Part 8: Photovoltaic power station energy storage side secondary system, EMS and overall station comprehensive troubleshooting SOP
This article corresponds to the panoramic topology diagram of the energy storage power station [Module 8: Secondary and EMS system], covering the local EMS server, SCADA monitoring background, dual-machine hot standby switching, time series database management, core control strategy (AGC / AVC / peak and valley arbitrage), relay protection device setting review, microgrid seamless switching and black start, station communication optical ring network, NTP/PTP time synchronization, telecontrol RTU four-remote testing and full-station cross-system comprehensive troubleshooting SOP.
1. Local EMS server, SCADA backend and control strategy maintenance SOP
EMS (Energy Management System) is responsible for station-wide power dispatching, strategy calculation, data storage and real-time interaction with the power grid dispatching main station.
(1) EMS server dual-machine hot standby and redundant switching test SOP
Master/standby server status check: Check the running status of the master server (Master) and the backup server (Slave), confirm that the dual-machine heartbeat cable (Heartbeat Cable) is connected normally, and the real-time redundant mirror synchronization of the database is without delay.
Dual-machine hot standby automatic switching test SOP (performed every six months): When the power station is running smoothly, manually unplug the main server network cable or cut off the main server power supply (simulating a sudden host failure). The backup server must automatically take over control of the entire site within < 1.0s, the SCADA monitoring screen must not freeze or crash, and the charging and discharging strategies being executed must not be interrupted.
(2) Database storage management and performance optimization
Database check: Check local time series database (such as InfluxDB, TDengine or MySQL) disk partition space. Disk remaining space must be > 30%.
Data archiving specifications: Second-level sampling data of the entire site must be saved locally for ≥ 90 days; minute-level historical data and alarm logs must be saved for ≥ 3 years. Database index reconstruction and cold data offline backup must be performed every quarter.
(3) Core control strategy execution and accuracy accounting SOP
Peak-Shaving strategy: Check the charge and discharge schedule within the EMS strategy engine to be 100% aligned with the latest peak-to-valley flat price period of the local power grid (such as off-peak charging: 23:0007:00, peak discharge: 11:00-15:00, 18:00~21:00).
AGC (automatic generation control) response and tracking accuracy: When receiving the AGC power adjustment instruction issued by the power grid dispatcher, the EMS must complete the instruction decomposition and issue it to each PCS within < 1.0s. The tracking accuracy of the actual output power of the whole station to the AGC instruction must be ≥ 99.0%.
AVC (automatic voltage control) reactive power regulation: Automatically controls the PCS output reactive power according to the grid-side node voltage fluctuation, and controls the 35kV bus voltage deviation within ± 2.5% of the rated voltage range.
2. Maintenance SOP for microcomputer relay protection devices, microgrid controllers and anti-backflow devices
Secondary protection equipment is the "last barrier" to prevent the expansion of electrical faults and ensure the safety of equipment and people.
(1) Microcomputer relay protection device self-test and setting sheet review SOP
Protection device panel self-inspection: Check the microcomputer protection devices (overcurrent protection, quick-break protection, zero-sequence overvoltage protection, differential protection) on the high-voltage switch cabinet and the main control room protection screen. The green light of "Run" of the device is always on, and the yellow and red light of "Alarm" or "Abnormal" are off.
Protection setting sheet review SOP (every year or when the setting changes): Derive the current setting value (Protection Settings) inside the protection device, and check it word for word with the "Relay Protection Setting Notice" issued and stamped by the power grid dispatching department. The current/voltage setting error must be 0.0% (absolutely consistent).
(2) Microgrid controller on-grid/off-grid seamless switching and black start (Black Start)
On-grid/off-grid seamless transfer (Seamless Transfer): In a microgrid or distributed energy storage scenario, when a grid drop or power outage is detected, the microgrid controller must control the static switch (STS) to disconnect the grid connection point within < 20ms, and at the same time control the main energy storage PCS to switch from PQ mode to V/F (voltage/frequency) mode to ensure that important loads do not lose power.
Black Start control logic: When there is a large-scale power outage in the entire network, the microgrid voltage is established by relying on the energy storage system's own battery and the station's DC panel to start.
Steps: Start the station UPS ➔ Activate the main control BMS and PCS ➔ The PCS establishes 380V/690V AC voltage in V/F mode ➔ Charge the booster transformer ➔ Close the important load circuit breakers step by step.
(3) Station anti-backflow protection device
When the energy storage/photovoltaic system is prohibited from sending power back to the public grid, the anti-backflow protection device monitors the current at the grid connection point in real time. When it is detected that the reverse power reaches the set threshold (for example, set to 10 kW), the anti-backflow device must send a derating or shutdown command to the EMS and PCS within < 200ms.
3. Station communication network, precise clock synchronization and telecontrol RTU "four remote" maintenance SOP
(1) Industrial optical ring network switch and packet loss rate test
Optical fiber self-healing ring network (RSTP/ERPS) self-healing test: artificially unplug an optical fiber in the ring network (simulating optical cable interruption), the ring network must complete self-healing reconstruction within < 20ms, and communication will not be interrupted.
Network packet loss rate and delay test: Use a network analyzer in the backend of the main control room to conduct a ping test on all BMS, PCS and gateways of the entire site. The network delay of the whole site LAN must be < 5ms, and the 24-hour continuous test packet loss rate must be < 0.01%.
(2) Precise clock synchronization of the whole station (NTP / IEEE 1588 PTP)
The clock source is combined with the satellite time synchronization receiver in the main control room (GPS/Beidou dual-mode time synchronization source). The clock error of all servers, BMS cluster controllers, PCS control panels and relay protection devices in the entire site must be ≤ 1.0 ms to ensure that the timestamps recorded in the SOE accident sequence are completely consistent when a failure occurs.
