WO2005075894A1

WO2005075894A1 - Thermal energy distribution system

Info

Publication number: WO2005075894A1
Application number: PCT/SE2005/000161
Authority: WO
Inventors: Svend Frederiksen; Sture Andersson; Ingemar BJÖRKLUND; Tommy Gundmundson; Janusz Wollerstrand
Original assignee: Svend Frederiksen; Sture Andersson; Bjoerklund Ingemar; Tommy Gundmundson; Janusz Wollerstrand
Priority date: 2004-02-10
Filing date: 2005-02-10
Publication date: 2005-08-18
Also published as: SE0400279L; CA2555494A1; SE0400279D0; EP1718901A1

Abstract

A thermal energy distribution system for supplying several buildings with thermal service, in which the buildings are fully detached from each other. The system comprises a branching station and several conduits forming an essentially unbroken connection from the branching station to a thermal connection close to each of the building. The conduit has essentially the same cross-section over the entire length. At least two conduits extend adjacent to each other over a first portion (CONa), and extend to each building over a third portion (CONc), and have a second portion (CONb) forming a transition between the first and the portion. One group of the first portions (CONa) of the conduit extend within or along a street.

Description

APPLICANT: SVEND FREDERIKSEN et al

TITLE: THERMAL ENERGY DISTRIBUTION SYSTEM FIELD OF THE INVENTION Thermal energy systems, typically in the form of networks of conduits forming part of a district heating and / or district cooling system supplying a few, or more commonly, many buildings with heat and / or cold for space heating / cooling, as well as hot service water inside buildings.

BACKGROUND OF THE INVENTION In a number of countries, district heating (in the UK sometimes termed: 'community heating') systems for many years have expanded successfully; in many cases, district heating comprises heating services to a majority of all buildings. In later years, a number of district cooling systems, for air conditioning and other cooling services inside buildings, have been built as well, often alongside with existing district heating systems. In this patent application, the term: 'thermal energy distribution systems' will be used, to stress the generality of the invention. As will become appar- ent, the invention can be adapted to pure district heating systems, to pure district cooling systems, and to combinations of the two types of system. Usually one or more heated or cooled, pressurised water fluid flows are maintained inside channels of conduits, mainly installed underground to form networks . Heat exchange equipment is often installed inside buildings connected to the networks, but sometimes hydronic space heating systems are connected directly to the network, to save equipment and to avoid tempera- ture losses across heat exchange surfaces. Hot service water can be produced centrally to be distributed to, but is normally provided locally in the building by heating of cold drinking water, either in a heat exchanger or via a coil or some other heat transfer sur- face inside a hot water storage tank. Although water is by far the most common carrier of thermal energy for heating or cooling, attempts have been made to use other fluids or fluid-like flows, such as for instance ice-slurry flows for district cooling. Thermal energy distribution systems tend to be costly and time consuming to install and may cause inconvenience to traffic during underground installation, especially in centres of cities. In built-up areas of relatively low heat / cold load density (amount of thermal services per square unit of land) , for instance areas dominated by single-family houses, specific installation costs (costs per unit of heat / cold connected load) tend to be rather high, sometimes making thermal energy distribution less competitive when there is an alternative solution to the heating / cooling problem, such as for instance a local boiler inside the building . Therefore, much work has been invested into developing distribution systems with as a high a degree of pre-fabrication as possible. The introduction some 30 years ago of pre-insulated, plastic shield pipe systems represents a major step forward in this direction. In areas of low thermal load density, and elsewhere, where relatively small cross-sections of con- duits are applicable, flexible conduits are nowadays often used instead of stiff conduits. One of the advantages of flexible conduits is that they can be supplied in great lengths, to be rolled out from a spool on- site, avoiding a lot of joints, compensators and other piping elements. Another attractive feature is that flexible pipes can rather easily be adapted to follow curved lines, for instance to circumvent obstacles (trees of gardens, for instance) on their way through the landscape. A third attraction with flexible pipes is that they, as flexible electrical cables, optical cables, etc., are more adapted to speedy installation underground, such as horisontal drilling of holes or automated processes for trenching, such as e.g. illustrated by the machine disclosed in US 6,651,361. Still, installation of thermal energy distribution conduits calls for quite an amount of on-site work, for instance caused by the necessity of establishing branching by T-elements at many points of a network, such as where service pipes lead up to buildings. A particularly problematic case can arise when at some later stage of network expansion there is a wish to connect an extra building, already situated or new- built amongst already connected buildings. An energy company would usually welcome such a situation, since a new customer will become a contributor to amortisation of investments already made, and the relative size of heat losses vs. heat deliveries will decrease. On the other hand, connecting new service pipes to an existing network usually is more demanding in terms of in-situ work for T-branching, breaking up existing pavements, etc. Special techniques for establishing a branching on an existing network conduit have indeed been invented, but the result is not always satisfactory. Sometimes blind T-branching elements are installed from the be- ginning, to facilitate later connections, but there is an added first cost associated with this, and such branching elements do not necessarily turn out to have been located optimally. Another problematic issue, especially with single- family houses connected to thermal energy systems, is metering of heat and / or cold supplies . Although much advance has been made in metering technology, for instance by the introduction and refinement of ultrasonic flow meters, metering of thermal services is ge- nerally significantly less accurate than metering supplies of electric energy, and meters need regular checks, which is rather time consuming, due to the substantial amount of manual work being involved in this . In the case of bigger buildings, when there is a need for personnel of the energy company to get access to meters or other network equipment inside the building, there will usually be a caretaker with whom appointment can be made for access during day-time hours. But in the case of single-family houses, often nobody is at home in daytime hours. A solution can be to install meters in a casing outside the building, to be opened by a key; but such a casing will not always be appreciated from an architectural point of view. A method of arranging underground pipes for district heating that has been tested is based on placing un-insulated pipes made of cross-bounded polyethylene (PEX) in grooves of insulation blocks made of expanded polystyrene (EPS) . Due to the simple way such a system can be arranged it offers a number of interesting possibilities compared with state-of-the art systems, for instance reduced installation time. Still, tested systems require branching of pipes which call for special elements, and where branchings occur they represent a potential weakness of the system. A prior art document EP 0027676 discloses a pipe system for conveying a heating fluid to a plurality of edifices or houses to be separately heated, by supply and return pipes extending from a common division gully, avoiding branchings of the pipes to be arranged directly in the ground by instead arranging branchings in one or more division gullies, each gully within a container. Such an arrangement can speed up the time needed to install the system, as well as further advantages pointed at in the document. A subclaim (7) and a drawing (fig. 2) shows that when a block of houses or edifices placed adjacent to each other, sharing a common partition walls , pipes extending from a gully to such a group of houses can be accommodated in a single, common casing. However, the general principle of the invention, as it would apply to separate houses or edifices, teaches that pipes be laid separately in the ground, with an increasing distance between pipes all the way from a division gully to such a groups of houses, as can be seen from fig. 1. This calls for a lot of ground work and is a serious limitation of the invention, since built-up areas, although they may comprise groups of buildings sharing common partition walls, equally often comprise buildings being fully detached from each other. The purpose of the invention is to develop a new kind of thermal energy distribution system, applicable to buildings and edifices mainly situated apart from each other within a built-up area, which: * Increases the degree of pre-fabrication of con- duits further from what has been achieved by state-of- the art technology and thereby reduces the need for site work, not only when installing equipment in an initial phase of system development, but also in later phases when there a need to connect further buildings arises, especially within an existing supply area, * Speeds up installation work, which will tend to reduce first costs and reduce disturbance to traffic during installation, * Reduces the risk of malfunction caused by defi- ciencies in manual work on site, * Increases possibilities of standardisation and mass-production in pre-fabrication of conduits and other equipment, * Reduces first costs for metering and facilitates meter reading, as well as checks of metering accuracy / reliability, * Makes cheap and fast detection of leaks possible, thereby reducing the need for heat exchangers inside buildings, * Is suitable for occasional or continuous, centralised control of individual building fluid flow supply, whenever such control is suitable and is in accordance with good service of customers, - All adding up to a cheaper, more reliable sys- tem, especially when systems provide thermal energy services to buildings of low heat density areas, such as single family dwellings .

