WO2008113969A1

WO2008113969A1 - Electricity network monitoring

Info

Publication number: WO2008113969A1
Application number: PCT/GB2008/000728
Authority: WO
Inventors: Jens-Otto Ravn Andersen
Original assignee: Ec Power A/S; Jackson, Robert
Priority date: 2007-03-22
Filing date: 2008-03-03
Publication date: 2008-09-25
Also published as: RU2009137483A; GB0705537D0; EP2140532A1; GB2447633A; GB2447633B; RU2470439C2

Abstract

A method of monitoring an electricity distribution network comprising an electricity network connection for supplying power to nodes N1to NN, the electricity network connection S being connected to node N1 and the nodes being connected in series, comprises: starting at the final node NN and progressing along the nodes toward the electricity network connection S, calculating a value representing the amount of electricity supplied to each node from the preceding node based on power consumption for the node and losses in the components between the node and the preceding node; and calculating a value representing the amount of electricity supplied from the electricity network connection to the first node N1 based on power consumption for the node and losses in the components between the electricity network connection and the first node N1.

Description

ELECTRICITY NETWORK MONITORING

The present invention relates to the monitoring of electricity networks.

Electricity networks have been built up over time with the aim of connecting ends users of the electricity to power stations and the like where electricity is produced. Historically, networks of this kind have been designed to distribute power from a small number of large power stations. Thus, the traditional electricity supply system is one-directional: electricity flows from power-plant to high voltage grid, from grid to high voltage network (HVN), from HVN on to medium voltage network (MVN), from MVN on to low voltage network (LVN) and finally from LVN to end- user. Further voltage levels may also be present.

The electricity network in the United Kingdom can be considered as a network of this downward flow type. In the UK there are mainly large power stations with only a comparatively small amount of power being generated by lower output units such as windmills, small-scale combined heat and power units (CHP), etc. The power stations are connected to a national grid that enables power to be distributed to consumers throughout the country as required. Thus, to some extent it acts as a large reservoir of power, although it does not generally store energy (pump storage being a rare exception).

With consumers operating on fixed tariffs, asset and congestion management (ACM) in the true sense is only relevant in relation to the high voltage grid, to which the controllable generators are connected. Down through the voltage levels ACM becomes a matter of simply dimensioning the systems for applicable loads. For this type of system, it is relatively simple to monitor power usage and losses in order to allow the performance of the system to be assessed. Table 1 below illustrates a very simple example where a HVN through a transformer feeds a medium voltage power line, which in turn feeds nodes 1 - 10. Distances (cable length) from, node to previous node (node 1 to transformer) are shown. Table 1.

The nodes may be end-users (i.e. industrial users using medium voltage), they may be sub-radials on the MVN, or they may be transformers for low voltage networks (LVN), in which case the node may be an LVN network of similar shape to the MVN represented here. Likewise, this MVN represents a node on the HVN, which in turn may represent a node on a grid, to which producers are connected.

There is no operational metering in the system, only accumulated readings (monthly, quarterly or yearly) from end-user meters. Customers apply loads as they please, while production is balanced by maintaining grid- and HVN voltage and frequency by increasing or decreasing generator loads. In case of severe HVN problems (i.e. major generator failure), MVN's are disconnected until balance is reestablished ("Brown-out").

The real cost of delivering an amount of electricity at a given time and given point or node in the system is unknown, costs are accumulated and distributed among customers according to agreed tariffs.

Based on these accumulated readings, statistics and information from installers regarding special electrical equipment, system dimensioning is done by use of "best practice", i.e. based on a set of standard assumption and calculations. Direct metering (i.e. on the transformer) is generally only performed in case of complaints from customers suffering problems, such as voltage drop, "flicker" etc.

The real situation might be as shown in Table 2 below, and these figures could vary from the figures arrived at via a best practice assessment. To obtain these figures, a careful analysis of each component in the system would be required. Table 2.

Power flow at each node is shown as "flow leaving previous node to supply this and following nodes". Thus, the node 1 value is power leaving the transformer to supply nodes 1 to 10, the node 2 value represents power leaving node 1 to supply nodes 2 to 10, and so on. The consumption at each node is shown, which could be a metered user consumption for example. Losses are shown as "transformer loss" and "line loss" (cable) and can be calculated as loss per unit entering the component (transformer or power line). Knowing the losses, the real cost of delivering one unit of electricity at each node can be calculated. Such loss and cost figures are highly valuable in assessing the performance of a network and identifying areas where maintenance or upgrading would be most effective in improving network performance.

However, many electricity networks no longer follow the traditional one- directional model as set out above. Smaller scale production, especially environmentally attractive power generation methods such as wind and tidal power, and CHP units, is growing rapidly. The presence of power producers at lower levels of the system is known as embedded or distributed generation.

For example, in Denmark, by contrast to the situation in the UK, around 70 per cent of electricity is produced by the smaller scale producers. The overall arrangement can be described as a distributed system because of the geographical distribution of generating capacity. There are a large number of small CHP units and also wind turbines. The wind turbines produce power whenever the wind blows and there is no centralised control of the smaller generating units such as CHPs. So when Denmark produces more power than it needs it has to export the power, e.g. by means of a sub-sea cable to Norway, or alternatively the large power stations can be shut down. There is only limited control over the supply of power from CHPs and wind turbines to the grid. A few medium sized CHPs and large wind farms may be stopped centrally in case of an extreme situation.

When a substantial amount of embedded /distributed generation is present the flow of power within the network changes radically. Table 3 below illustrates how the situation above changes with two power generators connected to the same MV power line, forming new nodes 4 and 9. The node number as used in Tables 1 and 2 is shown in brackets.

Table 3.

