IntelliGrid Architecture
|
|
|
Parent Menu
Same Level Menu
Child Menu
Questions/Comments
|
Wide-Area Monitoring And Control – Automated Control FunctionsContentsNarrativeTransmission Automated Control (Baseline) describes a set of functions that are typically automated within a substation, but are not directly associated with protection, fault handling, or equipment maintenance. In general, they serve to optimize the operation of the power system and ensure its safe operation by preventing manually generated faults. These functions include:
The functions described in this use case were traditionally performed by individual devices acting alone. When implemented this way, they did not have any effect on the communications system. However, in the last five to seven years, these functions have been distributed across the substation. That is, the software logic controlling the function now often resides on a different device than the one which provides the inputs or outputs to the process. This change has taken place because the use of substation LANs has made it economical to place Intelligent Electronic Devices (IEDs) close to the equipment they are monitoring and controlling. Logic has therefore either been centralized, with a single Substation Computer using the IEDs as remote controllers, or it has been distributed among the IEDs themselves. In either case, the communications system has now become part of the automation functions. Voltage Regulation using Tap ChangersIn voltage regulation, the automation system ensures a constant voltage on the substation bus by adjusting the tap of one or more transformers. A monitoring IED provides a voltage value to the Substation Computer, which has been programmed with threshold and hysteresis logic. The IED is usually monitoring the bus side of a transformer. In more complex situations, IEDs may monitor multiple voltages throughout the station and pass them all to the Substation Computer as input to the logic. When the logic indicates that the bus voltage must be adjusted, the Substation Computer issues a control operation to the IED connected to the transformer tap. This will change the monitored voltage, which will be fed back through the logic. The voltage control logic typically has a pre-programmed qualification delay in the tens of seconds – adjusting the tap causes wear on the equipment, so adjustments should not be made lightly. Therefore, an appropriate update time for the monitored voltage is on the order of one-half second to one second. Because of the wear on the transformer and tap, and the impact on the rest of the system if adjusted wildly, tap raise/lower operations are typically performed with select-before-operate logic. Redundancy and reliability of the communications path is important. Volt/VAR Regulation using Capacitor or Shunt ControlIn capacitor bank control, the automation system optimizes the voltage and inductive load on a line or bus by connecting or disconnecting one or more capacitor banks. It prevents the imaginary part of the load from becoming too large, reducing voltage and the efficiency of the system. The banks may be widely located across the power system, or within the substation. There are many different logic algorithms for performing capacitor bank control. The simplest is calendar or time of day control, in which the load on the power system is not even monitored. The logic simply assumes that the inductive load will be higher at certain times of the day or year. In areas where inductive load is largely caused by air conditioning, logic may switch based on ambient temperature. Some algorithms monitor voltage only and switch when it passes certain thresholds. More sophisticated algorithms monitor both current and voltage and switch based on either the calculated power factor or directly on the calculated inductive load (VARs). There are typically hysteresis settings on such logic to prevent frequent switching. Shunt control occurs under similar conditions, but with the addition or removal of inductive loads. In distributed Volt/VAR control, one IED controls one capacitor bank on a given line, and each IED makes switching decisions individually. In centralized Volt/VAR control, each IED reports monitored values back to a Substation Computer. The Substation Computer may make switching decisions based on averages or groupings of voltages. When it decides a switch is necessary, it sends a control message to the appropriate IED, which may or may not be the device reporting the controlled measurements. The hysteresis in some Volt/VAR algorithms may often be in hours, so communication delays in tens of seconds are easily acceptable. It is fairly common to broadcast capacitor bank control messages, without any select-before-operate logic, since the effect of any given control is usually small. When capacitor banks are located remotely, pagers have sometimes been used as a communications media – one number to switch the bank in, one to disconnect it. InterlockingInterlocking prevents unsafe operation of the various switches and breakers within a substation. When an Operator or software application attempts to operate a control, the automation system evaluates the state of the entire system and may reject the control request based on pre-programmed logic. This logic corresponds directly to the topology and interconnection of the substation. Simpler substations may have little or no interlocking. The most complex logic is associated with complex transformer and bus redundancy systems. For instance, an operator will close an earthing switch on a section of the bus to ground it prior to permitting maintenance on the equipment. However, the operator may not be aware that the bus section is still live due to an interconnection with another bus section or a feeder fed by another bus section. The automation system must prevent a fault by rejecting the Operator’s request. Interlocking is most reliably and efficiently performed by the device that must perform the requested operation. In the past, it may have been performed by the substation GUI or SCADA master station, when those were the only locations that could perform switching. It is still often performed by a Substation Computer or Data Concentrator which serves as a clearinghouse for all control operations to the substation. This centralized logic mechanism is still used, especially because deregulation has increased the number of master stations that require access to the substation. However, more and more frequently, interlocking is performed by logic on the IEDs themselves, operating on data distributed by peer-to-peer communications between the devices. This peer-to-peer communications has been made possible by the introduction of the substation LAN. Performing interlocking at the IED permits the same logic and performance to be in effect regardless of whether the control request originates at a remote site, at