SUMMARY OF THE INVENTION As a general statement it can be said that the invention utilises the potential of flexible district heating / cooling conduits a major step further from state-of-the art practice. A key idea of the invention is to concentrate branching of network elements to a single, or a limited, number of branching stations, STA, from which flexible conduits, CON, typically of equal outer cross- section, run adjacent to each other to form an assembly of conduits extending from the branching station, each conduit leaving the assembly of conduits by a curvature, typically to proceed along a line running roughly perpendicular to the initial main direction. By using this type network configuration it becomes possible to establish an unbroken line of conduit all the way from the branching station, whereby one omits many T- branchings directly buried in the ground of a conventional pre-fabricated system. A known method of avoiding T-branchings buried in the ground relies on adopting a network topology by which a series of buildings are connected by a chain of conduits, i.e. a conduit leads up to the first building, from which a next conduit leads to the next building, etc. This is in contrast to conventional network topology in which service conduits, branching off from a main conduit, lead up to each building. The chain-type topology does not omit branchings, but they can be established inside buildings which simplifies work and omits a number of weak points, whereby joints of sleeve pipes are exposed to the ground and in-situ after-insulation has to be done at T-branchings . Such chain-typology appears especially attractive with semi-detached houses, where conduits can be installed mainly in basements of buildings. However, with fully detached houses the chain arrangement involves conduits not solely serving a given building to be installed in the ground surrounding the building. Sometimes it even becomes necessary to install conduits on private property ground owned by people who are not connected to the district heating or cooling system. This in practice can cause a lot of trouble to the energy service company. Installation of main pipes in, or in conjunction with, public roads is therefore generally preferred. At first look some arguments seem to speak against the arrangement proposed by the invention: In compari- son with an equivalent conventional network, the new type of distribution system requires a larger total length of conduits, which may tend to increase both heat and pressure losses. Also, the assembly of conduits running in parallel tends to constitute a bigger combined cross-section than an equivalent conduit of a conventional system, maintaining the same amount of fluid flow to branch out in T-connections . Thus, when bundles of parallel conduits carrying heating fluids are found in prior art, use of such bundles is being restricted to arrangements within buildings, sometimes extending across common building walls of a group of houses . But, as will be explained and exemplified in detail below, the disadvantages pointed out can bemore than outweighed by many advantages, and there is potential for limiting the increased size of the combined cross-section of conduits running parallel. In particular it will be made clear that the adoption of superinsulation, i.e. insulation involving vacuum inside insulation materials, in designs of conduits, not only opens up for the possibility of reducing heat losses, but can also be utilised for reducing the size of an assembly of co-extending conduits significantly. As has been pointed out, T-branchings represent quite an amount of site work and are potentially weak points of a conventional system, so that replacing them by simple curvatures of conduits significantly speeds up installation and increases system reliability. The problem of connection of buildings at a later stage can be handled rather easily by drawing an extra conduit from the branching station, where hydraulical connection is established with a minimum of work to be done. Use of more than 2- lines, for instance 4-line systems with a separate loop for central provision of hot water services in buildings, becomes less complicated in terms of on-site work on networks. Another advantage of the invention is that the number of conduit sizes can be reduced significantly, compared to standards in state-of-the art conduit tech- nology. This is favourable to pre- and mass-fabrication, which will lower costs. In the branching stations, metering equipment can be concentrated, as opposed to conventional systems where metering equipment has to be installed in each building. This lowers first costs for metering, facilitates meter reading, and makes convenient on-site accuracy checks of meters possible. In addition the invention opens up for the use of a whole array of further, innovative developments, which can roughly be divided into two sub-groups : 1. New types of conduits can be used since there is no longer a need for making T-branchings which could be problematic to design for with the particular type of conduit, 2. Leak checks and further check procedures can be performed conveniently in the branching stations .

BRIEF DESCRIPTION OF DRAWINGS Further objects, features and advantages of the invention will be described below with referenc to embodiments shown on the attached drawings, in which: Fig. 1 is a simplified view from above of a first embodiment of the invention. Fig. la is another simplified view from above, showing how the system of the invention can be arranged in an urban landscape. Fig. 2 is an (compared to fig. 1) enlarged, cross- sectional view of the first embodiment. Fig. 3 is schematic showing the branching station of the first embodiment more in detail. Fig. 4a is a cross-sectional view of an assembly of conduits, and a longitudinal section of a single conduit, according to a second embodiment of the inven- tion. Fig. 4b is an enlarged view of a detail of the second embodiment Fig. 5a is a cross-sectional view of of an assembly of conduits according to a third embodiment of the invention. Fig. 5b is an enlarged, cross-sectional view of a single conduit of the third embodiment. Fig. 6 is a cross-sectional view of a fourth embodiment of the invention, being the first of several examples of how super-insulators can be integrated into conduit design. Fig. 7 is a cross-sectional view of a single conduit according to a fifth embodiment of the invention. Fig. 8a is a cross-sectional view of a single con- duit according to a sixth embodiment of the invention. Fig. 8b is a schematic of a 2-line system, supplemented by a vacuum line, according to the sixth embodiment . Fig. 9 is a schematic of a less common, but per se known type of 2-line system, which can be incorporated into the invention according to a seventh embodiment. Fig. 10 is a schematic of a 4-line system according to an eighth embodiment of the invention. Fig. 10a is an alternative schematic of a 4-line system according to the eighth embodiment of the invention . Fig. 11 is a drawing of a branching element according to any embodiment of the invention. Fig. 12 is a cross-sectional view of a ninth em- bodiment of the invention. Fig. 13a is a cross-sectional view of a tenth embodiment of the invention. Fig. 13b is an enlargement of one conduit of the tenth embodiment. Fig. 14 is a longitudinal sectional view from above of the tenth embodiment. Fig. 15 is a longitudinal sectional view of an eleventh embodiment of the invention. Fig. 16 is a view from above of the eleventh embodiment . Figs. 17a-c are three cross-sectional views of the eleventh embodiment. Figs. 18a-g are seven views of a twelvth embodiment of the invention. Figs. 19a-f are six views of an apparatus and a method of arranging the twelvth embodiment of the invention . Figs. 20a-e are five views of a thirteenth embodiment of the invention.