The amount and direction of power flow between the transformer and node 9 is unknown. Indeed, it is not even known whether this particular MVN node constitutes a load or a generator in relation to the HVN. In an embedded /distributed network, electricity is both injected and consumed everywhere in the system, which means that the load on / "real value" of network assets changes significantly. The existing best practice estimates will not operate effectively with this type of embedded generation situation. This gives rise to substantial problems in assessing the performance of the network and it is not possible to effectively manage network assets. When the network cannot be effectively monitored, congestion can appear without warning anywhere in the network, and maintenance and investment into network components and assets cannot be efficiently carried out.

It is thought that this type of arrangement is likely to become more common in future as more power is generated by environmentally friendly means such as wind and wave power. It also favours potentially more efficient technologies such as small scale CHP. In order for this system to work efficiently, both in terms of energy efficiency as well as cost efficiency, a method of accurately monitoring the electricity network is required.

Viewed from a first aspect, the present invention provides a method of monitoring an electricity distribution network comprising an electricity network connection for supplying power to a plurality of nodes N₁ to NN, the electricity network connection being connected to node Ni and the nodes being connected in series, the method comprising: starting at the final node N_N and progressing along the nodes toward the electricity network connection, calculating a value representing the amount of electricity supplied to each node from the preceding node based on power consumption for the node and losses in the components between the node and the preceding node; and calculating a value representing the amount of electricity supplied from the electricity network connection to the first node Ni based on power consumption for the node and losses in the components between the electricity network connection and the first node Nj.

It will be appreciated that number of nodes is not limited beyond the need for multiple nodes. In the limiting case where there are two nodes, the end point node N_N will be the same as the node N₂.

The use of this method allows an analysis of an electricity distribution network to be carried out on the basis of only the power consumption at each node and the losses in the components between the nodes and the electricity network connection.

In a traditional downward flow network, the electricity network connection would always be a supplier of power that is then consumed at the nodes. However, in the present invention the electricity network connection can either supply power to or receive power from the nodes. Thus the electricity between the connection and the first node could be flowing toward the node, which is represented by a positive value of electricity 'supplied' in the discussion below, or flowing from the node toward the electricity network connection point, which is represented by a negative value of electricity 'supplied' in the discussion below. The amount of electricity supplied to each node is similarly represented by a positive value, indicating a consumption of electricity, or a negative value, indicating a production of electricity. The amount of electricity supplied to a node includes the amount of electricity used by that node, as well as electricity transmitted onward to following nodes, i.e. to nodes further from the electricity network connection. Both the electricity used and the electricity transmitted onward can of course take a negative value, hence being in practical terms electricity produced and electricity received. The method provides a way of calculating a figure representing the power flows between components and hence allows the performance of the network to be assessed without the need to exhaustively measure and analyse each part of the network.

Whilst finding particular beneficial effects in embedded networks, the method of the invention applies equally well to less complex networks, and in fact can be used to analyse the performance of the traditional network as well as networks with only a small proportion of distributed generation. Further, as the method of the invention does not rely on an analysis of the actual components and their performance and is independent of the number of nodes, the method can be used to predict the effect of changes in customer load and power production. For example the capability of the system to cope with a sudden load peak can be assess, or the effect of a new power station or wind farm can be predicted. When new nodes are added to a simulated or real network, the method will be able to monitor the expanded network without any modifications.

The measurement of the amount of electricity referred to in the method may be measured in units of current, which allows a direct calculation of losses, although other convenient measurements of the amount of electricity may also be utilised. For example, a measurement of electrical power or energy may be used, and is advantageous as it allows the method to be applied where voltage and current are not constants, for example where a transformer forms part of the connection between nodes.

The power consumption at each node may be the metered consumption of an end user on a LVN or MVN for example, or the consumption of a lower network, such as a LVN transformer connecting onward to an number of users and acting as a node on a MVN. The nodes themselves may be single devices, or they may be groups of devices, such as devices connected to a junction in the network. The electricity network connection may be a transformer connecting the nodes to a higher voltage network. The losses may be line losses in power transmission lines and/or transformer losses.

In a preferred embodiment, the method further includes the steps of: starting at the first node Ni and progressing along the nodes away from the electricity network connection, estimating the total losses at each node with reference to the electricity network connection; and using the total losses to calculate an electricity cost for each node relative to the cost at the electricity network connection.

Thus, advantageously, the method can be used to assess the local losses as well as the local costs. This allows the true cost of any losses to be established and enables maintenance and investment into new infrastructure to be directed where most effective. A large loss at a low cost can be assigned a lower importance than a small loss at a significantly higher cost.

The step of calculating a value representing the amount of electricity supplied from the electricity network connection and the nodes to the following nodes may be performed by:

(a) determining the amount of electricity delivered to the end point node N_N based on the power consumption for the node;

(b) determining the component losses for the connection between the node N_N and the preceding node N_N- i;

(c) calculating the total amount of electricity provided for the node N_N from the connection point of the preceding node N_N-1 based on the amount of electricity delivered to the node N_N and the losses between nodes N_N and N_N-U

(d) determining the amount of electricity used at the preceding node N_N- i based on the power consumption for the node, and then calculating the value representing the amount of electricity supplied to the node N_N-_I by summing the amount Qf electricity used at the node N_N-_I and the amount of electricity provided for the node NN;

(e) repeating steps (b) to (d) for the node N_N-_I and preceding nodes as necessary in order to determine the value representing the amount of electricity provided for each of the nodes from the corresponding preceding node, until node Ni is reached and thus the value representing the amount of electricity supplied from the connection point of node Ni to node N₂ is known; and (f) calculating the value representing the amount of electricity provided for the node N₁ from the electricity network connection based on the amount of electricity delivered to the node Nj and the losses for the connection between the electricity network connection and the node Ni.