(3) Telecontrol RTU and dispatch gateway "four remote" test SOP
Communicate with the power grid dispatching main station through the IEC 60870-5-104 protocol, and conduct regular "four remote" function checks:
Telemetry (YC): Check the total active/reactive power of the entire station, bus voltage, total cluster current and other analog quantities, with an error ≤ ± 0.5%;
Remote signaling (YX): Test switch circuit breaker opening and closing status, fault alarm signal, displacement response time < 100ms;
Remote control (YK): Test and dispatch remote control of the main circuit breaker opening and closing function, and the logic locking is safe and correct;
Remote adjustment (YT): Test dispatch to remotely set PCS active/reactive power settings.
4. Typical cross-system comprehensive troubleshooting and recovery SOP (code-level diagnosis) for the entire site
When a complex fault occurs in the power station that spans the battery, PCS, temperature control and grid side, the operation and maintenance team must conduct unified command in the main control room and conduct troubleshooting according to the following SOPs:
(1) Comprehensive fault one: Insulation resistance of the whole station is too low (Riso Low) Cascade troubleshooting and positioning SOP
Step 1: Environment investigation. Check the temperature and humidity of each battery compartment and PCS compartment on the EMS interface. If the humidity in the cabin is > 85% RH, start the cabin dehumidifier for strong dehumidification for 2 hours, and then reset the test; if the humidity is normal (< 70% RH), it is determined that the physical insulation is damaged and enter step 2.
Step 2: AC and DC splitting. Open the PCS DC side main circuit breaker and decouple the PCS from the battery array. Use a 1000V megger to measure the PCS DC bus impedance to ground (if the impedance is < 1MΩ, check the PCS DC filter capacitor and IGBT) and the battery array total bus impedance to the ground (if the impedance is < 1MΩ, go to step 3).
Step 3: Cluster-level positioning. Open the DC circuit breakers of all battery cluster high-voltage units (BDUs). Close the single cluster switches one by one. When the Cluster N switch is closed, the insulation monitor alarms and the impedance drops suddenly, and Cluster N can be locked as a fault cluster.
Step 4: Module location and repair. Operation and maintenance personnel wore PPE and entered the cabin, unplugging the series high-voltage lines between the Packs in Cluster N. Use a megger to measure the resistance of the positive/negative poles of the Pack to the shell one by one, and find faulty Packs with impedance < 1MΩ. ➔ Open the cover and check whether there is leakage of liquid cooling pipelines, antifreeze water accumulation, or broken high-voltage cables. ➔ Clean the leakage/bandage and replace the cables. ➔ Test the insulation resistance impedance > 10MΩ. ➔ Restore closing.
(2) Comprehensive fault two: battery cell thermal runaway precursor (CO/H2 gas and temperature abnormality) three-level response SOP
Step 1: First-level millisecond automatic response. When the BMS/fire probe detects that single CO > 50ppm or H2 > 200ppm and the temperature change rate dT/dt ≥ 2.0 ℃/min, the BMS automatically cuts off the high-voltage contactor of the fault cluster, and the EMS issues an instruction to trip the AC/DC main circuit breaker, and at the same time automatically starts the outdoor explosion-proof exhaust fan for strong ventilation.
Step 2: Secondary automatic fire linkage. If the sensor further triggers a temperature/smoke secondary fire alarm, the cabin air conditioner and air duct fire damper (damper) will be automatically closed. After the audible and visual alarm outside the cabin counts down for 30 seconds, the electric explosion valve will be automatically activated to release the heptafluoropropane gas fire extinguishing agent. The cabin will be extinguished and left to stand for no less than 20 minutes.
Step 3: Three-level manual treatment and backup cooling. The operation and maintenance personnel wore positive-pressure air respirators and explosion-proof clothing to reach the safe area 50 meters away from the upwind outlet, and used infrared thermal imaging cameras to continuously monitor the temperature of the outer wall of the battery compartment through the partition wall. If the temperature stops falling and rebounds (there is a tendency to re-ignite), start the external Pack-level water spray/immersion system to continue cooling the battery.
(3) Comprehensive fault three: Communication interruption between PCS and grid side or grid overvoltage/undervoltage/abnormal frequency trip troubleshooting SOP
Communication interruption: Check the PCS network port indicator corresponding to the optical ring network switch in the main control room. If the light is off, test the fiber attenuation (must be < 3.0 dB); if the fiber is normal, restart the PCS control board communication module.
Abnormal grid-side voltage trip: retrieve the power quality analyzer waveform in the main control room. If the grid side voltage exceeds 1.1 Un (overvoltage) or is lower than 0.85 Un (undervoltage), check the over/under voltage protection action time of the PCS. After the grid voltage returns to normal, reset the fault on the EMS interface and resume operation according to the logic of "boost transformer ➔ AC circuit breaker ➔ DC contactor ➔ PCS standby ➔ power start"
Photovoltaic power station energy storage side secondary system, EMS and comprehensive fault site inspection sign-in form for the whole station
Power station name: _______________________________ Inspection date: ____ month ____ day, 2026
Inspection area: Main control room secondary system and EMS control backend Inspection person in charge: ____________________
1. Field equipment inspection and test data records:
2. Inspection defect registration and defect elimination dispatch:
3. Inspection sign-in and review signature:
Inspection engineer (signature): _______________ Supervision engineer (signature): _______________
Review (signature) of the shift leader: _______________ Sign-in and filing time: ____ hour on ____ month ____ day 2026