a substation GUI, at the control panel of the IED, or even a manual panel switch. In an ideal system, the state information required to perform interlocking would be updated simultaneously throughout the system. Any delay provides a window in which a control could be mis-operated. However, in practice, it is sufficient to update the state of the system in less than a second or two. This interval represents the typical time between the moment an Operator checks the state of the system on a GUI or display panel, and the moment the Operator makes the control request. As more automation applications are deployed in the substation, human reaction time will become less of a factor, and the demands on interlocking will increase. Today, a challenging interlocking requirement for an advanced substation is less than 200 milliseconds between updates. The control itself is typically issued with select-before-operate logic. The distribution of state information for interlocking may be broadcast or multicast. Redundancy and reliability is extremely important. Sequenced ControlsWhile interlocking is intended to prevent Operator-initiated faults by rejecting invalid controls, sequenced controls automate some portion of the Operator’s tasks to eliminate the possibility of an invalid control ever being issued. For example, consider a substation with two transformers and a normally open switch between the two bus sections connected to each transformer. There are two different philosophies that an Operator may employ to take one of the transformers out of service. In “make before break”, the Operator should (1) connect the two buses, (2) disconnect the transformer from the bus, and (3) disconnect the transformer from the upstream transmission line. This method ensures there will be no outage of service. In “break before make”, however, the Operator should (A) disconnect the transformer, then (B) quickly connect the bus and then (C) disconnect the transmission line. This results in an outage, but prevents side effects resulting from mis-matched transformers sharing the same bus.
In a sequenced control, the Operator simply requests the isolation of the transformer, and the automation system performs the controls in the sequence required by the utility. The Operator is not permitted to perform any other sequence. In the “break before make” case above, it also ensures that the resulting outage is as small as possible, because the automation system can perform the sequence faster than a human. The speed of a sequenced control is related to the components involved in the sequence. For instance, the logic may need to wait for motorized switches to connect or disconnect before proceeding with the next control in the sequence. In a “break before make” sequence as described above, however, the length of the outage must be minimized and a value of less than half a second is typically desired. All sequenced controls are typically service-affecting and are therefore executed with select-before-operate logic. Load BalancingLoad balancing is typically a distribution operation, performed between two transformers within a substation, but may also be performed in transmission systems between substations. In the distribution case, two feeders serviced by separate transformers are connected at their remote ends by a normally open switch. A pole-top IED controls the switch. Other IEDs monitor load on the line. The IEDs report the state and load of the system to a Substation Computer. The Substation Computer detects the condition when one transformer is heavily loaded and the other has excess capacity, and sends a message to the pole-top IED to close the switch. Now, instead of one line loaded at 90% and the other at 25%, both may be loaded at 50%. Since resistive losses vary with the square of the current, this action improves the efficiency of the power system and reduces wear on equipment. In transmission systems, two substations having lines feeding the same third substation may share load. This type of logic is typically centralized, not distributed.
As with tap changing, load balancing is not an action that is typically performed lightly. Qualification times for the logic may be measured in minutes or even hours. Therefore, update times and control transmission times may be measured in seconds. In distribution operations, this is fortunate because IEDs controlling the switches may be remote and only reached via slow links. Some utilities may prefer that the process not be completely automated, and that the automation system request confirmation from the Operator before taking action. Reliability of the data is important and redundant links may be used. Automated Service RestorationAutomated Service Restoration is typically a distribution operation, but may be performed in transmission systems when “loops” are possible between substations. In the distribution case, two feeders are connected at their remote ends by a normally open switch. Several other switch / breaker combinations are located at other points along the feeders. All the switches and breakers are monitored by IEDs. When a fault occurs, the IEDs on the upstream side of the fault trips its breaker. It notifies the Substation Computer of its action. The IED on the downstream side of the fault notifies the Substation Computer of the loss of current and estimates the direction of the fault. Based on that information and pre-configured logic, the Substation Computer recommends to the Operator that the breaker of the downstream IED should be opened and the normally open switch should be closed. The Operator typically directs the Substation Computer to do so, and the Substation Computer forwards the decision to the IEDs. When the IEDs perform the operations, power is restored to all portions of the feeders except the section in which the fault occurred. The more break points there are in the feeders, the fewer customers will be affected by a given fault.
Time is of the essence in service restoration, but utilities typically require an Operator approve the decision of the system, so the human Operator is usually the slowest link in the system. Communications times may be measured in seconds. The breaker tripping is done by an individual IED without need for communications.
An alternative scenario occurs if the fault is not on the main feeder but on a lateral. In this case, the fault causes the protection IED at the substation to trip and attempt reclosure. While the current is zero between reclosure attempts, the IED nearest the fault opens its switch to clear the fault. This is called “auto-sectionalization”. Then, when the next reclosure occurs, service is restored to all subscribers except those on the lateral. In this case, there is no effect on the communications system other than to monitor that the events occurred.
Normal Sequence Steps
Steps – Alternative / Exception Sequences
|
|
IntelliGrid Architecture
|