The method may include estimating the losses at each node with reference to the electricity network connection by:

(g) calculating a ratio "P/P" for each node, wherein for the first node Ni the ratio P/P is the amount of electricity leaving the electricity network connection divided by the amount of electricity delivered to the node, and for each of nodes N₂ to N_N the ratio P/P is the amount of electricity leaving the preceding node divided by the amount of electricity delivered to the node;

(h) calculating a ratio "Ptnuis/P" f°^{r eacn} node, wherein the ratio Ptran/P is the multiple of P/P for the node with P/P for the preceding node or with P/P for the electricity network connection where the node is the first node Ni, where P/P for the electricity network connection is one; and

(i) determining the total loss for each node as P_tnm/P minus 1.

The total losses may be used to calculate an electricity cost for each node by:

(j) calculating the cost at the first node Ni by multiplying the cost at the electricity network connection by the ratio P/P for node Ni; and

(k) calculating the cost at the following node N₂ and each subsequent node by multiplying the cost at the previous node by the ratio P/P for node N₂ and each subsequent node.

Thus, by a series of simple repeated steps a complex electricity distribution network can be easily monitored. The method of these preferred embodiments is also not limited as to the number of nodes beyond using multiple nodes. Thus, in the limiting case where there are two nodes, the end point node N_N will be the node N₂ as above, and the node N_N- i will be the first node Ni.

The simplicity of the calculation, and the small, amount of data passed from node to node at each step allows real time monitoring to occur. The effect of losses on network efficiency can be established with reference to varying conditions over time periods of days or hours, rather than relying on monthly or even quarterly accumulated values to try to estimate congestion points. For example, a node showing relatively low losses overall may have maximum losses at a period of high cost or high demand, and these losses can have more effect on efficiency than a node showing higher overall losses, where these losses occur over a period of low demand.

The method can be repeated for each voltage level of the network to monitor the whole network. Thus, the electricity network connection may form a node in a higher level network consisting of further similar supplies and other users or producers at that voltage level. Similarly, the node may form a electricity network connection for a lower level network of lower level nodes.

The values for losses and consumption at each node may be stored. This storage may be at a memory located at a calculation device for each node. The values can thus be read by any suitable instrument and transferred to a computer for later analysis. Alternatively, some analysis may be carried out at each node before indicating the results at the node, or transmitting data elsewhere.

The determiiiation of values for each node may be performed remotely. Preferably however, the determination of values is carried out by a calculation device situated at each node.

Viewed from a second aspect, the present invention provides a method of monitoring a node N_x in an electricity distribution network having a plurality of nodes, the method comprising: assessing the amount of electricity consumed by the node N_x and determining the losses occurring in power transmitted from the preceding node N_x-I ; receiving, from the following node N_x+i, the amount of electricity drawn from the node N_x by the following node N_x+i; and, based on the amount of electricity consumed, the transmission losses and the amount of electricity sent to the following node N_x+], calculating the amount of electricity drawn by the node N_x from the preceding node N_x-] and transmitting this value to the preceding node N_x-].

The method of the second aspect enables values local to each node to be determined based only on data transmitted from the adjacent nodes. Thus, knowledge of the entire network is not required. Expansion of a network to include further nodes therefore does not present a problem for the invention.

Preferably the method of the second aspect includes: calculating a ratio "P/P" for the node N_x based on the amount of electricity drawn by the node N_x from the preceding node N_x-I and the transmission losses between the node N_x and the preceding node N_x-i; transmitting the value of P/P for the node N_x to the following node N_x+1; receiving a value of P/P for the preceding node N_x-1; calculating a ratio "Ptmns/P" for the node N_x, wherein the ratio P_tran/P is the multiple of P/P for the node N_x with P/P for the preceding node N_x-i; and determining a total loss with reference to a electricity network connection for the node N_x as Pt_mn_s/P minus 1. The method may further include: receiving a cost value from the preceding node N_x-i; calculating the cost at the node N_x by multiplying the cost at the preceding node N_x-1 by the ratio P/P at the node N_x; and transmitting the cost value for the node N_x to the following node N_x+1.

Thus, the local loss and cost figures can also be calculated without knowledge of the entire network.

Viewed from a third aspect, the present invention provides an apparatus for monitoring a node N_x in an electricity distribution network having a plurality of nodes, the apparatus comprising: a data receiver for receiving, from the following node N_x+1, the amount of electricity drawn from the node N_x by the following node N_x+i ; a calculation device configured to assess the amount of electricity consumed by the node N_x, determine the losses occurring in power transmitted from the preceding node N_x-I and to calculate the amount of electricity drawn by the node N_x from the preceding node N_x-I, based on the amount of electricity consumed, the transmission losses and the amount of electricity sent to the following node N_{x+ 1}; and a data transmitter for transmitting the amount of electricity drawn by the node N_x from the preceding node N_x-] to the preceding node N_{x- 1}.

The apparatus may be arranged to carry out the method of the preferred embodiments of the second aspect as set out above.

The apparatus may be located remote from the node. However in a preferred embodiment, the apparatus is located at the node. As each node can be provided with an apparatus to monitor that node and transmit/receive data from adjacent nodes, a new node can be added to the network at any point without difficulty, as the new node can simply be provided with an apparatus to monitor itself, and can be connected between the adjacent nodes, which may previously have been connected with each other. Viewed from a fourth aspect, the present invention provides an apparatus for monitoring an electricity distribution network comprising an electricity network connection for supplying power to a plurality of nodes Ni to NN the electricity network connection being connected to node Nj and the nodes being connected in series, the apparatus comprising a data processing device configured to carry out the method of the first or second aspects as discussed above.

The apparatus may be arranged to monitor a number of nodes or an entire sequence of nodes in a network. Preferably however the apparatus comprises a monitoring apparatus at each node of the network. An apparatus as discussed above in relation to the third aspect may be provided at each node.

Electricity meters may be provided to monitor electricity consumption at the nodes. Some metering will always be required at the end user, but consumption at nodes above the end user can be calculated by the method above, and then fed into the next network up as a nodal consumption. With meters provided at every end user, the method and apparatus described above can operate based on only the meter readings and basic information on system characteristics to enable losses to be determined. Thus it is not necessary for all nodes of a network to be monitored. However, the use of more meters allows a more precise monitoring, because additional metering allows values calculated from lower down the network to be checked against measured values. Thus, consumption at each node may be monitored by a meter.

