Optional 128/256/512MB and 1GB RAM Optional 256MB and 4/8GB eMMC Optional RS485/RS232, GPIO, CAN, DI/DO/AI/AO/RTD/TC/Pulse Counting/PWM and other I/O modules

Case Details

Prerequisites

Before proceeding with the installation, ensure you have the following:

NodeJS Version – The Device Agent requires Node.js 18 or later.

Operating System/Image – It can be installed on most Linux distributions, Windows, and MacOS, or via the provided Docker image. (Most ARMxy series industrial edge gateways can be installed)

BLIIOT ARMxy BL340 – A functioning device with internet access.

FlowFuse Account - Ensure you have a FlowFuse account. If not, you can create a free account that allows you to manage up to two edge devices for free. For more information, refer to FlowFuse Free Tier

Sudo Privileges – Administrator access to install required packages.

Getting Started

This guide explores how to install and run Node-RED through the FlowFuse Device Agent on the BLIIOT ARMxy BL340, enabling you to build, manage, and scale Node-RED flows efficiently from a remote location.

Installing FlowFuse Device Agent

Before we start, it is recommended to update and upgrade your system to ensure all your packages are up to date:sudo apt update && sudo apt upgrade -y

Next, let's install the FlowFuse device agent with the following script.bash <(curl -sL https://raw.githubusercontent.com/FlowFuse/device-agent/main/service/raspbian-install-device-agent.sh)

This script installs the Node.js runtime (if not already installed), sets up the FlowFuse device agent, and configures the device to automatically run the FlowFuse agent on boot and restart it in case of a crash.

To verify that the service is running, use the following command:sudo systemctl status flowfuse-device-agent.service

Registering the Device to Connect to FlowFuse

Once you have installed the FlowFuse Device Agent, you need to register the hardware to connect it to your FlowFuse team.

For instructions on how to register the hardware with your FlowFuse team, follow the documentation: Register your Remote Instance.

When registering your hardware, you will be presented with a dialog containing a one-time passcode command that the Device Agent uses to retrieve its configuration. Make sure to copy it.

Connecting Device

Execute the command you have copied with sudo as shown belowsudo flowfuse-device-agent -o <insert-your-three-word-token> https://app.flowfuse.com

Once executed, you should see an output similar to the one below, indicating that the FlowFuse Device Agent has been successfully configured:[AGENT] 3/21/2025 7:09:25 PM [info] Entering Device setup... [AGENT] 3/21/2025 7:09:27 PM [info] Device setup was successful [AGENT] 3/21/2025 7:09:27 PM [info] To start the Device Agent with the new configuration, run the following command: [AGENT] 3/21/2025 7:09:27 PM [info] flowfuse-device-agent

Now, when your device reboots, the FlowFuse Device Agent will automatically start, ensuring that your BLIIOT ARMxy BL340 remains connected to the FlowFuse platform.

Accessing Node-RED Editor.

Login into your FlowFuse account.

Click on the remote instances option in the left sidebar.

Click on the device and enable the developer mode by clicking on the top right-corner switch.

Once Developer Mode is enabled, click on the Open Editor option located next to the that switch.

Case Details

What is FUXA?

FUXA is a powerful web-based software to rapidly build and deploy scalable SCADA, HMI, Dashboard or IIoT systems. With FUXA you can create modern process visualizations with individual designs for your machines and real-time data display, as well as control instrument of automated industrial plants.

What does FUXA offer?

Toolbox for industrial network HMI, web applications and dashboard applications

Graphical View Builder Fully drag-and-drop WYSIWYG development environment

100% pure web technologies: HTML5, CSS, JavaScript, REST API and WebSocket communication

HMI and applications run in all HTML5-compatible web browsers on touchscreens, PCs, tablets or smartphones without any plugins

Connectivity: OPC-UA, Modbus RTU/TCP, MQTT, Siemens S7 protocols, WebAPI, Ethernet/IP (Allen Bradley), BACnet IP (native fully customizable connectivity available as a service)

Server runs on Windows, Linux and ARM architectures

SCADA functionality with integrated SQLite database engine: storage and display of large amounts of process data as well as alarm events and contextual data

User-based access with detailed permission settings

How ARMxy and FUXA improve Industrial Automation？

Universal connectivity: ARMxy is paired with FUXA. It supports multiple industry standard protocols (OPC UA, S7, Modbus, MQTT, WEBAPI, BACnet) - this allows the SCADA/HMI software FUXA to seamlessly collect data from all your devices, no matter what "language" they use.