DETAILED DESCRIPTION OF EMBODIMENTS Figs. 1, la, 2, and 3 together show a first embodiment of the invention. Fig. 1 is a top view, for simplicity showing each conduit, CON / CON*, schematically as a single line, although, as shown in the sectional view given in fig. 2, each conduit comprises 2 channels, CHA, one being part of a forward line sending out fluid (typically pressurised water) flow to the building, BUILD, in question, the other line returning flow from the building. The channels are surrounded by heat insulating material, INS, typically contained within an outer shield pipe (here squared) . Fig. 3 shows how the per se known, closed loop type 2- line system is incorporated into the first embodiment and gives a more detailed, schematic view of the branching station, STA, including metering equipment, as well as monitoring equipment inside the building Fig. 1 shows a hierarchical system, in which the connection principle of the invention (as an example) has been utilised on two levels: The lowest level, in which all conduits lead up to buildings, is represented by the distribution system at the top, termed DISTR. Below, there are two more distribution systems at this level. From each lowest-level distribution system conduits, CON (and CONBIG) branch out from a branching element, BRA, inside a branching station, STA. At the next hierarchical level, distribution system DISTR*, conduits CON* branch out from a fourth branching sta- tion, STA*. Three of these conduits extend to the lower-order branching stations, STA, whereas a single conduit, CONBIG* extends to a single, big building, BUILDBIG*. All buildings, BUILD, could be single-family dwellings, while the two bigger buildings, BUILDBIG and BUILDBIG* could be multifamily buildings, office buildings, a school, etc. Inside branching stations, BRA, distribution systems, DISTRl ..., are connected to the higher-order level system, DISTR*, via heat exchangers, HE. The detailed geometry adopted in fig. 1 is a little schematic and more regular than what will typically be found in practice. One of the deviations from the regularity is that all conduits in DISTR are shown to bend by an S-like curvature close to the branching station, STA, where they are arranged adjacent to each other. Each conduit CON is composed of three parts: A first portion, CONa, arranged adjacent to other conduits, a small curved, second portion, CONb, leading the conduit away from the other conduits, and a third portion, CONc, running roughly perpendicular to CONa and up to the building, BUILD, in question . Fig. la shows a top view of an arrangement by which conduit first portions CONa are arranged within or along a STREET, forming part of a MAIN extending from a branching station, STA. Conduits are indicated in a way that is further simplified than the way adopted in fig. 1: Where conduit portions CONa of different conduits are arranged adjacent to each other, they are indicated by a single line only. This simplification has been made to provide a simple, geometrically realistic impression of how a system according to the invention can preferably be arranged in an urban landscape. At the top of the figure part of a main STREET* is shown. A smaller STREET extends downwards (only an inlet part of STREET is shown, i.e. it extends further downward than shown in the figure) . Both streets are furnished with a pavement, PAV / PAV*, on both sides of the street proper. Close to the right-hand side of the corner where the streets meet a branching station, STA, according to the invention has been arranged on the ground or (wholly or partly) below the ground surface. Thermal fluids are led to STA by conduits CON* extending below pavements PAV* and across STREET. From STA an assembly of conduits, CON, extend down STREET, under pavement PAV, such that all conduit portions CONa extend below PAV. Departing portions, CONb, of conduits are situated wholly or partly below PAV, depending on the size of curvatures of conduits. Conduit parts CONc, leading up to buildings, BUILD, extend either to the right or to the left; in the latter case parts of CONc portions extend below the street proper. Whether PAV should be considered a part belonging to, or not belonging to, STREET, is a matter of definition. Also, PAV may or may not be part of the ground property belonging to a certain building, as defined by the limits, LIM of the property. Often arranging conduit parts CONa below the pavement will be convenient in that on the one hand there is a minimum of interference with an asphalt layer of STREET, and on the other hand legal conditions will permit an energy company to arrange CONa portions without separate agreements with each property owner. This can be especially convenient when there are property owners who prefer not to become district heating customers and therefore naturally may not be willing to permit that parts of the system are arranged within their properties.

Furthermore, pavements can often easily be re-laid in such a way that there will be no visible signs of the district heating system in the street when site work has been finished. In some cases it may be cheaper or more convenient to arrange CONa portions in the middle of the STREET, as indicated by alternative 1, ALT1. This may happen if the ground below pavements is already too crowded' with other types of mains, such as town's water lines, sewage pipes, underground electricity wires, optical fibre wires etc. In further circumstances it may instead be beneficial to arrange CONa portions wholly or partly outside the STREET proper, viz. along STREET, as indicated by alternative ALT2. This may for instance be the case if there is a piece of ground between a pavement and the limits of properties belonging to building owners. By moving CONa away from the pavement it may be permissible to dig less deep when arranging conduits, since local regulations often prescribe a certain minimum height of an overhead layer of soil when mains are arranged within streets, including pavements. Finally, ALT3 shows how one by arranging CONa portions below pavements on both sides of STREET can minimise the interference of underground work with asphalt layers of streets. In this case an extra branching station, STA', has been added. As an example, fig. 2 shows that conduits can be arranged directly underground (claim 9) and can be made of an outer, squared shape, with rounded edges, to make the conduit less sensitive when transported to the site and when being laid or drawn underground during installation. The squared shape minimises voids between conduits, which maximises the amount of heat insulating material. In order that unwanted heat exchange with the surrounding soil be minimised, in each of the 8 conduits in contact with surrounding soil, the channel providing part of the forward, F, line (carrying relatively high-temperature fluid) is arranged farther from the envelope of the assembly than is the channel pro- viding part of the return, R, line. All 9 conduits are of the same outer shape and dimensions; 8 conduits are in addition of identical cross-section, while the centre conduit, CONBIG has channels of bigger diameter, this conduit carrying a bigger flow rate to BUILDBIG. In spite of the slightly differing central conduit, system DISTR displays much more standardisation of conduit sizes than what would be found in most corresponding distribution systems of a conventional design. Each conduit has the same cross-section all the way from the branching station, STA, to the building, BUILD, in question. The conduits are flexible, which is used for creating curved portions CONb of conduits, where they leave the other conduits, but can also be used for making bendings of conduit portions CONa and

CONc, e.g. to circumvent obstacles in the landscape, as is commonly done with flexible conduits according to prior art. DISTR1 can be a conventional thermal energy dis- tribution system or it could be a further distribution system according to the invention, where conduits lead up to, either branching stations, STA*, or to a combination of one or more branching stations and one or more bigger buildings. In the conduits shown in fig. 2 one or more signal CABLE (s) has / have been integrated into each of the conduits. This is a possible facility, not a provision for, the invention. The idea of equipping a DH conduit with a signal cable, by which communication with internal systems of the building becomes possible, is not per se new. As such the idea of 'splitting' costs for digging and laying of conduits and cables lies close at hand. However, with the configuration of the invention, avoiding branchings of conduits from the branching station to the buildings, it becomes particularly easy to employ this idea. Also, as will be explained below, it provides possibilities for extremely reliable and detailed metering, along with possibilities for extended system surveillance. Fig. 3 shows an example of a connection scheme, from the branching station, STA, of fig. 1 to and including the connected buildings with their internal distribution systems and a local temperature recording unit providing information to the building owner and collecting signals for centralised handling in the branching station. In accordance with the conduits shown in fig. 2, fig. 3 shows a (closed-loop), branched 2-line system, comprising a forward line, F, and a return line, R, respectively, providing space heating and hot service water, HW, being incorporated into the invention. For simplicity, only the internal distribution systems of one building, BUILD, has been drawn, and a single ra- diator, RAD, and a single hot water faucet, FAC, have been shown, to represent the normally greater number of such elements inside each building. Hot water, HW is produced from cold water, CW, in a heat exchanger HE1 inside the building, the HW distribution temperature being determined by a thermostatic valve control, THW. District heating water from the forward line, F, is led directly into the hydronic distribution system for space heating, the room temperatures being controlled by thermostatic valves, THR, fitted to each radiator, RAD. As a general comment to fig. 3 it can be said that quite a lot of transducers, valves, and signal transmission facilities in the figure have been adopted in the figure. This has been done in order to illustrate a multitude of facilities which can be adopted with the invention. As will be understood, various of these facilities could very well be left out in particular applications, depending on various priorities. Thus, a particular system in practice may appear simpler than the one shown in fig. 3. One example of a simplification would be to dispense with CABLE (s) . A central pump, PU, inside the branching station, STA, can maintain circulation in all 9 individual building heating loops, when needed. An expansion tank, EXP, controls return line pressure and allows for thermal expansion and contraction of circulating water volume. Heat insulation (not shown in fig. 3) should be applied to all pipes, heat exchanger, etc. inside the branching station, to minimise unwanted heat transfer with the surroundings and heat transfer between individual system elements operating at differing temperature . Inside the branching station, branched lines, BRAL, are connected to channels, CHA, of the forward, F, and return, R, lines leading flow to and from each building. These branched lines are inside STA shown to be equipped with flow meters, FM1 " and FM1 ' ' and with temperature sensors, TS1' and TS1'', all communicating measured values to a signal handling equipment unit, SIG. These values can be measured on an instantaneous value basis or on the basis of a sampling time values. In addition to meters / sensors, the return branched lines leading flow to the branching element, BRA, are equipped with valves, VA1, ... whose position can be controlled from SIG. The first embodiment of the invention, by example of fig. 3, shows a branching station capable of performing several metering and check procedures, as will be explained. Again, If one or more of these facilities is / are not wanted, one or more of the elements just mentioned can be left out to simplify the station. The ambient temperature, TA, is being recorded by SIG; thereby SIG can adjust the supply temperature TS ' , such that this temperature is raised above a minimum level when the ambient temperature TA falls below a certain level, say 0°C. As an overall check that substantial amounts are not lost by any leakage of the closed-loop circuit, a LEVEL indication from the expansion tank, EXP, is transferred to the signal handling unit SIG. Sometimes district heating supplied to a customer is accounted on the simple basis of the amount of flow sent to and returned from the building in question, irrespective of temperature levels. In that case, either flow meter FM1 ' or FM1 can be used for registration, for instance of the total amount of flow circulated in each quarter of the yearly season. More often, though, accounting is made on the basis of the amount of energy supplied. In that case, temperature sensors TS1' and TS1"' in combination with either flow meter FMl ' or