The meters may be integrated into the apparatus of the third aspect, allowing direct communication of the consumption to the calculation unit.

Viewed from a further aspect, the invention provides a computer program product containing instructions which when executed will configure a data processing apparatus to carry out the method of the first or second aspect as discussed above.

Viewed from a still further aspect, the invention provides a method of operating an electricity distribution network comprising an electricity network connection for supplying power to a plurality of nodes N₁ to N_N, the electricity network connection being connected to node Ni and the nodes being connected in series, the method comprising: transferring electricity between the electricity network connection and the first node Ni, and between each node and the following node; determining the consumption or production of electricity at each node and the amount of electricity transferred between the first node Ni and the electricity network connection; determining the losses in the transfer of electricity between the electricity network connection and the first node Ni, and between each node and the following node; and monitoring the electricity distribution network or the nodes therein using a method as discussed above in relation to the first or second aspect and the preferred features thereof.

Preferred embodiments of the present application will now be described by way of example only and with reference to the accompanying drawings in which:

Figure l is a representation of a national power generation network;

Figure 2 is a schematic view of nodes on a generic network; and

Figure 3 is a plot illustrating the changes in electricity price over a week.

As an example of an embedded/distributed network, Figure 1 shows a power generation network of the sort used in Denmark. In this example there are five voltage levels. The largest scale power-generation capacity is provided by one or more conventional power stations which supply the order of thousands of megawatts into a 400 kilovolt high voltage grid 13. This network is used to transmit the power over long distances. It is then transformed down to 150 kilovolts and fed to regional network 14. This network also has power generation capacity connected to it such as large offshore wind turbines 15.

The power from the 150 kilovolts network is then transformed down to 60 kilovolts where it is supplied to district networks, which can be viewed as the HVNs discussed above. To this network there may be connected inland wind turbines 16 and large district combined heat and power units 17.

In order to distribute the power to various local areas it is then transformed down to 10 kilovolts to form MVNs 20. Smaller-scale CHP units 18, and smaller inland windmills 21 may also feed power in to this part of the network. It is from this part of the network that large industrial buildings 19 are supplied and these may in some cases have the option of providing power to the network.

Local distribution networks 22 and 23, which operate at 400V are supplied from each medium voltage network 20 via transformers. Local network 22 comprises a number of consumers 22a, 22b, etc. In local network 23 some of the consumers have their own power generating capacity. This includes combined heat and power units 23a and 23b.

It will be seen that in this system power is fed into the distribution network at all different levels. If there is excess power in the network then it may be exported from the 400 kilovolt level 3 by international sub-sea cables.

In accordance with the invention, the network is monitored by calculation devices using data at nodal points. For example, on the MVN 20 there are five nodes, comprising in sequence starting from the transformer which is the electricity network connection for the network: the local network 23, the windmill 21, the industrial user 19, the CHP unit 18 and finally the local network 22. Further nodes could exist at further points away from the transformed beyond the local network 22.

Figure 2 is a schematic representation of a generic network having a electricity network connection S, in the form of a transformer connecting a lower voltage network 24 to a higher voltage network 25. The connection of the electricity network connection S to the higher voltage network 25 forms a node 26 on that network 25. On the lower voltage network 24, three nodes are illustrated, as a sequence of nodes N_x-i, N_x and N_{x+ 1}. Each node has a nodal load, which consumes power from the network. The nodes are connected by transmission lines 27, 28. It will be appreciated that any number of other nodes could be connected preceding and following the nodes N_x-], N_x and N_{x+ 1} shown.

Each node is provided with a calculation unit, or node calculator, which is capable of calculating loss in a distribution network component (i.e. a length of cable, transformer) between the node and the previous node (towards point of common connection) in this network and calculating accumulated loss and unit cost at this node, based on information on: unit characteristics such as loss characteristics, nodal the previous node. The calculation unit passes information regarding total load, comprising the nodal load plus the load of the following nodes to the previous node and passes information regarding accumulated loss and unit cost to the following node.

Returning to the notional network of Table 3, an example calculation will be used to illustrate the monitoring method and the operation of the node calculator. In this case, the "network components" (apart from transformer) are sections of the MVN cable with lengths shown ("dist.")- The results of this calculation are shown below in Table 4.

Table 4

The calculation of losses is easiest understood by starting at the last node and working towards the transformer. Table 5 shows the results of a loss calculation for nodes 7 to 12 from Table 3.

Table 5

The various values shown in Table 5 are as follows:

"Comp. loss" (component loss) is the total loss in the MVN cable from previous node connection point, calculated as current at nominal (transformer) voltage "Total current" is the nodal current plus current for following nodes as calculated at previous node connection point. A negative value means that the direction of flow is towards the transformer / previous node.

"Nodal current" is the nodal current calculated at the previous node connection point. A negative value means the node is a power producer rather than a consumer.

"P/P" is a ratio of the power leaving the previous node connection point to the power entering the present node connection point. If P/P is greater than 1 then the direction of power flow is towards previous node.

"Ptrans/P" ("Ptrans/P") is a ratio of the power leaving transformer to the power entering the present node connection point.

"Total loss" is relative loss per unit at the node referring to the transformer.

"Cost" is the unit cost/value at the node referring to the price at the transformer.

"Ref. current" is a reference value for electrical current delivered to a node, which is calculated at a nominal voltage (transformer voltage, in this example 10 kV) from electrical power (kW). This electrical power value may be obtained from a nodal meter, or it may be the result of a series of calculations like this network calculation and based on end-user meters.

In this example a power factor of 1 is assumed, other power factors may be used, either as fixed values or dynamic values i.e. transmitted from the HV/MV transformer.

