FAQ
Frequently Asked Questions
Find answers to common questions about ZHIYUAN equipment health management — protocol integration, AI alerting, industry scenarios, deployment, and security. Contact us if you have more questions.
Platform & Protocols
Which industrial protocols does ZHIYUAN support?
ZHIYUAN supports three major industrial IoT protocols: MQTT, Modbus-RTU, and Modbus-TCP. It is compatible with mainstream third-party DTUs and sensors already on the market. Custom protocol adaptation can be negotiated for proprietary protocols, but standard-protocol devices achieve the fastest integration.
Do I need to replace existing sensors and DTUs to use ZHIYUAN?
No. ZHIYUAN is designed for "like-new legacy integration" — it is compatible with 95% of third-party DTUs and sensors on the market, directly reusing your existing data acquisition hardware. As long as devices support MQTT or Modbus standard protocols, no hardware replacement is needed.
How long does it take to connect standard devices to the platform?
Standard-protocol devices (MQTT / Modbus) can complete data reporting integration within 30 minutes once the point table is prepared. The process: submit point table template → platform-side access configuration → integration testing and verification → alert rule configuration. Complex or non-standard protocols are evaluated separately.
What types of sensors and monitoring parameters does the platform support?
The platform supports common industrial sensor signals including temperature, humidity, vibration, current, voltage, power, ammonia concentration, water meters, electricity meters, and gas meters. Any analog or digital value readable via MQTT attribute reporting or Modbus registers can be connected, regardless of brand or model.
AI Alerts & Intelligence
What are the advantages of AI time-series alerting over traditional threshold alerts?
Traditional threshold alerts only trigger when parameters cross a set limit, producing high false-alarm rates (a common industry observation — figures vary by study). AI time-series alerting learns each device's own historical operating baseline, identifying slow drift and hidden anomalies. Combined with same-type alert merging within time windows and delayed confirmation to filter transient fluctuations, false alarms can be significantly reduced. The exact reduction depends on the equipment and operating conditions — validate it in your own pilot.
What notification channels does the alert system support?
Multi-channel graded push notifications are supported: enterprise WeChat, SMS, and web dashboard pop-ups. Alerts are graded as urgent/normal: urgent alerts push to on-duty personnel within seconds, while normal reminders are aggregated into daily reports. Unacknowledged alerts automatically escalate to supervisors, ensuring every alert receives a response.
Can the work order system integrate with enterprise WeChat?
Yes. After an alert generates a maintenance work order with one click, work order notifications are pushed via enterprise WeChat. Maintenance personnel can receive, accept, check in on-site, capture photos, and complete records through the WeChat mini-program, achieving full closed-loop management from alert to work order archiving.
How much historical data does the AI model need to establish a device baseline?
Typically 1–2 weeks of normal operating data is sufficient to establish an initial baseline. The baseline continues to optimize as operational data accumulates. For newly connected devices, the platform provides same-model population baselines as a cold-start reference, gradually transitioning to individual baselines.
Gateway Hardware & Access
What communication methods does the gateway support? Is network cabling required?
The gateway supports 4G wireless, Wi-Fi, and Ethernet. For most factory scenarios, 4G wireless mode is recommended — simply insert a SIM card to deploy, with no on-site network cabling needed. Cloud/local dual-mode with flexible switching, local data caching during network outages, and automatic backfill upon recovery — no data loss.
Which devices are suitable for connection? Which are not recommended?
Suitable: general electromechanical equipment with standard Modbus/MQTT interfaces — air compressors, pumps, motors, fans, chillers, elevators. Not recommended: core production-line equipment with complex PLC logic (requires custom solution), high-security devices (requires physical isolation), old equipment without communication interfaces (external sensors can be added but cost-effectiveness should be evaluated).
How do I choose the right gateway model for my scenario?
Lite Gateway: for up to 10 monitoring points with standard Modbus/MQTT protocols (single air compressor, small pump room), supports 4G/Wi-Fi dual-mode reporting. Pro Gateway (Industrial): for 10–50 points with edge computing preprocessing, supports offline caching, protocol parsing, and local alert logic — recommended for multi-device workshops and pump station clusters. Pro+ Gateway (Heavy Industrial): IP66 rated, -20~70°C wide temperature range, supports RS485/Ethernet/LoRa multi-interface — suitable for high-temperature, high-humidity, and dusty environments. If unsure, our solution advisor can recommend the optimal model based on your point table and site conditions.
How long does integrated gateway + AI platform deployment take from kickoff to go-live?
