Local automation - no human intervention


Andy Nesling, electronic systems engineer at Lucy Switchgear, examines how it is possible for local automation schemes to perform essential switching functions without the need for human intervention and with little or no communications bandwidth.

In the last five years, automation of secondary distribution networks has grown rapidly in the UK. With the benefits of rapid network fault diagnosis and reconfiguration, DNOs have invested heavily in SCADA control systems, Remote Terminal Units (RTU) and reliable communications infrastructures to facilitate network automation. As the number of automated points on the distribution networks grow, communications bandwidth and control engineers’ resources are becoming more of an issue. Combine with this the continued requirement to reduce customer minutes lost (CML) and it becomes evident that remote control of medium voltage (MV) switchgear in isolation cannot provide the whole solution for optimum network automation.
A local automation scheme comprises the following fundamental building blocks:
• MV switchgear rated for intended mode of operation
• actuation for MV switchgear
• intelligent actuation control electronics (RTU / IED / PLC)
• a communications module (optional).
The function of a local automation scheme is to perform a MV switching operation, or series of operations, based on a set of criteria being met and without the need for human intervention. There are effectively two main types of local automated switching:
• protection switching – performed by intelligent relays and circuit breakers
• network configuration switching – performed by ring main units (RMU) and pole-mounted switchgear
For the purpose of this discussion we will primarily cover the advantages and issues associated with ‘network configuration’ switching as the issues associated with ‘protection switching’ are well known and documented. It is worth noting that there are many parallels between these two modes of switching, which are generally considered to be quite different.
Figure one shows a typical switching point on a MV network that can benefit from the implementation of a local automation scheme.
The figure shows a RMU at a system open point, running with the LHS normally closed and the RHS open. The RTU controlling the actuators on the RMU would be programmed with the following simplified scheme.
• On permanent loss of LV – check that the fault passage indicator (FPI) is not set
• Check that the RHS has full voltage present (MV)
• Open the LH switch and close the RH switch
• Check that the LV has been restored
The above scheme can restore the customer’s LV supply within 30s, providing the control engineer with valuable time to analyse the network and isolate the fault. In an actual scheme, there are many more validations, including actuators, low gas and user intervention, which all ensure safety and reliability of operation.
There are a number of advantages of de-centralising the monitoring and control of MV switching.
• Reducing customer minutes lost: a local automation scheme is able to carry out all the pre-switching checks and control of the switches in a fraction of the time it takes a control engineer to perform the same operations via a central SCADA system. This is especially true where communications to outstations is via a GSM, GPRS or PSTN connection (dial-up). The result of this is that customers are without power for seconds rather than minutes.
• Communications: a local automation scheme can be set-up to report only essential MV network data and automation scheme status, greatly reducing the amount of data that must be reported back to the central SCADA system. This is of particular benefit to systems using scanning radio modems. Alternatively, local control schemes can be setup with no communications at all (or just SMS message reporting). This can benefit DNOs with limited current SCADA capabilities that require all the advantages of network automation but with a clear upgrade path to future remote monitoring and control.
• Availability: as local automation schemes can be configured for operation without any remote dependence, they can continue to function even during a communications failure, greatly increasing the availability.
• Control engineer involvement: adverse weather conditions can result in large fault incidents on the MV network. A number of carefully planned, local automation schemes can free up control engineers to locate and isolate faults quickly, while the minimum
Although it is clear that there are many advantages of local automation schemes, there are a number of factors that must be given careful consideration when planning and implementing such a scheme:
• MV network complexity: when planning the location for a local automation scheme, careful thought must be given to all the possible modes of operation around that part of the network. On many projects, provision for ‘remote disabling’ of an automation scheme is essential. This facility then allows the control engineer full control over part or the entire network during maintenance or extreme circumstances.
• Commissioning time: with all of the extra functionality that a local automation scheme can offer, there is the additional time required to install and test it on-site. Switchgear manufacturers are increasingly working towards minimising this cost to the end customer by providing a full turnkey solution from the MV switchgear through to the RTU and communications device. The complete system can be delivered to site fully-tested, greatly reducing the time required to install and commission the system.
• Modes of failure: as any system becomes more complex, the number of ways in which it can fail inevitably increase. When considering local automation of an MV switch, the addition of a rapid mechanical disconnect mechanism is one way to mitigate this risk, allowing the user to continue operating the MV switchgear manually in the event of an automation failure.
In October 2004, Lucy Switchgear, in partnership with a UK DNO, successfully commissioned a local auto-changeover scheme for the MV supply to a sewerage pumping station. The scheme consisted of two Sabre RMUs with circuit breaker relays, motorised actuators and a Gemini-RTU. A key requirement of the client, a privatised water company, was to be able to restore supplies within 60s as the overflow tanks could reach the maximum capacity beyond this time (figures two, three and four illustrate the commissioned system).
The scheme programmed into the RTU was developed in close consultation with the DNO to cater for all the different operating scenarios and maintenance procedures. Initially, a communications link between the main SCADA system and the pumping station did not exist so a local control panel was provided, allowing an engineer to enable and disable the scheme locally. Following the successful commissioning of the system a high-powered radio link was added to the site allowing the DNO to monitor the RTU via the DNP3 communications protocol.successful implementation of local automation schemes
Over the past few years we have found that the key points to consider when planning and implementing a local automation scheme include gaining a clear understanding of all the different operating scenarios that must be considered and ensuring that the MV switchgear and actuation are designed to operate under all the expected modes of operation.
In addition, selecting a robust RTU / PLC that has the ability for future communications expansion using standard industry protocols is important, as is working closely with the system integrator to develop a scheme with high reliability and availability.
Finally, thorough testing prior to on site commissioning is advisable. Systems can go weeks or months without operating and so all scenarios need to be fully tested prior to commissioning. This ensures that issues with schemes do not arise years later or, worse still, after many more identical schemes have been installed.
In summary, local automation schemes can be an excellent first step into secondary network automation, with a clear upgrade path to future SCADA control and monitoring. They can be used as a method of reducing the communications load on an existing automated network and in critical applications to return supply with minimum downtime.