The calculation proceeds as set out below, where "*" represents a multiplication operation.

A total reference value for electrical current arriving at a node connection point (c.p.) is found as the sum of "Ref. Current" and current for the next node connection point. This value is not shown in Table 5.

For node 12 c.p. the total reference value = "Ref. Current" as there is no node 13. The component loss ("Comp. Loss") from node 11 connection point (c.p.) to node 12 c.p. is a straight cable loss which can be determined as (current)square * component value (cable unit R) * cable length. In this case the cable component value is set at 0.0001 /unit, and the cable length at 2.4 units. Based on this information the node 12 calculation unit calculates the amount of electricity (in this case the number of units of current) leaving node 11 as follows:

→ Comp. Loss = 0.0001 * 2.4 * 7.506 * 7.506 = 0.0135 — ^•> P/P (Units leaving node 11 c.p. per unit at node 12 c.p.) =

(7.506+0.0135)/(7.506) = 1.0018

— * Total current (amount for node 12 c.p at node 11 c.p. reference.) = 7.5 *

1.0018 = 7.5191

The value for "Total current" is transmitted to node 11. Thus, the reference current arriving at node 11 c.p. equals:

("Ref. current" node 11) + ("Total current" node 12) = 12.715

The node 11 calculation unit then calculates the units delivered from node 10 in the same manner as above:

→ Comp. Loss = 0.0001 * 1.9 * 12.715 * 12.715 = 0.0307 -→ P/P (Units leaving node 10 c.p. per unit at node 11 c.p.) =

(12.715+0.0307)/12.715 = 1.0024

→ Total current (amount for node 11 c.p. at node 10 c.p. reference) = 12.715

* 1.0024 = 12.7459

The value for "Total current" is then transmitted to node 10

Reference current arriving at node 10 c.p.:

("Ref. current" node 10) + ("Total current" node 11) = ,23.138

The node 10 calculation unit then calculates the units delivered from node 9 in the same manner as above:

→ Comp. Loss = 0.0001 * 1.2 * 23.138 * 23.138 = 0.0642 -→ P/P (Units leaving node 9 c.p. per unit at node 10 c.p.) =

(23.138+0.0642)/23.138 = 1.0028

→ Total current (amount at node 9 c.p. reference for node 10 c.p.) = 23.138 *

1.0028 = 23.2025

The value for "Total current" is then transmitted to node 9.

Reference current arriving at node 9 c.p.:

("Ref. current" node 9) + ("Total current" node 10) = -11.439

Here the "Ref. current" is negative, meaning node 9 is a producer, but the calculation of delivery from node 8 c.p. is the same. → Comp. Loss = 0.0001 * 1.0 * (-11.439) * (-11.439) = 0.0131 — > P/P (Units leaving node 8 c.p. per unit at node 9 c.p.) = (-

11.439+0.013 l)/(-l 1.439) = 0.9989

— > Total current (amount at node 8 c.p. reference for node 9 c.p.) = -11.439 *

0.9989 = -11.4254

The value for "Total current" is then transmitted to node 8.

In this case the "Total current" is also negative, giving a "negative" power flow from node 8 c.p to node 9 c.p., which means the power flow direction (and loss) is from node 9 c.p. to node 8 c.p.

Reference current arriving at node 8 c.p.:

("Ref. current" node 8) + ("Total current" node 9) = -6,2292

Here the "Ref. current" is positive, as node 8 is a consumer, while "Total current" from node 9 is negative with power flow from node 9, but the calculation of delivery from node 7 c.p. is the same.

→ Comp. Loss = 0.0001 * 2.7 * (-6.2292) * (-6.2292) = 0.0105 — »^■ P/P (Units leaving node 7 c.p. per unit at node 8 c.p.) = (-

6.2292+0.0105)/(-6.2292) = 0.9983

— > Total current (amount at node 7 c.p. reference for node 8 c.p.) = -6.2292 *

0.9989 = -6.2188

The value for "Total current" is then transmitted to node 7.

In this case the "Total current" is still negative, giving a power flow direction

(and loss) from node 8 c.p. to node 7 c.p.

Reference current arriving at node 7 c.p.:

("Ref. current" node 7) + ("Total current" node 8) = 0,1321

Here the "Ref. current" is positive, as node 7 is a consumer, while "Total current" from node 8 is negative with power flow from node 8, but the calculation of delivery from node 6 c.p. is the same.

→ Comp. Loss = 0.0001 * 3.5 * (0.1322) * (0.1322) = 0.000006 → P/P (Units leaving node 6 c.p. per unit at node 7 c.p.) =

(0.1322+0.000006)/(0.1322) = 1.00005

-→ Total current (amount at node 6 c.p. reference for node 7 c.p.) = 0.1321 *

1.00005 = 0.1321 The value for "Total current" is then transmitted to node 6.

In this case the "Total current" turns positive, giving power flow directions (and losses) from both node 6 c.p. and node 8 c.p. to node 7 c.p.

This procedure is repeated by every node calculator, with the "node 1" - calculator calculating delivery from the HV/MV -transformer. The whole calculation sequence is continuously repeated at short time intervals (for example 1/sec.)

The HV/MV -transformer may itself have a calculator, and can thus form a node with calculator on a HV network.

Likewise, a node on this MV network may be a MV/LV -transformer to a LV network with nodes & calculators, they may also be industrial end-users or they may be generators, but here information for the calculators regarding "Reference current" may in any case come from end-user meters, forming the metering basis of the system.

Having thus established the loss, P/P ratio and total current for each node, the "Total loss" and hence the "Cost" values for each node can be determined. As both "Total loss" and "Cost" (or value) in the calculations refer to the MV side of the HV/MV -transformer, they are easiest understood by starting at the first node from the transformer (node 1) and working towards the end of the line. Table 6 shows the results for nodes 1 to 6, with the calculation proceeding as set out below.

