INCH Duo Smart Charging at the Location
2. Smart Charging Solutions at the Location
Overload Prevention
Main functionality of the overload prevention algorithm is to decrease the EV charging load to prevent the overload of grid connection point. This function has the highest priority among all the power management algorithms. The EV charging load shall be controlled so that the total load (household load + charging load) of grid connection point is below the rated current of the main fuse.
When the user connects EV to charger, and prior to beginning of charging, the charger determines the current available for charging as the difference between the rated current of main fuse (reduced by a safety margin that can be pre-set by the user via charger’s web interface) and the last measurement received from Load guard.
Figure 4: Use of household consumption data to prevent overload
By using Load Guard device, the household load can be measured and used in overload prevention algorithms:
Static limit of maximum allowed charging current per phase.
Static limit of maximum allowed charging current per phase in case connection with Load Guard sensor / back-end is lost.
Detection and visualisation of available supply and automatic adjustment of charging power.
Detection and visualisation of surplus energy returned to the grid (Production from renewable energy sources).
Frequency Regulation
Frequency regulation can prevent grid breakdowns. Frequency depends on the balance of production and consumption in the electricity grid. When the consumption is bigger than production the frequency falls below nominal value. In the event when the production is bigger than the consumption frequency can raise above the nominal value of frequency.
To bring the frequency back to the optimal, either the consumption or production needs to be changed. With charging stations, it is very easy to manage the load and with this, influence the frequency of the grid.
Frequency regulation works on a charging station level. Each charging station measures the frequency value 3 times per seconds and based on the measured frequency automatically reacts by lowering the current.
In the case when the frequency is lowered because the consumption is too high, the charging station can simply decrease the charging power (currents). Regulation of the charging current is made on basis of the frequency limit settings. If the system frequency is lower than nominal value, the charging current will also be lowered. Nominal value of frequency in the power system can be different, usually defined as 50 Hz or 60 Hz.
Start when frequency under [Hz] | Delay start (when persistently under) [s] | Decrease output current to [A] | Delay end (when persistently above) [s] |
50.0 | 10 | 32 | 4 |
49.0 | 8 | 25 | 8 |
48.0 | 6 | 15 | 10 |
47.0 | 4 | 6 | 20 |
46.0 | 2 | 0 | 40 |
Table is showing an example of defined stages of frequency regulation and setting of one stage in charging station’s web interface is shown at the figure below.
Figure 5: Setting of one stage of frequency regulation
The settings of under-frequency regulation are usually defined by local distribution system operator and can be a requirement or only a recommendation. All the frequency regulation settings are defined by the circumstances at the actual location, local legislation, and requirements.
It is possible that the under-frequency regulation can be subsidised and represents an income to the charging station’s owner.
Charging Process Optimization
Economic/Price Optimization of Charging
When different energy tariffs are set, the charging station can optimize the charging process to ensure that the price of charging will be the lowest possible. Based on the set parameters, the charging algorithm calculates the optimal charging schedule plan.
For example, normally, the charging during the night would have the lowest price, however the exact time interval of low tariffs could influence the calculation.
The possible production of energy at the location could be treated as being of high tariff, or the use of this energy for charging would be preferred and designated with low tariff. In most cases of public charging stations, the user wants the electric vehicle to be charged as soon as possible and the price optimization of charging is not desired.
The calculation of optimal charging plan:
Based on energy tariffs.
Time scheduling of charging towards lower tariffs or self-consumption when user preferences and pricing allows it.
Evaluation of on-site production (e.g., photovoltaics).
Operation Optimization
In some cases, the easiness of the charging process can be supported with the use of machine learning and pattern recognition to evaluate the user habits and offer the charging schedule plan, optimized to specific user (by evaluation of the historic data of charging).
With the setting of user’s departure time, the charging process could optimize the charging of a whole cluster. The electricity price can be a part of optimization algorithms, or not. In case of public charging stations, the operation optimization is usually not wanted, however its main use are semi-public charging stations (charging at workplace) and individual users.