FM1 ' can be used. The signal handling equipment in a known way can be equipped with calculation procedures for compensating for temperature dependence of water density and specific heat. According to the invention, and in contrast to conventional district heating / cooling systems, some or all metering of services to individual buildings is concentrated to branching stations, instead of being provided for inside or close to each building. Concentrating metering equipment to the branching station has a number of advantages: A single signal handling unit, SIG, replaces individual signal handling units inside each building. This reduces first costs and allows for more advanced and reliable equipment to be chosen for this unit. Also, one might dispense with individual thermal sensors, TS1' ... TS9', since all should register essentially the same temperature, and replace them by a single, central sensor, for instance TS' adjacent to the central heat exchanger, HE, in fig. 3. Access to meters and sensors for reading, checking, and replacement becomes easier, in particular in the case of single-family houses, whose inhabitants may not be at home in daytime hours to give energy company personnel access to their house. First costs for meters and sensors can be lowered for several reasons: Fabrication costs of meters can be lowered, since measurement equipment can be built into a common mechanical unit. This will also significantly simplify exchange of meters and sensors for calibration in a laboratory rig, since all meters and sensor can then be taken out and replaced by a whole new assembly of meters and sensors, instead of such replacement work being done individually in each building. Also, due to better possibilities of checking meters, as will explained below, it may be permissible to opt for cheaper types of meters and sensors for the individual metering. Effects of individual installation geometries on measurement accuracy can be eliminated / reduced significantly: Flow meters and temperature sensors are known to be more or less susceptible to individual geometries of installation and surrounding piping. For instance, a bend may cause a distorted velocity profile within the pipe leading up to a flow meter inside a building, which will cause a measurement error. Accordingly, flow meter installation is usually made subject to requirements for certain minimum up- and downstream lengths of straight pipes (usually specified in terms of number of pipe diameters) . Temperature sensors are required to be installed in ways that will reduce measurement errors due to heat losses and / or thermal stratification within the fluid. But space is not always available for long, straight pipes, and even when in a certain buildings conditions are favourable for correct installation of meters and sensors, less careful in-situ work may result in unnecessary measurement errors due to installation effects. When instead, as shown in fig. 3, individual meters and sensors are installed in a branching station, the design can be made, in a standardised way, applying to many stations to reduce installation effects, and remaining installation effects can to a certain extent be eliminated by calibrating meters and sensors when situated exactly as in the branching station; this will be explained more in detail below, in relation to the example of fig. 11. On-line and other check procedures can be adopted in various ways to check the accuracy of meters and sensors installed in the branching station: For instance, if a flow meter is installed both in the forward line, FM1 ' , and in the return line, FM1 ' ' , as exemplified in fig. 3, and given there is no leakage in the flow route from FM1 ' to FM1 ' ' , they should read the same mass flowrate. Another kind of check possibility is provided if equipment is installed for measurement of overall flowrate (FM' and / or FM' ' ) and / or overall thermal energy rate. In the example of fig. 3, supplied thermal energy rate can be registered by combining values read by temperature sensors TS' and TS'' with the flow rate measured by one of the flow meters, FM' and FM''. Such a registered overall flow rate and overall thermal energy can be compared with summed values recorded for individual supplies to buildings, and the size of any deviation will provide an evaluation of the reliability of the registered values. According to fig. 3, overall thermal energy rate is measured on the secondary side of the central heat exchanger, HE, inside branching station, STA. Another possibility is to make measurement on the primary side of the heat exchanger, which should give identical values, provided heat losses from the heat exchanger and pipes can be considered negligible, and no flow leakage occurs across the heat exchanger surface. When overall flow rate and central temperatures are measured in addition to individual values for each building, there is the possibility to choose rather sophisticated and carefully calibrated equipment for measurement of overall values, and cheaper, less accurate equipment (but not necessarily equipment more susceptible to failure) for individual measurements. This can bring down the total cost for measurements. When such a strategy is used, there is the possibility to divide a deviation registered in the checking procedure among individual readings, for instance on a pro rata basis, as a correction towards values that are probably closer to true values . Such a procedure can be combined with an alarm criterion that draws attention to a deviation value exceeding a pre-set limit of acceptance, to prompt an individual action, such as taking the assembly of measurement equipment out for laboratory checking. Naturally, such an alarm criterion will not provide a fail-safe guard against the risk that two or more gross errors happen to more or less cancel out each other. Thus, laboratory calibration of equipment on a regular basis cannot be dispensed with completely, but since check procedures like the ones described will reduce the risk of component failure, they could be taken advantage of to accept longer intervals between laboratory calibrations; this will help bring down costs for metering. According to fig. 3 a local status collection and display unit, STAT, in each building collects a number of temperature recordings, for visual display to the building owner, and for further transmission to the central signal handling unit, SIG, via signal CABLE (s). When this is done it becomes natural to base calculations of delivered amounts of heat on the basis of, not centrally measured supply and return temperatures, TS1' and TS1'', respectively, but on the basis of the locally measured values, TS1''' and TS1¹¹¹'; thereby heat losses between the branching station, STA, and the individual building is accounted for. Even if conduits do not comprise any CABLE (s), transmission of locally recorded fluid temperatures may be possible, e.g via the telephone network. If there is/are no signal transmission facility, one would use the centrally recorded, individual building temperature signals, TS1' and TS1''. This raises the issue of taking calculated or estimated heat loss values into account in some appropriate way. In an extended system (with CABLE ( s ) ) like the one shown in fig. 3, even though temperature reading in each building there are significant advantages associated with moving individual flow metering away from the building to centralised measurement inside the branching station, as will become apparent by the example of fig. 11 here below. Sensing supply and return temperatures both at STA and at BUILD provides a possibility of keeping check on individual heat losses in conduits CON, which can be used for fast detection of various defects that might arise. Recording and displaying all the temperatures as shown in fig. 3 also provides possibilities for various types of checks that may be useful to the customer. For instance, a satisfactorily high primary water temperature, TSl''', but a too low hot water temperature, THW, may indicate excessive heat exchanger fouling. Observing this will help prevent hot water temperatures that may represent lowered comfort level or even a risk of multiplication of the dangerous Legionella bacterium. In fig. 3 inside the branching station is shown a valve, VA1, fitted into the return line from the first building. This valve can be used for several purposes: As has been noted, in the first embodiment of the invention illustrated in fig. 3, hydronic space heating systems of buildings are connected directly to the local distribution system, DISTR, i.e. for this service there is no heat exchanger to provide hydraulical separation. Some energy companies hesitate to dispense with such a further heat exchanger for various reasons. One argument is that, in case of a leaking radiator, a heat exchanger separating the hydronic system from the district heating network will maximise to the potential amount of water leakage to the water volume of the hydronic system, not the volume of the district heating system, which will usually be much bigger. As a precaution against the risk of a greater leak in case of direct connection, valve VA1 can be used for leak detection: By closing the valve and measuring the flow rate in the forward line by flow meter FMl ' any leak can be detected, provided the flow meter is capable of measuring relatively small flow rates, and measurement is made on a continuous basis, so that any transients attributable to thermal / pressure driven expansion / contraction of the loop can be assumed to have died out. Instead of a flow meter, temperature recordings can be used to check against leaks, since in a tight loop closed off by a valve, water will cool off, due to heat losses. When both the forward and the return lines are equipped with flow meters, FMl' and FMl'', as in fig. 3, the difference between the mass flow rates registered by the two flow meters constitutes a further, online, possibility of detecting at least a major leak which would result in a difference exceeding any difference that could be attributable to inaccuracy in flow measurements . Another possibility provided by individual valves, VA1... , inserted into loops serving buildings, is to centrally control individual fluid flow rates and thermal energy