Table 6

As noted above, "P_trans/P" represents the units leaving transformer per unit at the present node c.p. and it is calculated as:

Ptrans/P ⁼ (Ptrans/P value from previous node)* P/P

At the transformer "P_trans/P" = 1. Therefore, at node 1, "P/P" = "P_trans/P". Total loss" is a relative figure in the same manner as "P_trans/P", here meaning the units lost by delivering one unit at this node referring to the transformer. It is calculated as:

TOIaI IoSS = CP_h3nSZP") - !

As a result of this, losses due to power injections inside the network can be given negative values. This could of course be remedied by using absolute values, because they are indeed losses, but in this case negative values have been maintained in order get a clear picture of network congestion "losers" and "winners", i.e. a "negative" loss at a consumer node represents a "winner", but at a producer node it represents a "loser".

"Cost" represents the cost of delivering 1 unit (kWh) of energy from the transformer or the balanced value of electricity from a producer in the network. It is calculated as:

Cost = (Cost previous node)*P/P, because P/P represents "Units leaving previous node c.p. per unit at this node c.p.".

Node 1 calculation unit (c.u.) - receive values for "P_trans/P" and "Cost" from previous node (in this case transformer)

P/P (node I) = LOOOl

→ Ptrans/P = 1*1 0001 = 1.0001 (transformer value = 1 )

→ Total loss = 1.0001 - 1 = 0.0001

→ Cost = 10*1.0001 = 10.001

In this case, only a fraction of the supply comes from the transformer. Therefore, the loss is very small and the cost almost identical to transformer price.

Node 2 c.u. - receive from node 1 values for "P_tran/P" (=1.0001) -and "Cost" (=10.001)

P/P (node 2) = 0.9983

→ P trans./P = 1.0001*0.9983 = 0.9985

→ Total loss = 0.9985 - 1 = -0.0015

→ Cost = 10.001*0.9983 = 9.9846

For node 2, all of the supply comes from node 3 c.p. and there is power flow to node 1 c.p., giving a P/P < 1 and, in this case, "negative" total loss. Being a consumer, this makes node 2 a "winner". The cost is reduced compared to the transformer price.

Node 3 c.u. - receive from node 2 values for "Pt_13nJP" (=0.9985) -and "Cost" (=9.9846)

P/P (node 3) = 0.9971

→ Ptmns/P = 0.9985*0.9971 = 0.9956

→ Total loss = 0.9956 - 1 = -0.0044

→ Cost = 9.9846*0.9971 = 9.9561

For node 3, all of the supply comes from node 4 c.p. and there is power flow to node 2 c.p., giving a P/P < 1 and, again, "negative" total loss. Being a consumer, this makes node 3 a "winner".

Node 4 c.u. - receive from node 3 values for "P_trans/P" (=0.9956) -and "Cost" (=9.9561)

P/P (node 4) = 0.9974

→ Ptrans/P = 0.9956*0.9974 = 0.9930

→ Total loss = 0.9930 - 1 = -0.0070

→ Cost = 9.9561*0.9974 = 9.9305

Node 4 is a producer with power flow to node 3 c.p., giving a P/P < 1 and still more "negative" total loss. Being a producer, where "Cost" = "Value", this makes node 4 a "loser".

Node 5 c.u. - receive from node 4 values for "P_trans/P" (=0.9930) -and "Cost" (=9.9305)

P/P (node 5) = 1.0051

→ Ptrans/P = 0.9930*1.0051 = 0.9981

→ Total loss = 0.9981 - 1 = -0.0019

→ Cost = 9.9305*1.0051 = 9.9807

Node 5 is a consumer with all supply coming from node 4 c.p., giving a P/P > 1 , but total loss remains "negative". Being a consumer, this makes node 5 a "winner".

This procedure is repeated by every node calculator through to the last calculator in the network (node 12). The whole calculation sequence is continuously repeated at the same time intervals (like 1/sec.) The HV/MV -transformer may itself have a calculator, and form a node with calculator on a HV network, from which "Cost" is obtained.

Likewise, a node on this MV network may be a MV/LV -transformer to a LV network with nodes and calculators, they may also be industrial end-users or generators, but here the last calculator will be embedded in the end-user meter, which will use the calculated "Cost" directly for counting money as well as kWh's and pass information on to price-responsive end-user equipment, forming the responsive basis of the system.

The true cost (as opposed to power loss) basis for customer billing and congestion management is very important, as electricity prices are subject to extreme volatility. There is no point in investment if the losses which are avoided by the investment only occur when prices are very low (or even at zero), for example, in the case of high wind-turbine production at night.

Figure 3 shows a plot of energy prices in a normal week in the Kontek area (part of Nord Pool in Denmark). Prices range from zero to 149 0MWh (=14.9 cent/kWh, which is a "low" peak. The record price level for this area is more than 10 times higher.

It is quite obvious that power lost at zero price means next to nothing, only sheer power system capacity is of importance in a situation like that. On the other hand, at 14.9 cent/kWh at HV-grid level power losses should obviously be minimised.

In the "true cost" system as set out above, customers may be consumers, producers - or both. Every node may contain production equipment of varying capacities and varying operational costs, every network may have different characteristics for capacity/loads/losses varying in time and grid connections may be strong or weak with varying injections of wind-power. With this calculation system, price-responsive customer equipment can optimise an electricity supply system down to the last component by continuously adjusting loads and inputs in pursuit of optimum performance for the individual and for the whole network alike. In this way, congestion management is turned into an ongoing process.

The shortcomings of the present "fixed tariff system can be seen from Figure 3, where price peaks represent "shortage" of relevant generating capacity and a fall to zero price represents "excess" capacity. In the fixed tariff system the customers pay the same price per unit at all times. In the specific example of the Kontek area, this might be around 6 cent/kWh in addition to grid & network tariffs and taxes. Even if price-responsive customer equipment accounting for only a fraction of the total load was present, the impact on peak prices would be very substantial, as the price / load ratio in extreme peak situations can be 50:1 or higher. Consumers could react to increased prices (reflecting a shortage of capacity) by reducing non-essential consumption, and producers could react by increasing production where possible.