Standard scenario is divided into five phases: ① Site survey and point confirmation (1–2 days, client provides equipment list and site photos); ② Gateway shipping and installation (3–5 days, including SIM card activation); ③ Platform-side device access and data commissioning (0.5–1 day, completed remotely); ④ AI baseline learning (7–14 days, automatic modeling during normal equipment operation); ⑤ Trial run and acceptance (7 days, verifying alert accuracy and work order workflow). From contract signing to official go-live: approximately 3–4 weeks. Scenarios with existing sensors can shorten to 2 weeks. Actual timeline depends on device count and site conditions — a detailed schedule is provided during the solution phase.
What does the customer need to coordinate on their side?
Three main coordination items: ① Provide equipment list and site photos (we provide the point table template — just fill in device name, model, communication protocol, and installation location); ② Confirm gateway installation location (requires 220V power supply + 4G signal coverage, no equipment shutdown needed during installation); ③ Assign operations personnel to participate in alert rule configuration discussions (who receives alerts, which channels for which severity levels). All of the above are non-intrusive — no equipment modification or production stoppage required.
What if my equipment uses proprietary protocols or has no communication interface?
For proprietary protocols: protocol adaptation development can be evaluated, typically taking 2–4 weeks — client needs to provide protocol documentation for integration testing. For legacy equipment without communication interfaces: external sensors can be added (adhesive vibration sensors, CT current clamps, temperature probes) with data uploaded to the platform via DTU. External sensor costs are evaluated based on point count and sensor type. For equipment with no viable retrofit path (no power, inaccessible for sensors), we recommend excluding it from the monitoring scope and prioritizing key accessible equipment first.
Industry Scenarios
Which industries and scenarios is ZHIYUAN suitable for?
The platform covers five core scenarios: ① Industrial Manufacturing — electromechanical equipment status monitoring (air compressors, pumps, motors) and energy management; ② Energy Management — water/electricity/gas sub-metering and leakage/waste detection; ③ Smart Aquaculture — shed temperature, humidity, ammonia monitoring; ④ Elevator IoT — operational status monitoring and digital maintenance work orders; ⑤ Warehouse & Server Room — cold storage, data center, and laboratory temperature/humidity monitoring.
Does deploying this system require production line shutdown?
No production line shutdown is required. The platform reuses existing sensors and DTU equipment. Sensor installation can be completed during normal equipment operation (e.g., adhesive temperature sensors, magnetic vibration sensors) without production line interruption. This is a core advantage over traditional SCADA retrofit approaches.
What is the approximate investment for a small-to-medium factory to deploy this system?
Investment depends on monitoring point count and feature requirements: the platform uses a per-point annual subscription model, with minimum-scale (3–5 key points) annual fees being manageable. Reusing existing DTUs and sensors significantly reduces hardware costs. We recommend starting with 1–3 key pieces of equipment as a pilot, validating ROI with real results before expanding.
Deployment & Security
How is data security ensured?
The platform implements multi-layer data security: TLS encryption at the transport layer; tiered time-series data storage with retention based on subscription plan; customer self-service periodic export of key data; device access uses per-device unique key authentication. The platform only collects equipment operational parameters and does not access sensitive information such as production recipes or process parameters.
What if on-site network conditions are poor?
The acquisition gateway supports 4G wireless upload with local data caching during network interruptions and automatic backfill upon network recovery, ensuring no data loss. This solution has been validated in farms, remote pump stations, and other scenarios with unstable network conditions. For signal dead zones such as underground spaces, gateway placement must be confirmed during point planning.
Does the platform support private deployment?
The current primary offering is SaaS subscription to minimize the barrier for small and medium customers. For enterprise customers requiring private deployment, a private deployment solution can be evaluated. Please contact our sales team for details.
Purchasing & Partnership
Can we try before purchasing?
Yes. The platform offers a pilot program for first-batch customers: select 1–3 key pieces of equipment or a single area for initial connection. Implementation fees are waived during the pilot period, allowing you to validate alert and work order closed-loop effectiveness with real data. The pilot period is typically 3 months; both parties evaluate results before deciding on a formal subscription.
What purchasing methods are supported? Can we use corporate contracts and bank transfers?
Corporate contract signing and bank transfer payments are supported, with VAT invoices available. Annual subscription follows either the calendar year or the contract signing date. Renewal reminders and pre-expiry data export ensure service continuity.
Didn't find your question?
Feel free to contact our technical support team, and we'll provide detailed answers.
Contact Us