Innovation for utility companies is going beyond the hardware that is used to monitor energy consumption on the grid, bringing forth analytics to understand meter accuracy, which previously could not be tracked in field. In collaboration with Helen Electricity Network (a distribution system operator in Helsinki, Finland) and Aidon (an established supplier of smart grid and smart metering technology and services in the Nordics), a field trial was performed utilizing Energy Analytics Studio, an Analog Devices’ state-of-the-art, edge-to-cloud meter analytics solution using mSure® technology. This solution monitors the accuracy of an electricity meter through its deployed life and detects a wide range of tamper types. Meter accuracy monitoring is particularly pertinent to the Finnish market, and thus, the main focus of this trial.
Value of Accuracy Monitoring
Meters deployed in industrial, municipal, and residential environments are subject to varying conditions over time, including harsh weather, unpredictable loading, lightning strikes, and more. As such, the measurement accuracy of the meter may drift or change, resulting in overbilling or underbilling, the implications of which mean that time and money need to be spent to fix the resulting problem rather than catching an error soon after or even before it happens.
Even worse, it means that customer trust in the utility company is lost if they have been incorrectly billed by the electricity meter. Today, most utility companies initiate periodic sample testing and replace meters at regular intervals, methods that are both costly and intrusive to energy consumers.
The solution consists of a new technology called mSure, which can be integrated in every single new meter in the field, and a cloud-based analytics service that continually monitors and reports measurement accuracy of each meter in situ. The analytics service provides a utility company with visibility into the accuracy of all meters in its meter population to get ahead of meter issues, quickly replace meters that are outside of their allowed accuracy limits, and, if allowed by regulation, reduce and eliminate sample testing. A utility company can therefore take better advantage of the existing, powerful AMI network.
In addition, as energy consumption becomes more dynamic due to renewables, electric vehicle charging, and other variables, consumers’ electricity costs become more erratic, resulting in consumer inquiries or complaints. The solution allows for a utility company to quickly assess the accuracy of a specific meter, avoiding a costly field visit, and therefore improve customer satisfaction.
Field Trial Deployment
With the cloud-based analytics service, Helen Electricity Network has visibility to meter accuracy information for 40 evaluation devices using mSure technology deployed in the field since August 2018. Validation of the accuracy of these devices has been conducted by VTT/MIKES, an independent testing house in Finland. Phase 1, where 19 working devices were removed from the field for accuracy testing, concluded in October 2018. Phase 2, where the same 19 devices were subjected to accelerated lifetime testing by VTT/MIKES, was concluded in November 2019. Testing with high accuracy test equipment was performed to find a baseline accuracy for all the devices prior to the trial and to validate accuracy drift in the devices. The results from the drift found by VTT/MIKES and analytics service after phase 2 is shown in Figure 3.
The cloud-based analytics service is used in conjunction with purpose-built evaluation devices installed on premise, in series with a primary meter. The evaluation devices shown in Figure 1 feature Analog Devices’ ADE9153B energy measurement IC, which includes mSure technology to enable advanced diagnostics. This way, the meter can pass raw diagnostic information to the analytics service, which performs analysis to provide alerts, observe trends, and give reports on the health of the meter. In a real deployment, utility companies can deploy meters based on the ADE9153B energy measurement IC and use the analytics service to seamlessly gain the benefits of mSure.
Field Trial Results
In phase 1, the data from the cloud-based analytics service when compared to the reference measurements performed by VTT/MIKES shows that, for these 19 devices, the analytics service was able to track the accuracy drift to better than 0.1%. All 19 devices were grouped tightly and near 0%, showing minimal drift.
In phase 2, the meters were allowed to age for 8 months in an accelerated environment that simulated about 10 years in the field at 30°C average ambient temperature. Phase 2 was performed in a controlled laboratory environment, as opposed to the field, in order to accurately assess the performance of the analytics service and to speed up the aging process for this small batch of meters. Similar to phase 1, the accuracy drift of the 19 devices was tracked to better than 0.1%, as shown in Figure 4, but now both the accuracy testing and the analytics service show an average negative drift of about –0.05%.
As part of the laboratory experiments, one meter was artificially aged to show the capability of the analytics to accurately track larger drifts. The artificial aging was performed by placing a resistor in parallel with the shunt to modify the shunt value. The shift caused by this aging was measured by VTT/MIKES to be –1.91%, while the analytics determined the accuracy shift of this meter to be –1.96%, or only a 0.05% difference.
In conclusion, phase 1 of this field trial showed that the analytics service is able to track the accuracy of mSure enabled devices deployed in the field very closely, within 0.1% but little meter drift was seen in that time. For phase 2, in a simulated 10 years in the field, the accuracy drift continued to be tracked at 0.1% as the accuracy testing and analytics showed the meters drifting in the negative direction. The field trial demonstrated the ability for mSure technology coupled with an analytics service to monitor meter drift with enough accuracy to be used in place of meter sample testing.