Further, with a calculation unit for the transformer as well, the entire network supplied by the transformer can be related to the HV network as one virtual powerplant with extreme response capability. For example, the Kontek peak price of 14.9 cent/kWh illustrated in Figure 3 could, including transmission and transaction costs, give a true price of 18 cent/kWh on the HV side of the HV/MV transformer. In response to this price level, in the responsive network the CHP 's at nodes 4 and 9 will be in operation, while other price-responsive equipment at nodes 3 and 10 could reduced consumption. In consequence, the MVN can act as a generator in relation to the HVN. Power losses are very small, but, because of the high price level, they still economically significant (up to 0.47 cent/kWk). Table 7 below shows the results of these changes.

Table 7.

At the other end of the scale a zero price could result in 2 cent/kWh on the HV side. In that situation, "CHP-nodes" 4 and 9 would obviously stop producing. In the example illustrated in Table 8 below node 4 has taken this further and now uses the cheap electricity for power-to-heat, while nodes 3 and 10 are back to normal consumption. In this case, the power losses are very substantial with only 85% of HV-delivery reaching the nodes, but the economical "loss" is less than before (up to 0,43 cent/kWh), and everybody is getting very cheap energy.

Table 8.

It is clear that price volatility is greatly reduced under such circumstances. In a fully developed scenario, with the entirety of a network monitored and controlled as set out above, the entire supply system function would be turned upside-down: system balance would primarily be maintained by end-users absorbing fluctuations in supply, for example from wind-power injections in the HV grid. Existing HV transmission connections to neighbouring areas in a larger grid (for example, EEX - E.ON, DK east) would suddenly prove very adequate.

Congestion management is consequently greatly aided by the monitoring and calculation procedures set out above.

There are also advantages in asset management of electricity networks. The requirements for asset management are very different from the requirements for congestion management. In this context, asset management is all about gatheπng relevant information in order to establish a solid ground for decision-making with regard to investment.

This has two implications with regard to this calculator system:

Operational values from the congestion management functions must be accumulated, and The calculators must be able to communicate beyond "nearest neighbour".

In a very simple version, the calculator simply accumulates values for component power load, "economical load", power losses and economical losses - in a memory accessible by a suitable device for read-out, whereby operational costs of that particular component could be established.

In a more advanced version, the calculator compiles the accumulated values in relevant fractions, i.e. economical losses at different power loads, based on which the cost/value of an alternative component could be assessed.

In a still more advanced version, the calculator performs a parallel set of calculations and compilation of values for a virtual component and/or a virtual load, based on which the cost/value of the virtual component could be established. hi a still more advanced version, the calculator communicates the parallel set of calculations including Component loss to neighbouring calculators and receive, process and communicate parallel sets of information from neighbouring calculators in the same manner as the basic, operational procedure. Based on this, the network cost/value of a virtual load and a virtual component could be established.

In a still more advanced version, the HV/MV transformer calculator communicates the above information to a computer accessible for the network operator.

In a still more advanced version, the calculator can, on request or at fixed time intervals (i.e. 1 min.), communicate an information package with compiled, accumulated values to the computer, where they could be stored with time-stamp as "nodal information" and re-compiled and stored as network information.

A value for total transmission loss in a network (here 1OkV) can be calculated as:

Network transmission loss = (sum(Comp. Losses)) *V3*1 OkV (kW), assuming that a power factor of 1.

Other power factors (metered or calculated) may of course be used.

An alternative value for the total transmission loss in a network can also be calculated as: Network transmission loss* = (Transformer load (MV-side)) - (sum(nodal loads)). However, this would call for metering on the MV side of the transformer. On the other hand, such a meter could be used to calibrate the individual calculation units by calculating a "network power factor" based on the difference between "Network transmission loss" and "Network transmission loss*".

The total economical loss in a network then can be calculated as: (Network transmission loss)* (Unit price at transformer MV side)

Claims

CLAIMS:

1. A method of monitoring an electricity distribution network comprising an electricity network connection for supplying power to a plurality of nodes Ni to N_N, the electricity network connection being connected to node N₁ and the nodes being connected in series, the method comprising: starting at the final node N_N and progressing along the nodes toward the electricity network connection, calculating a value representing the amount of electricity supplied to each node from the preceding node based on power consumption for the node and losses in the components between the node and the preceding node; and calculating a value representing the amount of electricity supplied from the electricity network connection to the first node Ni based on power consumption for the node and losses in the components between the electricity network connection and the first node Ni.

2. A method as claimed in claim 1, comprising: starting at the first node Ni and progressing along the nodes away from the electricity network connection, estimating the total losses at each node with reference to the electricity network connection; and using the total losses to calculate an electricity cost for each node relative to the cost at the electricity network connection.

3. A method as claimed in claim 1 or 2, wherein calculating a value representing the amount of electricity supplied from the electricity network connection and the nodes to the following nodes comprises:

(b) determining the component losses for the connection between the node N_N and the preceding node N_N- i; (c) calculating the total amount of electricity provided for the node NM from the connection point of the preceding node N_N- i based on the amount of electricity delivered to the node N_N and the losses between nodes NN and N_N-_I;

(d) determining the amount of electricity used at the preceding node N_N-_I based on the power consumption for the node, and then calculating the value representing the amount of electricity supplied to the node NN-_I by summing the amount of electricity used at the node N_N-_I and the amount of electricity provided for the node NN;

(e) repeating steps (b) to (d) for the node N_N-_I and preceding nodes as necessary in order to determine the value representing the amount of electricity provided for each of the nodes from the corresponding preceding node, until node N₁ is reached and thus the value representing the amount of electricity supplied from the connection point of node Ni to node N₂ is known; and

(f) calculating the value representing the amount of electricity provided for the node Ni from the electricity network connection based on the amount of electricity delivered to the node Ni and the losses for the connection between the electricity network connection and the node Ni.