Cost-effective performance: All are open source, saving you licensing fees. With open source technology, you have access to source code, so you can customize it to perfectly meet the needs of your factory.

Scalability and future-proofing: ARMxy's powerful hardware, developed based on ARM architecture, has flexible and stable functions, and you can also choose IO boards such as Ethernet/RS485/RS232/CAN/GPIO/DIN/DO/AIN/AO, which is scalable to adapt to the future development of the factory.

More Application Scenarios for FUXA?

With FUXA, the possibilities are endless. Although it is often associated with industrial automation, its versatility goes far beyond that. Imagine using FUXA's rich connectivity and open source features to transform building control systems and SCADA applications for energy, water supply, manufacturing plants, laboratories, etc.. Utilizing ARMxy's various hardware interfaces with FUXA, you can flexibly meet different project requirements without customization, making it a powerful tool for a variety of applications in different fields. Whether you are optimizing energy use, monitoring environmental conditions or improving productivity

Case DetailsUnder the general trend of Industry 4.0 and intelligent manufacturing, it is crucial to collect, store, analyze and visualize all kinds of data in the production process in real time. This helps enterprises better understand the production status, optimize the process flow, predict equipment failures and ultimately improve production efficiency and reduce operating costs.

Traditional industrial data collection and visualization solutions often rely on expensive PLC, SCADA systems and host computer software, which is a considerable burden for small and medium-sized enterprises or cost-sensitive application scenarios. In recent years, with the maturity of ARM technology and the prosperity of the open source software ecosystem, a more cost-effective and flexible solution has gradually attracted attention: an industrial data collection and visualization system based on ARM industrial computers and Grafana.

Why choose ARM industrial computers?

ARM architecture processors are widely used in the embedded field due to their low power consumption, low cost and small size. Applying them to industrial computers brings the following advantages:

Low power consumption and fanless design: Suitable for industrial environments with high requirements for power consumption and heat dissipation, reducing energy consumption and maintenance costs.

Cost-effectiveness: Compared with traditional x86 architecture industrial computers, ARM industrial computers usually have lower hardware costs, which lowers the threshold for overall system deployment.

Small size: Easy to integrate into various industrial equipment and control cabinets, saving space.

Rich interfaces: Many ARM industrial computers are equipped with common industrial interfaces such as GPIO, RS-232/485, Ethernet, etc., which are convenient for connecting various sensors and devices.

Good software compatibility: With the wide support of open source operating systems such as Linux, the software ecosystem on the ARM platform is also becoming more and more perfect, providing rich options for data collection and visualization.

Why choose Grafana?

Grafana is an open source data visualization and monitoring platform. With its powerful functions and ease of use, it is becoming more and more popular in the industrial field:

Powerful data source support: Grafana supports various mainstream databases (such as InfluxDB, TimescaleDB, PostgreSQL, MySQL, etc.), time series databases, and third-party APIs, which can easily connect and query industrial data from different sources. Rich visualization panels: Provides various chart types (line charts, bar charts, pie charts, dashboards, heat maps, etc.) and plug-ins, and users can create intuitive and easy-to-understand industrial data visualization dashboards according to their needs. Flexible alarm function: According to the set thresholds and rules, key industrial data can be monitored and alarmed in real time, and abnormal situations can be discovered and handled in time. Easy to install and configure: The installation and configuration process of Grafana is relatively simple, and it has complete documentation and community support. Cross-platform compatibility: Grafana can run on a variety of operating systems and hardware platforms, including ARM architecture industrial computers. Strong scalability: Supports plug-in extensions, and users can add new data sources, panels, and alarm methods according to specific needs. User-friendly interface: Provides an intuitive web interface for users to easily create, edit and manage dashboards.

