Determining if a hot runner predictive maintenance system is faulty requires a three-tiered diagnostic logic: "abnormal indicator monitoring + health index assessment + multi-dimensional status linkage," to achieve accurate judgment from local anomalies to system-level failures.
1. Basic Anomaly Indicators: Directly Measurable Fault Signals
System faults first manifest as quantifiable and observable sensor or communication anomalies. These are the first line of defense:
|
Anomaly Type |
Specific Manifestations |
Judgment Basis |
|
Temperature Anomaly |
Single-point temperature consistently exceeds limits (e.g., >105℃), multi-point temperature difference >15℃, trend drift >3℃/cycle |
Exceeds dynamic threshold range and persists for more than 3 cycles |
|
Current Anomaly |
Heating circuit current fluctuation amplitude >±20% . Baseline, periodic spikes |
Combined with temperature analysis to determine if heating coil short circuit/open circuit |
|
Communication Interruption |
Gateway fails to receive any sensor data for 5 consecutive seconds, heartbeat packet loss ≥3 times |
System automatically marks as "communication failure" |
|
Sensor Reading Anomaly |
Thermocouple displays "over", "000" or negative values, current sensor output is consistently 0 |
Complies with IEC 60584 standard fault diagnosis logic |
Operation Prompt: Any single abnormal indicator triggers a Level 1 warning, requiring manual review; if three or more indicators are abnormal simultaneously, the fault diagnosis process begins.
2. Health Index (HI) Assessment: System-Level Status Quantification Model Modern systems employ a multi-sensor fusion Health Index (HI) to transform raw data into a comprehensive health score of 0–100, serving as the core basis for determining whether the system is "on the verge of failure."
Health Level Classification:
|
HI Value Range |
Status |
Handling Recommendation |
|
90–100 |
Normal |
Maintain as planned |
|
70–89 |
Slight Deterioration |
Increase inspections and record trends |
|
40–69 |
Warning Status |
Initiate diagnostic process, check sensors or algorithms |
|
<40 |
Fault Status |
Immediate shutdown, trigger work order |
Industry Practice: After deployment in injection molding companies, the Yuanshuo PHM system achieved a 92% accuracy rate in determining HI<40, with a false alarm rate of less than 5%.
