for building automation systems.

 

 

First and foremost, the lighting installation with wide footprints in both horizontal and vertical dimensions, forming an infrastructure of wireless network, provides ready power and positions for other building services provisions, equipment and estates management systems to locate monitoring/actuating devices, indoor and outdoor, in enhancing respective system’s performance. By coordinating inputs from various systems within a defined zone, the optimal performance across services is achieved - at the area where the well-being of occupants and ambience is important and relevant.

 

Building Management System (BMS) software programme

 

With advancing technological developments in building services systems and the ever-present demands for efficiency, carbon neutrality and well-being by society, the current practice is to engage a master software, a BMS, to coordinate system-wise performance across building services provisions, equipment and estates management systems. A proprietary BMS software is often costly, but if desirable deliverables of building performance were clearly and explicitly delineated, typical platforms with satisfactory performance would emerge over time, driving down the costs. This had already happened in Hong Kong with independently run restaurants already offering a complex menu with entries capable to take on board the customers’ “well-being” choices. At present, most if not all BMS systems installed are only for dashboard display as an accolade of “No human inputs, No computer outputs.”

 

Self Photos / Files - c3

Typical BMS governance control - operator interventions required

 

This article demonstrates that local automation and on-field controls can supplement and enhance the original designs in real-time applications for each building services provision, equipment and estates management system without involving the BMS, and more importantly, in rendering optimal performances across all systems within the same area and ambience where it is most relevant in the quest of carbon reduction and ambient well-being.

 

 

Local automation and on-field controls can supplement and enhance the original designs...in rendering optimal performances across all systems within the same area and ambience where it is most relevant in the quest of carbon reduction and ambient well-being.

 

 

Wireless mesh network and IoT

 

Internet of Things (IoT) is only about a device with built-in processing capability. The common perception that IoT must be connected to the internet is incorrect. For building management, it is more a matter of an intranet.

 

An apparatus (namely, Wireless module) with a communication chip is required to transmit data and receive instructions. With a logic (here, an algorithm) burnt in the chip to provide the processing capability, the assembly of the Wireless module connected to a device like a sensor becomes an IoT—a Node. The algorithm allows using sensors of the simplest and most basic type as RS485, which suffices as an IoT for the purpose of systems control.

 

The configuration of data and information carried by an IoT follows a protocol which is different from that of the communication over which the data was transmitted. All the Nodes connected within range form a wireless mesh network. A router-node transmits all the data of the Nodes within the mesh network to the particular system’s desktop computer, or a master software, and onto the file server. Monitoring devices of all building services provisions, equipment and estates management systems are connected to the wireless mesh network to share and receive data information for prompt control.

 

Self Photos / Files - c4

Wireless mesh network governance control - extraneous data driven

 

Wireless mesh network is not a panacea of total solution for every system as control of certain equipment systems of linear nature, like metering, piping and arguably lift installations, works fine with fixed-line and hardware logic. The advantages of, at least in augmenting, using a wireless mesh network with IoT over the traditional fixed-wired with hardware logic installations are summarised below:

 

Data traffic

 

  1. Data travelling in opposite directions, along both hard-wired and wireless mesh network, must clash. Some hard-wired signal control systems had defined slots for monitored data reports. Some chose the “send and forget” arrangement. Both approaches defeat the purpose of upholding optimal and effective building management at all times. On the other hand, a wireless mesh network with multiple routings between Nodes provides numerous alternate paths with milliseconds apart in traffic to increase the odds of successful transmissions. Since data are not necessarily travelling head-on in a wireless mesh network at the same juncture as well as along the same path between Nodes, the statistic of head-on clash between two Nodes is much reduced. The worst scenario is with broadcast messages to all Nodes. Even in that scenario, the range of error of broadcasting is usually about five to ten failures per one million transmissions. The statistic of the number of Nodes active during broadcast can be further addressed by sending a package of a desist-to-report command preceding the one that contains data. Repeating the same package once is often adequate. But the designer and client can always stipulate or later amend the frequencies according to the weight of broadcast messages. Of course, a scheduled time for broadcast only can always be set, like the first message of the day in aligning the clock.
  2. Nodes are also programmed to acknowledge data received to ensure prompt delivery. Where all else fails, Nodes of other systems can pick up erratic performance from a lone system to adjust each respective response until abnormality finally subsides by the self-healing effect within the confined area – only if the sensor nodes of all systems were connected to the wireless mesh network. Some systems do provide safeguards against abnormal system performance from erratic signal, usually more for precautious measures at system-wise level. It could be difficult for a system alone to identify self-initiated commands of unusual performance.