To achieve better operation with existent devices, to connect to other systems, or to configure additional functionalities, the digital inputs and outputs can be used.
The communication protocol is Modbus. The support for Modbus protocol further extends the options for integration with external systems and means of monitoring and control of operation of individual charging station or a cluster. It can be implemented for control of operation on local level, or for the remote control from the back-end office.
Operation optimization is linked to other functionalities, e.g. price optimization and power management possibilities and is dependent on the exact use case scenario.
Different configurations are possible, using only charging stations to manage the whole cluster, using the Etrel Ocean monitoring system, or external monitoring system.
Because of it, the optimization of the complete system is subjected to the use case of individual customer and is not limited only to:
Machine learning and pattern recognition using built-in AI to predict and optimise each charging session.
Collection of user's departure time over to refine automatically suggested charging profile.
Support for Modbus protocol for integration with external smart building systems.
Power Management
Basic power management is performed on individual charging station and consist of several settings and is supported by other functionalities. It can also be performed on the level of back end for load areas, sub load areas, or individual charging stations.
Cluster Management
The goal of this functionality is to distribute the power available for charging among several chargers in the cluster when multiple EVs are charging. Cluster management on level of charging station is using one charging station designated as master of the cluster.
Based on user preferences and current installation's load conditions.
Master-slave relationship with floating master.
INCH Duo charging station: Connection of up to 18 charging stations possible.
INCH Duo can provide power management of up to 36 electric vehicles, meaning cluster of 18 charging stations INCH Duo. This is valid for the most unfavourable scenario with low power capacity available, meaning constant need for power management recalculations with inclusion of data obtained from Load Guard. INCH Pro could also control larger clusters, depending on the individual case.
Larger cluster (supply of up to 300 electric vehicles in most unfavourable scenario) is possible with use of industrial computer and connection to Etrel Ocean management software.
The total consumption of the cluster can be limited with the maximal capacity of the grid connection point. To optimize this functionality, user charging preferences and EV onboard charger’s characteristics must be known or estimated. User charging preferences can be obtained by the charger when user enters his preferences into the charging station’s settings. Charging station calculates them using the preferences prediction algorithm with the historic charging data.
When several chargers are installed behind the grid connection point the control algorithm must decide how to distribute the current available for charging to individual charging sessions (EVs) where all users’ charging preferences must be satisfied all the time.
Algorithm must also prevent grid connection point overload if the full charging power of all charging station in cluster would cause overload.
Operation
One of the charging stations is assigned role of a cluster master. Cluster master serves as a main communication channel between all the charge points in the cluster and is responsible for power management algorithms. This allows cluster to do the power management even if the communication with the back office is lost as all the calculations are done within the cluster.
Algorithm determines the current available for charging as a difference between the actual EV charging currents and the required reduction of grid connection point load.
Cluster management algorithm then distributes the current available to individual EVs with consideration of user's charging preferences and EV on board chargers' characteristics:
Calculation of required current: Charging current needed (per phase) to deliver the required energy till the end of time period available for charging is calculated for each charging session.
Required currents of all charging sessions are summed (per phase).
Calculation of current set points for individual EVs: Current available for charging (per phase) is distributed to individual sessions according to the share of their required current in the total of all required currents in the cluster.
A problem arises when EVs with 1- and 3-phase chargers are currently connected to chargers. Calculation of current set points for individual EVs may result in different phase currents for the same EV with a 3-phase charger; Consequently, the currents assigned to individual EVs must be recalculated to reach a symmetrical load of EVs with 3-phase chargers.
For this recalculation the algorithm considers a “fairness factor”. A high factor means, that all EVs are considered on an equal basis: the share of EV’s set point current in the sum of set point currents of all EVs in the cluster corresponds to the share of EV’s required current in the sum of required currents of all EVs in the cluster.
Additional Options
If the cluster is part of a household, multi-dwelling house or if additional loads are connected to the same connection point Load guard unit should be added to the installation. Load guard can measure values of all the loads in the installation and with this data charging station can manage the charging power.
Additional options for cluster management are supported on the level of back-office, with the use of Etrel Ocean monitoring system.