Application scenarios of ARM industrial computers with Grafana

The solution of ARM industrial computers with Grafana is suitable for various scenarios that require industrial data collection and visualization, such as:

Equipment status monitoring: Real-time monitoring of the operating status of industrial equipment (such as temperature, pressure, current, speed, etc.), timely detection of abnormalities, prediction of equipment failures, and improvement of equipment utilization.

Production process monitoring: Monitor the key parameters of the production line (such as output, yield, energy consumption, etc.) to help optimize the production process and improve production efficiency.

Environmental parameter monitoring: Monitor temperature, humidity, PM2.5 and other parameters in the industrial environment to provide a good environment for production.

Energy management: Real-time monitoring of the factory's energy consumption (such as electricity, water, gas, etc.) to help enterprises optimize energy and control costs.

Remote equipment monitoring: Through the network connection, ARM industrial computers deployed in different locations can be used to realize centralized monitoring and management of remote equipment.

Small automation system: For some cost-sensitive small automation projects, ARM industrial computers can be used as the core of data collection and control, and Grafana can be used for visualization.

Steps to build an ARM industrial computer and Grafana system

To build an industrial data collection and visualization system based on ARM industrial computers and Grafana, the following steps are usually required:

Choose a suitable ARM industrial computer: Choose an ARM industrial computer with sufficient processing power, memory, storage space and required industrial interfaces according to the needs of the application scenario.

Install the operating system: Install Linux or other operating systems that support Grafana on the ARM industrial computer.

Install data acquisition software: Select appropriate data acquisition software or write a custom data acquisition program based on the type of data and device interface to be collected. Common data acquisition solutions include:

Protocol drivers such as Modbus/OPC UA: Used to communicate with PLCs, sensors and other devices.

Tools such as Node-RED/Python: Used to build flexible data collection and processing processes.

Message queues such as MQTT: Used to receive real-time data from various sensors and devices.

Install and configure the database: Select a suitable time series database (such as InfluxDB, TimescaleDB) or other database to store the collected industrial data.

Install Grafana: Install Grafana on the ARM industrial computer. Usually it can be installed through the package manager or by downloading the installation package.

Configure data source: Add the configured database as a data source in Grafana.

Create and customize dashboards: Use the various panels and chart types provided by Grafana to create intuitive industrial data visualization dashboards according to needs.

Configure alarm rules (optional): Set alarm rules based on key indicators and send notifications when data exceeds the preset threshold.

Deployment and testing: Deploy the system to an actual industrial environment for testing and verification. 

Advantages

Case Details

1. Characteristics and Advantages of ARM Industrial Computers

Low Power Consumption, High Performance ARM-based processors (e.g., Cortex-A series) deliver strong computational capabilities with low power consumption, ideal for industrial scenarios requiring continuous operation.

Rich Interface Support Supports industrial communication interfaces such as GPIO, CAN, RS-485, Ethernet, and USB, enabling direct connectivity to sensors, actuators, and industrial bus devices.

Compact and Rugged Design Industrial-grade construction (wide temperature tolerance, vibration resistance, dustproof) ensures reliability in harsh environments, with a small form factor for embedded deployment.

Linux/RTOS Compatibility Runs Linux (e.g., Debian, Ubuntu Core) or real-time operating systems (e.g., FreeRTOS) to meet real-time requirements.

2. Core Features of OpenPLC

Open Source and Cross-Platform Complies with IEC 61131-3 standards, supporting programming languages like Ladder Logic and Structured Text (ST), and runs on Windows/Linux/Raspberry Pi platforms.

Flexible Deployment Compatible with x86/ARM hardware, deployable as a soft PLC (on general-purpose computers) or hard PLC (e.g., Raspberry Pi, ARM industrial PCs).