 

Data integrity

 

Data is encoded and processed. It has to be interpreted, managed or converted by software instructions so that the data become information. Most wireless communication protocols support encryption in data transmissions with both sender and receiver Nodes programmed to single out abnormal, faded and delimited data by checks and verification to maintain the integrity of the information. Without proper checksums or cryptographic measures, attackers can modify data during transmission, leading to data corruption or manipulation. Some of the fixed-wired control systems do not support encryption.

 

External devices

 

  1. The difficulty of installing external devices with fixed-wired configurations protocol is well known. The Wireless module installed in, or close to, the lighting fittings provides ready power and locations for the external devices needed for other systems. The frequent breach in data integrity is often with the external devices. The Wireless module installed in the lighting fittings provides due encryption. IT Technicians are capable of installing new or replacing faulty sensor with common products in the market as the operational logic was preset in the Algorithm. Historical data remained secured as it was stored in association with or under the address of the Wireless module. Miniature battery powered Wireless module is available for connecting critical monitoring devices at odd locations. There is bound to be a lighting fitting close by with dedicated Algorithm to process the signal received or initiate response.
  2. Devices with composite sensors mounted in integrated assembly can be connected to the wireless mesh network. Most of them are products with the manufacturers’ expertise in weighting the monitored data for all-round applications. But the wireless mesh network and the Algorithm allow designer and client to strategically place its own selection of sensors and decide the gravity of data for prompt responses to commensurate with the particular characteristic of the installation during design, Testing and Commissioning (T&C) and operation at will.

 

Wireless module

 

Products obtained from the market with the same communication protocol can be used. Assistance from the original vendor of the software programme is required to install the output configuration details due to network security, the need to “burn in” the Algorithm and assign the same address to the replacement.

 

Apart from the programming and services to the Wireless module, no other reliance is needed from the original vendors; in particular, no proprietary in-line hardware is required in maintenance as with many fixed-wired systems.

 

Automation

 

The advantages of IoT Nodes in enhancing cross systems performance and the well-being of occupants and ambience are summarised as follows:

 

Field Control

 

The Node’s Algorithm is programmed to assess data inputs against preset parameters, trends of input variations and other interrupting data; the level of intelligence in control increases and is tunable. For on-field operations, it is not necessary to seek instructions from the system’s controlling software to meet prompt, automatic and localised executions, as long as responses are within the programmed authority of the Algorithm. For instances, lighting fitting is dimmed by rapid steps with interims to alert a lone occupant who escaped motion detection working in off-office hours or the movement rate of CO2 concentration as forecast information for Heating, Ventilation, and Air Conditioning (HVAC) system. A master software cannot be written to attend such numerous and sundries well-being features, nor is it appropriate for the controllers watching the dashboard in fumbling to locate the right button.