supply to buildings (claim 27) . In normal operation an energy company will allow local control equipment inside each building to decide the appropriate amount of flow supplied to any building. But on rare occasions, it can be in the general interest of customers if the energy company has the capacity to exercise control on individual supplies. Such occasions may be when, for some reason, there is a general shortage of thermal energy supply, because of an exception- ally low outside air temperature, or a breakdown of generating plant. In a conventional district heating or cooling system, due to higher differential pressure between forward and return lines, such a shortage of supply will hit customers unevenly: Buildings located close to circulation pumps may experience no shortage at all, while other buildings, typically those in the periphery of the distribution system, may suffer a more or less complete stop of supply. By resorting to a centralised control, throttling a little on all flow supplies the energy company can avoid such a situation. A hierarchical system according to fig. 1 would typically be a beneficial arrangement with conduit designs with relatively small outer dimensions for a given energy rate transferred, so that the size of the outer envelope of conduits CON* extending from STA* will not be too large, and extensive use is being made of flexible conduits, saving time and money when installing conduits underground. Although, as shown in fig. 1, the number of con- duits extending from different branching stations, STA, may differ, a standardised, modularised branching station design can be adopted. For instance, differing stations may share a common outer shell design, which will lower costs. Fig. 11 below will show how metering associated with a branching element can be made in a both sophisticated and rationalised way. Fig. 4a shows a combined cross-sectional and longitudinal sectional view of a second embodiment of the invention. Here, the outer shape of the thermally insu- lated (INS) conduits is sexangular, and conduit portions CONa are arranged to be surrounded by thermally insulating material, INSa, inside a common casing, CASa, as opposed to fig. 2 where conduits are placed directly in the ground. This casing may be essentially stiff, if it is appropriate that it follows a straight line, or the casing my be elastically or plastically deformable, when (such as demonstrated in fig. 1) the assembly of first conduit portions, CONa, is supposed to follow a line being wholly or partly curved. The casing is provided with one or more openings where individual conduits depart from the conduit assembly. Such an opening can take the shape of individual holes, or the entire casing may have a cross- sectional shape of a circular arc not extending all 360 degrees round; this type of design makes later arrangement of extra conduits easier. As shown in fig. 4, a conduit portion CONc (and sometimes in addition curved portion CONb) can also be arranged to be surrounded by thermally insulating material, INSc, either arranged directly in the soil or, as shown in fig. 4, inside a casing, CASc. When extra heat insulation, such as INSa and INSc, is added, as shown in fig. 4, the outer dimensions of conduits can be cho- sen to be smaller than when the heat insulating material of the pipes (as in fig. 2) is supposed to provide virtually all heat insulation. Fig. 4b gives an enlarged view a detail of CONc, close to the forward, F, line channel, CHA, illustrat- ing a particularly appealing structure of the conduit: Here, the conduit is made from a single, polymeric material; an integrated polymeric structure comprises a body part with heat insulating CELLs, as well as inner (at CHA, F) and outer surface layers, SURF, which are compact and smooth, making all surfaces of the conduit mechanically robust. The smooth inner surfaces will reduce pressure losses of fluid flow inside the channels. The proposed polymeric structure is suitable for mass- production by use of modern fabrication methods for polymers. In the embodiment of the invention shown in figs . 4a and b, for minimisation of heat losses, channels, CHA, providing parts of the forward line, F, are located closer to the centre of the assembly than those channels providing part of the return line, R, as in the first embodiment whose cross-section is shown in fig. 2. Fig. 4a illustrates a further possibility of reducing heat losses: Designing channels, CHA, of the forward line, F, with a smaller diameter than those of the return line, R. A bigger diameter of the return line channels has the further advantage that, due to a smaller pressure gradient along the return line, by using pressure reducing valves in forward lines at or inside buildings, BUILD, it becomes possible to operate pressurised equipment inside buildings at a relatively low pressure, when directly connected to the district heating system, i.e. when there is no hydraulical separation by heat exchangers. If the flexibility of the conduits is of a purely or predominantly elastic nature (claim 7), and a casing CASa is used, as shown in fig. 4a, it becomes particularly easy to arrange conduits underground by drawing them inside the casing, either in the direction from the branching station, STA, or in the opposite direc- tion. Conduit parts CONb and CONc can also be drawn inside casings, or these conduit parts can be arranged underground in a more conventional way, for instance by laying them in a channel dug out in the ground. Conduits arranged by drawing should preferable have such surface finish and may be additionally prepared in an appropriate way (claim 6) , for instance by being lubricated or supplemented by a smooth folio, so that they can be drawn underground without use of excessive force . Provided sufficient space is left within casing CASa, as shown in fig. 4, extra conduits for later connection of new customer buildings to the distribution system can be added by drawing them inside the casing and by connecting them hydraulically to branching element (s) BRA of a branching station, STA. Branching elements can easily and at very low cost be prepared for this by use of blinded or valve-closed branch-off pipes, so that new conduits can be connected without interrupting thermal energy supply of already connected buildings. It is understood that in this way the invention in a convenient and robust manner facilitates later connections, avoiding the sometimes difficult establishment of branchings in previously established network parts according to prior art, as described when presenting here above the background of the invention. Both the squared and the sexangular shape of conduits according to figs. 2 and 4a, respectively, are somewhat unusual. Instead, as the third embodiment of the invention of figs. 5a and b illustrate, commonly used shapes, such as a circle or an ellipse may be used. In fact, the whole conduit can be pre-fabricated and / or designed according to methods and designs known in prior art. This can make introduction of the invention in practice particularly easy. Fig. 5b is an enlarged view of one of the conduits of fig. 5a. This cross-sectional view displays many features which are known from conventional, pre-fabri- cated, flexible district heating pipes: Inner pipes are made of copper, Cu, which is plastically deformable, and the insulation, INS, is a closed cell, flexible foam made of PEX, i.e. cross-bonded polyethylene, which can be supplied as a relatively heat-resistant polymer. Also the outer shield pipe is made of PEX. It can be seen that each conduit of the embodiment shown in figs. 5a and b comprises in total 4 medium carrying pipes or channels, CHA: A forward, F, and return line, R, both for carrying a space heating and / or cooling fluid, as well as a hot service water forward line, HWF, and a return water line WR of circulated water not being tapped off inside the connected building. The hot water, HW, is assumed to be prepared centrally from cold water, CW arranged to be fed in at the branching station, STA. Lines HWF and WR are shown to be arranged adjacent to each other; there is no reason for reducing heat exchange between these two lines - heat insulating material is better used for preventing heat exhange with, and between, the two other lines, F and R, as well as heat exchange with the surroundings . As an alternative to copper, one might also use PEX for the inner pipe. Normally, water will flow inside the channels. In that case, oxygen may diffuse from the outside of the conduit and through it, to become dissolved in the water. In case the system is such designed that this water flows through corrosion susceptible components, such as steel radiators inside connected buildings, this may not be tolerable. To pre- vent such a problem, it is common to use metallic or polymeric folio membranes, e.g. around inner pipes carrying the water flow in question, here lines F and R, and / or to employ a membrane between the insulation and the outer shield pipe. The latter arrangement will also help prevent any diffusion of gases / vapour in and out of cells of the insulating foam, which would increase heat transfer through the insulation. Arranging as much as 4 channels inside a conduit is per se not novel, but somewhat unusual. In conven- tional thermal energy distribution systems, with many T-branchings of conduits, there is a tendency that such branchings become more the difficult to design for and arrange by work in situ the more channels are comprised by each conduit. Also, if preparation must be made for T-branchings, this to some extent restricts exactly where it is appropriate to arrange channels in the cross-sectional geometry. The invention completely avoids any such difficulties or restrictions on conduit design . Very interestingly, the invention opens up for the use of un-common or quite new types of conduits which in some cases might have been considered previously, but may have been discarded, since branching with T-s would have been too complicated. Numerous variations of new types of conduits can be thought of. One example has been shown already, i.e. the polymeric structure illustrated by fig. 4b. Typically, in the design of conventional conduits, at least three different materials and or fabrication methods are used: For instance, there may be one or more fluid-carrying pipes of steel, copper or a heat- resistant polymer, such as PEX. Second, there will be an outer shield pipe of a polymeric material, such as high-density PEH. In-between these pipes a heat-insu- lating closed-cellular foam of a polymeric material, usually PUR or PEX, will be arranged. In fig. 5a the casing has been shown to be thin - it could be a metal pipe with thin inner and outer polymeric surface layers to make it corrosion resis- tant. Super-insulators constitute a class of per se known isolation arrangements of various types, whereby an enhanced insulation effect has been achieved by using vacuum inside the insulator. Some super-insulators which have been adopted for various applications include the following: - A multitude of radiation-reflecting, thin metal (e.g. aluminium) foils, kept apart to avoid heat conduction from layer to layer, - Powders made up of granules (e.g. silica-gel) of a suitable (complicated) micro-shape, such that the contact surface between adjacent granules becomes very small, - Fibrous structures, where the fibres can be predominantly oriented in parallel planes, so that the aggregate heat conduction in the direction perpendicular to the direction of the planes becomes smaller . Powder / fibrous structures are sometimes mixed, and platelets of radiation reflecting metal can be added to reduce heat transmission by radiation. Technological fields in which super-insulators have gained general acceptance include : Cryo-technology (e.g. in pipelines for transport of liquefied gases, such as nitrogen) , cooling technologies (including household refrigerators), and spacecrafts. Closed-cell foams, such as polyurethane foam commonly used in DHC conduits, exhibit heat conductivity in the order of 0.030 W/mK. Super-insulators generally have conductivities below 0.010 W/mK. The most sophisticated (and expensive) super-insulators can attain a conductivity even below 0.00010 W/mK. A basic problem in most applications is that the super-insulator and its surrounding design elements generally must be capable of transmitting force in a mechanical design. Separate design members, themselves not being super-insulators, will transmit heat in addition to force, which calls for ingenuity in devising the whole structure of high heat-insulating capability. In a number of applications, where low but not extremely low, heat conductivity is called for, one can select a type of super-insulator which by itself is capable of transmitting force, i.e. mainly a powderous material. Incorporation of a super-insulator into a DHC conduit appears attractive in systems according to the network configuration of the present invention, because it will relieve an inherent difficulty of the inven- tion, viz. that conduits with conventional heat insulation and of a small inner fluid-carrying diameter must have significantly larger outer dimensions, for heat losses to become acceptable; this means that the outer envelope of the assembly of conduits extending adjacent to each other from the branching station, tends to be rather big. Super-insulators offer a possibility of keeping the ratio between inner and outer dimensions moderate, with acceptable heat losses, even in the absence of supplementary insulation mem- bers: A thin casing CASa, as shown in fig. 5a can used, and conduit parts CONb and CONc, outside casing CASa, could be disposed of altogether. Of course, super-insulators also provide the possibility of lowering heat losses. A few attempts have been made to use super-insulators for DHC conduits, so far without any significant success in practice. The designs adopted have mainly been for stiff conduits, whereas in the concept adopted in this invention flexible conduits are called for. In other branches of technology, for instance in WO 01/14783 and in US2001/0035224 , flexible pipes mainly intended for carrying a low-temperature fluids for superconduction of electricity, are described. In principle these designs, directly or modified, could be used as incorporated into the present invention. They incorporate corrugated inner and outer steel pipes for attaining flexibility. Modified variants of these designs could for instance rely on smooth, flexible pipes instead of corrugated pipes. Fig. 6 shows a cross-sectional view of a fourth embodiment of the invention in the form of a novel type of flexible conduit, incorporating super-insulating material for part of the heat insulation, a type which has been devised directly aiming at DHC applications. An inner PIPE, e.g made of PEX, surrounds a channel CHA, which could be both a forward or a return flow channel. The outermost SHELL and the insulator, INS, can be made of dense and foamy PEX, respectively. INS has an inner circulator surface, which in combination with PIPE gives an annular space. In this space, 4 flexible supports, RUB, made of rubber, are interposed, as well as 4 bags containing super-insulator material, SUP, held under vacuum. The bags can be made according to known methods in prior art, of laminated foils, to provide good barriers to any diffusion of gas or vapour from the outside, which would destroy the vacuum, as well as good mechanical strength, which in combination with powder inside the bags provide a substantially constant shape of the bags in operation. The powder can be supplemented by granules of a so-called better material which captures any molecules that might transverse diffusion barriers, thereby helping keep up the vacuum condition of the superinsulator . The rubber supports are compressed so that they exert radial inward forces on PIPE. The residual segments of the annular spacing are not completely filled out by the 4 bags containing SUP, which allows for some deformation when bending the conduit, without any significant outer forces being exerted on the bags. Although the 4 RUBs constitute some resistance to heat transfer, they are less effective than are the 4 SUPs; therefore, RUBs have been positioned diagonally i relation to the outer square of the conduit, so that the thickness of INS has maxima where INS is in contact with RUB. Fig. 7 shows a cross-sectional view of a conduit according to a fifth embodiment of the invention. Here, a super-insulator, SUP fills out the space between two metal pipes, PIPE2 and PIPE3. The outer PIPE1 could also be made of metal, or of a polymer. The insulator, INS, inside PIPEl could be made of some polymer which combines flexibility with sufficient mechanical stability, so that the 3 inward protrusions of INS to establish contact with PIPE2 provide sufficient support for PIPE2 where the conduit is bent, i.e. its axis perpendicular to the cross-section showed follows a curved line. A third requirement for INS that it provides thermal insulation for fluid contained in the 3 partly annular sectors; as indicated by designation CHA, R, it is envisioned that these 3 sectors all convey return channel part flows. In the center, inside PIPE 3, we have the forward flow, CHA, F. That is, return flow essentially circumvents return flow. If the fluid flowing in CHA, R is of low temperature, the heat loss to the surroundings by this arrangement can be kept low even with moderate heat insulating capacity of INS. The super-insulator, SUP should be of sufficient compression strength to hold the inner pipe in such a position, that the thickness of INS nowhere becomes too small in cross-sections of a bend of the conduit. At the same time SUP should behave flexibly in bending of the conduit. This poses some demands on the character of SUP which can be fulfilled by a suitable microstruc- ture of SUP, permitting relatively smooth repositioning of the individual grains of fibre of SUP when exerted to mechanical forces from the outside. Figs. 8a and b show a sixth embodiment of the invention where vacuum inside super-insulators, SUP,F and SUP, R, surround a forward flow channel CHA, F, and a return flow channel, CHA, R, respectively. A vacuum suction channel, CHA, EVA is arranged adjacent to both flow channels, communicating with the super-insulating insulators by perforations, PERF. All three channels are embedded in an outer, flexible insulator, INS. The perforations should be sufficiently small to prevent any grains from the super-insulator to be sucked into the evacuation channel; alternatively, either the outer envelopes of the super-insulators or the perforations as such may be provided with a textile web with minute perforations, smaller than the size of any grains, but large enough that no excessive pressure drop occurs over the textile. For simplification, in fig. 8b, for the sake of simplification of the figure, only connection to a single building, BUILD, has been shown. An evacuation pump, EVAP, maintains vacuum by connection to the evacuation channel, EVS . Each conduit, CON, has its evacuation channel closed off at or inside the building by a SEAL. From fig. 8b it can be seen that it has been assumed that the building, BUILD, has been fitted with a storage type hot-water heater. In comparison with instantaneous water heaters this reduces demands on siz- ing of PIPE, F and PIPE, R. These pipes can be designed to combine strength with inner smoothness, to be maintained even after years of service, i.e. resisting corrosion and erosion, by making them of a metal, e.g. copper, and applying thin polymeric layer to the inner surface. This will permit rather high flow velocities, say up to around 2 m/sec, and pressure up to, say 10 bars. Thereby small flow section diameters can be attained. Thus, fig. 8a could be taken to show a cross- section in full size. If the conduit is arranged verti- cally (as it is shown in the figure) , it will only be necessary to dig a very narrow ditch in the ground, to embed the conduit. In particular when pipes carrying relatively small amount of time-average heat flow are concerned, as is the case in particular with service pipes leading up to single-family dwellings, there is a geometry-related benefit from lowering the heat conductivity of insulation materials, which goes beyond the direct effect of lowering the heat conductivity as such: This extra effect, which is well-known in heat-transfer theory, for the sake of simplicity can be discussed with reference to a conduit of circular-symmetrical