Protocol Compatibility Built-in support for Modbus TCP/RTU, MQTT, OPC UA, and other protocols, facilitating integration with SCADA, HMI, and cloud systems.

Extensibility Allows custom function blocks via Python/C++ to integrate AI algorithms or third-party libraries.

3. Typical Application Scenarios for ARM + OpenPLC

3.1 Automation of Small to Medium-Sized Production Lines

Example: Food Packaging Machinery Control

ARM industrial PC (e.g., NXP i.MX8) runs OpenPLC to control servo motors and pneumatic cylinders via Modbus RTU.

HMI touchscreens interact with OpenPLC via Modbus TCP for real-time monitoring.

Advantages: Costs 1/3 of traditional PLC solutions, with support for custom algorithms (e.g., visual quality inspection).

3.2 Distributed Energy Management

Example: Photovoltaic Power Station Monitoring

ARM gateway (e.g., Allwinner T507) runs OpenPLC to collect inverter data (RS-485) and upload it to the cloud via MQTT.

Local logic control (e.g., battery charge/discharge strategies) is executed directly by OpenPLC.

Advantages: Edge computing reduces cloud workload and supports offline operation.

3.3 Smart Warehouse Systems

Example: AGV Scheduling Control

Multiple ARM industrial PCs (e.g., Rockchip RK3568) run OpenPLC to coordinate AGV paths via CAN bus.

Integrates ROS nodes for SLAM navigation, with OpenPLC handling motor control and obstacle avoidance.

Advantages: Unified hardware-software design simplifies multi-device coordination. 4. Comparison with Traditional PLC SolutionsAspectARM + OpenPLCTraditional PLC (e.g., Siemens S7-1200)CostHardware cost reduced by over 50%High hardware/licensing feesCustomizationDeep customization (drivers/algorithms)Closed ecosystem, vendor-dependentEcosystemSeamless Python/Node.js integrationLimited to TIA Portal/CODESYSUse CasesSMEs, R&D, rapid prototypingLarge-scale, high-reliability systems5. Challenges and Mitigation Strategies

Real-Time Limitations: ARM + Linux lacks native real-time performance; use kernel patches or FPGA coprocessors.

Long-Term Maintenance: Open-source version updates may cause compatibility issues; lock versions and maintain forks.

Security Risks: Harden Linux systems (e.g., disable default SSH ports, enable SELinux).

Case Details

ARMxy Based Single Board Computer have been widely used in the field of Internet of Things (IoT) due to their low power consumption, high performance, and cost-effectiveness. As an open source time series database, InfluxDB performs well in storing and analyzing the massive time series data generated by IoT. Combining the two can build an efficient and reliable IoT data platform.

Advantages of ARMxy Based Single Board Computer in the IoT field

Low power consumption: ARM architecture has lower power consumption and can run 7*24 hours a day, stable and low power consumption

High performance: It can process a large amount of sensor data in real time to meet the real-time requirements of IoT.

Cost-effectiveness: Compared with industrial computers with x86 architecture, it has lower cost and reduces deployment costs.

Miniaturization: It has a small appearance and is suitable for deployment in a small space.

Advantages of InfluxDB in IoT data storage and analysis

High-performance writing: Ability to efficiently write large amounts of time series data.

Flexible data model: Supports multiple data types and tags, and can flexibly store various IoT data.

Powerful query function: Provides a rich query language that can perform complex analysis on time series data.

High availability: Supports cluster deployment to improve system reliability.

Typical application scenarios of ARM Edge Computer with InfluxDB

Industrial IoT: Collect various sensor data of production equipment, conduct real-time monitoring and analysis, and optimize the production process.

Smart home: Collect the operating status data of home devices to realize the automatic control of smart home.

Environmental monitoring: Collect environmental sensor data, monitor environmental quality, and achieve environmental protection.

IoT System architecture

Sensor: Collects various physical quantity data, such as temperature, humidity, pressure, etc.

ARMxy Based Single Board Computer : Collects sensor data and writes it to InfluxDB.