 

Self Photos / Files - c5

Local automation - on-field and real-time without operator interventions

 

Cluster Control

 

  1. The command of one-to-many Cluster Control is about a particular Node being assigned to relay changes initiated by occupants and/or driven by the ambient influences to other Nodes within a cluster either for prompt and actual action or information subject to the scrutiny of the Algorithm of the receiver Nodes. The cluster of Nodes is not necessarily confined within a zone or the same system. The level of intelligence is at the local arena, but enhanced by collective sanction or denial by the receiver Nodes as programmed. The data traffic path is short and effective for delivery. Latency is never a concern. The reverse arrangement of one (of the many) issuing commands to many (of the same many) is possible.
  2. The software of each system is usually not programmed to override the Cluster Control command within its own system. But where applicable, a particular command initiated by a particular Node to designated Nodes of other systems could be programmed to transmit first to designated system(s) via the mesh network to verify the state and nature of the command before prompt executions to preserve the integrity of safeguard control of the software of the receiver system. This is a form of real-time vertical scalability and automatic, different from the current practice of BMS in managing individual system.
  3. With or without interventions from the software programme(s), the subsequent status data of each Node, whether having responded to Cluster Control command and whether having subsequently shaped by ongoing adjustments of various systems according to each designed parameters, is of a de facto interactive nature. That is, whether the Cluster Control command was executed, the as-is status data of all the Nodes reflects the ambience of “adjusted” responses across systems without operators’ intervention. This is an operation of interoperability in real time and on the spot. This is also about putting engineering designs into operational practice in real time and on the spot.
  4. These consequential results are usually of more value for record. This locality-based Cluster Control transforms the traditional concept of one-to-many group control in ON/OFF only to a more intelligent level. As the command of Cluster Control could be initiated by the occupants and/or due to ambience changes, the well-being of both was addressed.
  5. Cluster Control among the cluster of Nodes is usually predetermined and/or modified during T&C on the basis that a response of one Node deserves concerted responses from Nodes either within its own system or the other ones. The configuration can be reset by a tablet to reflect changes in ongoing operational requirements.

 

Self Photos / Files - c6

On-field and real-time cluster automation - automation across systems

 

Edge Computing

 

  1. To truly achieve interoperability across building services provisions, equipment and estates management systems without operators’ interventions, a subprogram or function or Edge Computing of the BMS software programme should be delegated to monitor, assess and issue commands in real time to each and every system for concerted executions. The Edge Computing must be written for this purpose to circumvent the risks of issuing miscalculated and erratic data for systems coordination. The expertise in fine-tuning control parameters to improve and enhance systems performance is within the skill set of the building services engineers, but due to prevalent limitations, it was banned after the design stage.
  2. Edge Computing is an advanced development of Cluster Control requiring a more advanced logic (namely, Firmware) for assessment. With all systems connected to the wireless mesh network, the need of interoperability of various systems in cross operation control is, more often than not, confined within a certain area. Dedicating the “operation” of Edge Computing to controlling a specific and defined area sans the BMS main software is a logical choice with the obvious benefits of reduction in data traffic and latency. This is another form of scalability, vertical. With on-site information available, the expertise of the building services engineers can be utilised in the operation stage.
  3. The Firmware of Edge Computing required for this purpose can be simple and straightforward. It can be a software of a series of elimination of status data among conflicting and comparable parameters until all relevant status data, first as primary and then as feed-back control, had gone through the iterative cycles in reaching the final commands. After assessment, the Firmware issues either commands or data information, or both, to Nodes of respective system(s), either as status data, fresh input data, or execution command. Under most circumstances, the control software of various systems is not involved with the Edge Computing process. Like Cluster Control, Edge Computing is subject to the scrutiny of the receiver Nodes or, if first routed to other systems’ software, subject to further system-wise verification. Execution is calculated and automatic on field. The established history of a Firmware performance is a kind of idempotency records.
  4. The currently adopted stop-gap alternative (namely, collecting data from all systems for a software programme to flash up a display of status data for the operators to consider what actions needed to be taken) is not Edge Computing.