configuration, comprising a single fluid flow channel: Supposing that in a calculation the inner diameter is kept constant, and the outer diameter of the insulation is being gradually increased, the marginal benefit in terms of added heat insulation per unit of diameter increase gradually diminishes. In the case of an insulated pipe, which is not buried, but is surrounded by air, heat is given off from the surface by convection. In that case there will even be a certain outer diameter which minimises the heat loss, that is, further outer diameter increase will even increase heat losses. Another example of the geometry effect: If, for a given outer diameter the inner diameter is being diminished, the benefit from this depends upon the heat conductivity. Thus, when conventional heat insulating materials demand a relatively big ratio of outer-to-inner pipe diameter, as is the case with service pipes of single- family dwellings, there is a strong incentive to look for more efficient heat insulating materials. This consideration underlines the interest of the various types of conduits employing super-insulators. When integrated into the invention, such conduits become even more attractive, since they help keep the envelope size of co-extending conduits, and thus casings, CASa, of moderate dimensions. Below a set of criteria quantifying new, more ef- ficiently insulated conduits is given, taking as a reference state-of-the-art conduits employing insulating foam with a heat conductivity in the order of 0.03 W /mK: Criterion A: New conduits comprising heat insu- lating material (s) with heat conductivity being less than 0.03, less than 0.015, less than 0.007, less than 0.003, or even less than 0.001 W / mK . Criterion B: New conduits for which an average heat conductivity is defined as the conductivity of a theoretical conduit of uniform conductivity, completely filling out the space between inner and outer envelopes of the conduit in question, that is including parts of the conduits which are not made of material with exceptionally low heat conductivity, this average heat con- ductivity being equal to 0.03, less than 0.03, less than 0.015, less than 0.007, less than 0.003, pr even less than 0.001 W /mK. Criterion C: New conduits having the same, or only moderately higher, or lower heat loss per unit length of conduit than that of a comparable state-of- the art conduit operation at the same fluid temperature (s), transporting the same amount of heat rate, the new conduits having outer envelope size being less than 0.7, 0.5, 0.3, or even 0.1 times the size of the outer envelope of the state-of-the-art conduit. All the embodiments shown in figs. 6, 7, and 8a offer a possibility of designing flexible conduits with outer dimensions that are no more than roughly twice the size of the fluid medium-carrying, inner pipes. Such pipes can be supplied to the building site for arranging the thermal supply system on rolls carried on a truck. As has been pointed out already, in practice the diameter of such rolls cannot easily be allowed to exceed a dimension of around 2.5 meters. By reducing the outer dimensions it becomes possible to supply much longer lengths of conduits. Since the invention, dispensing with branchings of conduits, calls for long conduit lengths, super-insulated, flexible conduits open up a possibility of dispensing with casings for conduits in a system without branching elements, i.e. a system that can be arranged with a minimum of on-site work, including making excavations, which can be performed at great speed. Conversely, the system structure according to the invention, dispensing with branching elements, paves a way for implementing super-insulators for thermal energy supply networks, since branchings are more complicated to arrange on super-insulated conduits than on conduits with a conventional insulation. In spite of the many advantages of using super- insulation in conduits in systems according to the invention large-scale practical implementation of such systems cannot be made before extensive testing and possible modification of designs. Thus, there is a need for embodiments of the invention that rely on more conventional conduits that have a relatively small outer diameter. This can be realised by using un- insulted pipes inside casings in various configurations, as will be shown here below in the form of 10^th, 12^th and 13^th embodiments of the invention. Fig. 9 shows a schematic of a seventh embodiment of the invention, in which a less common, but per se previously known, 2-line system (displayed simplified with only one building etc., as in fig. 3), where cold (drinking) town's water, CW is arranged to be fed into the distribution system centrally, i.e. into the branching station, STA, to be heated for supply of hot service water, HW, to the buildings: The same fluid is also used for space heating services, giving off heat, via a heat exchanger, HERAD to a hydronic heating system with internal radiators, RAD, since oxygen is dissolved in the water. Big district heating systems with centrally produced hot service water have been built in a number of Russian cities. Smaller systems, according to fig. 9, and using polymer conduits, in Sweden are known as 'Grudis' systems, which have been built in some few cases, but have not yet been adopted in any great number. The present invention could provide a new start for such systems, which offer appealing possi- bilities of first cost savings. An advantage with a scheme essentially relying on distribution of hot service water, as in fig. 9, e.g. when compared with the closed loop system of fig. 3, is that in fig. 9 conduits may be designed without bar- riers to oxygen diffusion, such a metal folio. Thus, the scheme of fig. 9 lends itself to an all-polymeric conduit design. A concern may be that polymeric channels for hot water could promote microbial growth (e.g. the Legionella bacterium) , especially in the return line, where temperatures will typically be lower than 50 °C; one way of handling this problem could be to add anti-microbial copper to the polymer surface. In fig. 9, an Electrically driven heat pump, HP, drawing HEAT from an external environment, such as for instance outside air or ground water, is shown as an example of a centralised source of thermal energy comprised within the branching station, STA. Fig. 10 shows (like-wise simplified, to show only one building etc.) an eighth embodiment of the inven- tion, incorporating a 4-line distribution system (claim 21), composed of a 2-line closed loop forward, F, and return, R, line system for building heating or cooling, depending on the season (claim 22), and a 2-line loop for loading a tank, Ta, inside each of the buildings, with hot water to be supplied via a forward line, HWF, when cold water is taken out from the bottom of the tank and fed into return line, RW, for centralised heating of cold water, CW, into hot water, HW, in the branching station, STA. Heat and / or cold is supplied to the branching station, STA, from a 4-line district heating (HEAT) and district cooling (COLD) system, DISTR*. Whenever hot service water is needed, this is served via heat exchanger HE1, which typically will be in operation for some intervals during the day, or maybe more or less continuously, most or all the days of the year. In the cooler part of the season, forward line, F, serves distribution of district heating water, which is heated in heat exchanger HE2, and in a warmer part of the season the same line serves distribution of district cooling water which is cooled in heat exchanger HE3. That is, the 2-line system composed of F and R operates as a switch-over system. Settings of 3-way valves, 3VA', 3VA' ' , and 3VA' ' ' in the forward and return lines in- side the branching station and in the building, respectively, will determine which of the two alternative modes is in operation. The 3-way valves in the branching station are shown to be shifted automatically by signals from the signal handling unit, SIG, according to recorded level of the ambient temperature, TA. In- side the buildings, radiators, RAD, are in operation in the cooler part of the season, and fan coils, FC, are in operation in the warmer part of the season, as determined by controller C'. A simpler variation of the system shown in fig. 10 does not include cooling services, whereby a number of components, including 3-way valves, become superfluous. Today in most countries with a large district heating sector, air conditioning of single-family houses is not very common. But it can be envisaged that in future a possibility to cool private homes, including single- family houses, will be seen as attractive. Building simplified systems according to fig. 10, provides the possibility of changing individual F / R loops to in- elude the district cooling facility when cooling of the corresponding building becomes a reality. In the simplified scheme of 4 lines, circulation of the F / R loop can be shut off in the summer season. When the HWF and RW channels are designed with small diameters, the heat loss from such- a system will be lower in the summer season, compared to a similar 2-line system, where circulation is maintained all year round in bigger F / R channels. Inside the tank, Ta, hotter and less dense water is stored on top of colder water by thermal stratification in a per se known manner. Cold water, CW, is supplied to the tank from a drinking water system to the tank at its bottom, when hot service water, HW, is drawn off by opening of one or more hot water faucets, FAC, inside the building. As soon as the tank looses some of its stored amount of hot water, a temperature sensor at the bottom of the tank registers this, whereby re-loading of the tank by opening of valve, VA, is activated. A tank storage system of the kind shown in fig. 10 is known from district heating and from other methods of serving buildings with heat energy, e.g. local heat pumps serving individual buildings, except for the fact that in fig. 10 hot service water is heated centrally, i.e. not inside the building but at a distance from the building, and in a heat exchanger, which will typically be common to a group of buildings where tanks are installed. In comparison with simpler connection schemes, such as for instance the one shown in fig. 9, the