InfluxDB: Stores and manages time series data, and provides query and analysis interfaces.

Front-end application: Based on InfluxDB data, develops various visualization applications to display and analyze data. Grafana, IgnitionSCADA etc.,

Technical implementation details

Data collection: Use C/C++, Python and other languages to write data collection programs and communicate with sensors through RS485/RS232, CAN/GPIO/DIN/DO/AIN/AO/RTD/TC etc.,

Data writing: Use the InfluxDB client library to write the collected data to InfluxDB.

Data query: Use InfluxDB's query language (Flux) to query and analyze the stored data.

Data visualization: Use visualization tools such as Grafana and Kibana to visualize the query results.

NPU (Neural Processing Unit) is a specialized coprocessor designed to accelerate artificial intelligence (AI) and machine learning (ML) tasks in ARM-based industrial control systems. It optimizes neural network computations, enhancing real-time performance, energy efficiency, and complex data processing capabilities in industrial automation.

Key Features of NPU

Dedicated Architecture Hardware-optimized for neural network operations (e.g., matrix multiplication, convolutions), enabling higher computational throughput at lower power consumption compared to CPUs/GPUs.

Massive Parallelism Utilizes parallel processing units (e.g., MAC arrays) to handle multidimensional data (images, sensor signals, audio) efficiently.

Low Latency & High Energy Efficiency Minimizes redundant instructions and memory access, achieving millisecond-level inference speeds critical for real-time industrial applications.

NPU vs. CPU/GPU

Benefits of NPU in Industrial PCs

Real-Time Performance: Meets strict latency requirements for robotics, production lines, etc.

Power Savings: Reduces energy costs for 24/7 operations and simplifies thermal management.

Edge Intelligence: Local data processing improves reliability in unstable network environments.

Cost Efficiency: Eliminates external AI accelerators, lowering hardware complexity.

Real-World Examples

Smart Warehousing: NPU-powered robots improve picking accuracy by 30% via real-time object recognition.

Power Grid Inspection: Drones analyze infrared images locally to detect faults (e.g., broken insulators).

Food Packaging: Vision systems with NPU reduce sealing defect rates from 0.5% to 0.02%, cutting GPU costs.

Typical Use Cases in Industrial Control

Machine Vision & Quality Inspection

Example: NPU accelerates vision models (e.g., YOLO, ResNet) to detect product defects (scratches, misalignment) on production lines, replacing manual inspection.

Advantage: Higher frame rates (FPS) for high-resolution images with lower power consumption.

Predictive Maintenance

Example: Analyzes sensor data (vibration, temperature) using time-series models (e.g., LSTM) to predict equipment failures (motors, bearings).

Advantage: Real-time processing of multi-sensor data streams, reducing downtime.

Autonomous Robotics

Example: AGVs (Automated Guided Vehicles) leverage NPU-accelerated SLAM algorithms for obstacle avoidance and path planning via LiDAR/camera data.

Advantage: Ultra-low latency ensures safe navigation in dynamic environments.

Voice & NLP Integration

Example: Enables voice-controlled machinery (e.g., "Start Line B") using on-device speech recognition models.

Advantage: Offline operation ensures privacy and reliability without cloud dependency.

IIoT Edge Computing

Example: NPU processes video/sensor data at the edge, transmitting only critical insights to the cloud.

Advantage: Reduces bandwidth usage and enhances data security.

Conclusion

NPUs empower ARM-based industrial computers with localized AI capabilities, driving efficiency and reliability in automation, quality control, and predictive maintenance. As Industry 4.0 demands smarter edge devices, NPUs are becoming essential for next-generation industrial systems.

Downlink: IEC104, BACnet/IP, BACnet MS/TP, Siemens, Mitsubishi, Omron, Delta PLC protocols, Modbus RTU, Modbus TCP, Various air conditioning protocols Uplink: BACnet/IP, OPC UA, Modbus RTU, Modbus TCP, MQTT

Applicable to BAS, BMS, HVAC, Lighting system, Fire protection system, Security system, etc.