 

Self Photos / Files - c7

On-field and real-time analysis - interoperability across systems

 

Interoperability and system security

 

The most primal method in database programming is by storing, retrieving, sorting and calculating the data information. Management and exchanges of data information are made behind firewalls. Each software has its own embedded errors, and direct communication with other software increases the errors, or exceptions. The quality of the Application Programming Interface (API) between two software varies. Direct non-encrypted data information with commands to a software from other sources could escape detection and checksums. To avoid the risk of receiving and sending data information in the “dressed-up” format of computer language, a proper BMS requires communication protocols of various systems to transform data information in a more secured format. The most basic data protocol adopted by the industry is RS485, albeit it is not absolutely free of errors with embedded instructions.

 

 

A proper BMS requires communication protocols of various systems to transform data information in a more secured format.

 

 

Field Control, Cluster Control and Edge Computing channelled through the wireless network are forms of real-time inter-coordination of various building service provisions, equipment and estate management systems by managing encrypted secured data information. Any erroneous or abnormal commands will be rejected by the receiver Nodes. Consequential abnormal performance will be subsequently subdued by various systems’ self-adjustments as a self-healing feature. For critical concerted executions across systems, information can be programmed to first route to each software system for assessment before prompt executions for added scrutiny.

 

Master BMS software programme

 

The much-vaulted claim that building management software programme can also take on the role of Artificial Intelligence (AI) analysis was not correct. AI analysis software is basically about passive data. Building management software involves processing fresh inputs. This requires different expertise.

 

Field Control, Cluster Control and Edge Computing provide a certain scope of ready interoperability and valuable information of operation data. They were not abnormal or unusual data for AI statistical analysis. Instead, they were value zero at particular locations for the statistical analysis to converge upon.

 

Attempts to engage one operative software to coordinate cross systems performance are sometimes made for cost saving or convenience purposes. Many of them were written in high-level and more readable computer language meant for rapid applications. As the modus operandi of individual software was only catered to its own characteristic, it is difficult to see how an individual operative can interoperate cross systems of different natures.

 

Programmers have their own preferences in the choice of computer languages. The rudimental computer software is data based or Machine code. But low-level computer language is already capable of optimising the performance in writing precise control, highly efficient and specialised programme. As long as raw data are provided, programmers can execute their expertise in choosing whatever level or a mixed of low-level and high-level language as a further option.

 

Each building services and equipment system’s controlling programme has its own merit and quality and it should be retained to operate what the manufacturers consider adequate and proficient. The programmers of master software who aim to coordinate the interoperability should bear this in mind.

 

Pseudo software claiming to be “BMS” is not about real-time interoperability; it merely makes suggestions, the executions of which may be approved by the estates management as it sees fits. For a more secured option, the “instructions” from the “BMS” could be first transformed into the raw RS485 data format, stored at the file server for access by various and all systems to instigate concerted actions at a specific time— subject, again, to the scrutiny of the controlling software of each system and any escaped abnormality in turn subject to the intertwined effect of on-field responses across various systems. It is faster, safer, effective and with errors self-healed.

 

About the author

 

Ir Peter Y Wong is Past President of The Hong Kong Institution of Engineers and Founding Chairman of its Building Services Division. He is also Past President of The Chartered Institution of Building Services Engineers in the United Kingdom.

 

References

 

  1. Ir Peter Y Wong (2023). ‘Best design practice: Lighting installation and beyond – IoT structure’. Hong Kong Engineer, 51(6), pp.8-14.
  2. British Broadcasting Corporation. The toaster that changed the world. Available at: https://www.bbc.com/storyworks/future/the-new-wave/innovation [Accessed on 28 August 2024]
  3. Abbas M (2023). Kevin Ashton: Pioneering the Way for RFID Technology and the IoT. Available at: https://iotworld.co/2023/03/kevin-ashton-pioneering-the-way-for-rfidtechnology-and-the-iot/ [Accessed on 28 August 2024]
  4. Artefacts - Scientific Illustration & Archaeological Reconstruction. Heron of Alexandria – Automated Temple Doors. Available at: https://www.artefacts-berlin.de/portfolio-item/heron-of-alexandiraautomated-temple-doors/. [Accessed on 28 August 2024]

 

 

 

 

 

Explore Hong Kong Engineer