4. A method as claimed in claim 3, wherein the losses at each node with reference to the electricity network connection are estimated by:

(h) calculating a ratio "P_trans/P" f°^{r ea}ch node, wherein the ratio Ptrans/P is the multiple of P/P for the node with P/P for the preceding node or with P/P for the electricity network connection where the node is the first node Ni, where P/P for the electricity network connection is one; and

(i) determining the total loss for each node as P_trans/P minus 1.

5. A method as claimed in claim 4, wherein the total losses are used to calculate an electricity cost for each node by:

(j) calculating the cost at the first node Nj by multiplying the cost at the electricity network connection by the ratio P/P for node Ni; and

6. A method as claimed in any preceding claim, comprising storing consumptions and loss values for each node for use in determining network losses.

7. A method as claimed in any preceding claim, wherein the electricity distribution network comprises distribution networks at different voltage levels, and the method is repeated for nodes of each voltage level of the network to monitor the whole network.

8. A method of monitoring a node N_x in an electricity distribution network having a plurality of nodes, the method comprising: assessing the amount of electricity consumed by the node N_x and determining the losses occurring in power transmitted from the preceding node N_x-i; receiving, from the following node N_x+i, the amount of electricity drawn from the node N_x by the following node N_{x+ 1}; and, based on the amount of electricity consumed, the transmission losses and the amount of electricity sent to the following node N_x+], calculating the amount of electricity drawn by the node N_x from the preceding node N_x-1 and transmitting this value to the preceding node N_x-] .

9. A method as claimed in claim 8, composing: calculating a ratio "P/P" for the node N_x based on the amount of electricity drawn by the node N_x from the preceding node N_{x- 1} and the transmission losses between the node N_x and the preceding node N_x. i; transmitting the value of P/P for the node N_x to the following node N_x+i; receiving a value of P/P for the preceding node N_x-1; calculating a ratio "Pfcms/P" for the node N_x, wherein the ratio P_trans/P is the multiple of P/P for the node N_x with P/P for the preceding node N_x-i; and determining a total loss with reference to an electricity network connection for the node N_x as Ptrans/P minus 1.

10. A method as claimed in claim 9, comprising: receiving a cost value from the preceding node N_x-i; calculating the cost at the node N_x by multiplying the cost at the preceding node N_x-1 by the ratio P/P at the node N_x; and transmitting the cost value for the node N_x to the following node N_{x+ 1}.

11. An apparatus for monitoring a node N_x in an electricity distribution network having a plurality of nodes, the apparatus comprising: a data receiver for receiving, from the following node N_{x+ 1}, the amount of electricity drawn from the node N_x by the following node N_{x+ 1}; a calculation device configured to assess the amount of electricity consumed by the node N_x, determine the losses occurring in power transmitted from the preceding node N_x-I and to calculate the amount of electricity drawn by the node N_x from the preceding node N_x-1, based on the amount of electricity consumed, the transmission losses and the amount of electricity sent to the following node N_{x+ 1}; and a data transmitter for transmitting the amount of electricity drawn by the node N_x from the preceding node N_x-I to the preceding node N_x-].

12. An apparatus as claimed in claim 11, wherein: the calculation unit is configured to calculate a ratio P/P for the node N_x based on the amount of electricity drawn by the node N_x from the preceding node N_x-I and the transmission losses between the node N_x and the preceding node N_x-1; the transmitter is for transmitting the value of P/P for the node N_x to the following node N_{x+ 1}; the receiver is for receiving a value of P/P for the preceding node N_x-1; and the calculation unit is configured to calculate a ratio P_tran/P for the node N_x, wherein the ratio Ptram/P is the multiple of P/P for the node N_x with P/P for the preceding node N_x-1, and to determine a total loss with reference to an electricity network connection for the node N_x as P_tran/P minus 1.

13. An apparatus as claimed in claim 12, wherein: the receiver is for receiving a cost value from the preceding node N_x-] ; the calculation unit is configured to calculate the cost at the node N_x by multiplying the cost at the preceding node N_x-I by the ratio P/P at the node N_x; and the transmitter is for transmitting the cost value for the node N_x to the following node N_x+).

14. An apparatus as claimed in claim 11, 12 or 13, wherein the apparatus is located at the node.

15. An apparatus for monitoring an electricity distribution network comprising an electricity network connection for supplying power to a plurality of nodes N₁ to N_N, the electricity network connection being connected to node N₁ and the nodes being connected in series, the apparatus comprising a data processing device configured to carry out the steps of any of claims 1 to 10.

16. An apparatus as claimed in claim 15, wherein the data processing device comprises an apparatus as claimed in any of claims 11 to 14 for each node.

17. An apparatus as claimed in any of claims 11 to 16, comprising a meter for determining the power consumption of the or each node.

18. A computer program product containing instructions which when executed will configure a data processing apparatus to carry out the method of any of claims 1 to 10.

19. A method of operating an electricity distribution network comprising an electricity network connection for supplying power to a plurality of nodes Ni to N_N, the electricity network connection being connected to node Ni and the nodes being connected in series, the method comprising: transferring electricity between the electricity network connection and the first node Ni, and between each node and the following node; determining the consumption or production of electricity at each node and the amount of electricity transferred between the first node Ni and the electricity network connection; determining the losses in the transfer of electricity between the electricity network connection and the first node Ni, and between each node and the following node; and monitoring the electricity distribution network or the nodes therein using a method as claimed in any of claims 1 to 10.

20. A method of monitoring an electricity network substantially as hereinbefore described.

21. An apparatus for monitoring an electricity network substantially as hereinbefore described.