sixth embodiment of the invention, incorporating a 4-line system has some attractive features, as will be explained here below. These attractions would in principle apply also to any thermal energy distribution system built according to the scheme of fig. 10, i.e. including a conventional network necessitating use of many T-branchings of channels and conduits. In a conventional district heating system according to fig. 10, one might arrange all 4 channels inside the same conduit (as in fig. 5b), or one could for instance use 2 conduits running parallel, with 2 channels inside each conduits. In both these arrangements, it would certainly be possible to accommodate T- branchings, but quite an amount of on-site work would be required, and the 3-dimensional geometry of these branchings will place various demands on exactly where to place the 4 channels inside the cross-sections of conduits. When instead a 4-line scheme, as the one shown in fig. 10, is arranged by using an energy dis- tribution system according to invention, where no branchings are made directly on conduits, no such restrictions exist. In fact, it may be claimed that the more lines and channels are used in a particular district heating / cooling scheme, the more the advantage of dispensing with T-branchings will come into its right. For example, in distribution systems according to the invention, it will represent no great complication to accommodate as much as 6 channels within a single con- duit. In cases where heating of some buildings is required alongside with cooling of other buildings, it can be natural to design distribution systems with 6 channels: 2 channels for central production and distribution of hot service water, 2 more channels for dis- trict heating, and finally 2 more channels for district cooling. A switch-over system (claim 22), where 2 channels in the winter season are used for district heating and in the summer season for cooling of course in one sense is a simpler solution, but this solution requires that the group of buildings served perform rather equally in terms of when heating and cooling is needed, a requirement which may not always be fulfilled. As one example, in a temperature climate, like the Scandinavian, older single family houses with little thermal insulation may call for rather little cooling in summer, where it may attractive for the inhabitants to spend some time of the day in the garden, if they are at home at all in daytime hours. By contrast, modern office buildings, with substantial thermal insulation, and other features to keep down heating demand on cold days, with big windows and with a lot of heat generating PC-s and other equipment, as well as people generating heat, often call for of air conditioning i.e. cooling, not heating, in a major part of the year, i.e. not only in the summer. Now, returning to explaining the advantages with a connection scheme as shown in fig. 10: First: If hot service water has to be produced instantaneously in heat exchangers, not in prolonged time intervals, due to evening-out effect of a tank, the re- suiting peak loads on the distribution system are felt particularly hard in the case of single family houses, because of a big relative load variation. - In large buildings there is an automatic evening-out effect due to diversification in time of tappings in individual apartments. Thus, with single-family houses, tanks can be used for selecting smaller diameters of channels in service pipes leading up to buildings. In the system shown in fig. 10, very small diameter pipes can be adopted for the lines HWF and RW, which can contribute to decrease heat losses and / or to reduce conduit sizes . Second: When a group of single-family houses is served by the same branching station, STA, they will in fact often perform rather equally in terms of thermal load variation with outside air temperature, so that conditions may be rather favourable for adopting a switch-over solution. Often groups of single-family houses have been built within the same year, or within the same decade. In the first case, they may even have been made by the same firm. Still, even if groups of buildings behave almost identically, conditions may not be ideal for using the switch-over scheme: The temperature differential between the forward and the return line will tend to be substantially bigger in the district heating mode than in the cooling mode, which could present a mis-matching problem of channels for cooling needing to be bigger than channels for heating. How well channel sizes needed for heating and cooling match has to be considered specifically in a given case, to decide if a switch-over solution is appropriate . Third: According to fig. 10, hot service water, HW, is prepared from cold water, CW, which is (via WR) led to the branching station, STA (claim 15) , and heat- ing is catered for by distribution lines separate from lines serving hot water preparation. These conditions are favourable to the use of Friction-Reducing Additives (FRAs), such as tensides, which in the last few years have been developed, tested and applied (not least in Japan) with success. Such additives can be tailored (by modifying their chemical composition) to perform optimally at in various temperature intervals . Also, newer types of FRAs have been shown to be bio-de- gradable in soil. Still, so far, in some countries, like the Scandinavian countries, FRAs have mainly been considered for use in big district heating transmission systems, separated from local heat distribution networks by heat exchangers. For most existing district heating systems the use of FRAs is considered problematic, since one always has to consider the risk of district heating water by accident leaking into drinking water systems. Although the tensides in question are not considered particularly poisonous (as for in- stance the de-oxidising substance hydrazine, sometimes being added to district heating water) , the mere possibility of a 'foreign' chemical substance leaking into a drinking water system by authorities in most countries is not accepted. It is a fact that when hot service wa- ter is prepared separately in each building, experience shows that district heating water occasionally does leak into drinking water systems. This could happen in an event of a leaking heat transfer coil of a storage heater, combined with a higher pressure on the district heating side than on the drinking water side, combined with the mishap of mal-functioning non-return valve in the drinking water supply line. But with a system according to fig. 10, it can be argued that the risk of leakage into the (CW) drinking water network can be reduced so much that use of ten- sides could be regarded as acceptable. For instance, heat exchanger HE2 can be made with double walls, to make a mixing media virtually impossible. Using FRAs in local distribution systems, like the one shown in fig. 10 can be utilised for selecting small diameters of channels for space heating / cooling (F and R) . Fourth: Provided the tank is of sufficient size, it can to a great extent even out the load on the distribution systems, caused by tappings of hot water, which will vary substantially with time, especially in the case of single-family houses. Thus, for example, if a conduit cross-section of the type shown in figs. 5a and b is modified for application in a 4-line system as described, one can select small diameters of channels for hot water provision (HWF and WR) . Fifth: When service water is heated centrally to be fed into a tank, it becomes attractive to fit the signal handling equipment, SIG, with intelligent procedures to optimise loading of individual tanks, That is, in addition to the local (i.e. of the building) control, represented by thermostatic valve, VA, branching line channels of the branching station, STA, may be fitted with valves, VA1 ' , to be controlled by SIG, for instance to adjust the loading flowrate (measured by a flow meter, FMl'') according to the individual consumption pattern of the building and / or to spread out in time starts of loading of tanks belonging to individual buildings. Alternatively, the loading pump, PU1, may be controllable from SIG. This will cause evening out / diversification of the heat load posed by distribution system, DISTR, on the district heating, HEAT, part of distribution system, DISTR*, from which heat is supplied to the branching station, STA. This kind of centralised control should of course not be driven to such extremes that people living in the buildings will experience shortage of hot water. SIG may even be programmed to adapt loading of individual tanks to individual consumption patterns, so that a faster loading takes place in a building with an above average hot wa- ter consumption. Sixth: When space heating (+ possible cooling) and hot service water provision is served by 4 lines instead of 2 lines, separate metering of 2 energy rates (or mass flow rates) are called for, which provides an opportunity for the energy company to supply the customers with information about how the total energy consumption is divided between space heating and hot service water; such added information can be of value to the customer, e.g. if he wants to asses effects of measures taken to reduce energy consumption. The downside is that more metering is associated with this extra information service. Here, the invention comes in handy, because of the rationalisation of metering which becomes possible, due to concentration to the branching station as previously explained. Fig. 10a shows another hydraulical configuration with essentially 4 lines that can be adopted, either directly in the form shown, or in combination with features shown in fig. 10. Fig. 10a in particular exemplifies how conduits of relatively small diameter can be adopted even when there is no hot water tank, and how pressure drops in conduits can be handled in a relatively simple way. For the sake of simplicity only two building systems are shown, viz. that of a first building close to the branching station, STA, and an nth building at relatively great distance from STA. Forward line FI and return line Rl are connected to the first building radiator system. Forward line Fn and return line Rn are connected to the nth building. Within STA or close to STA, Rn spilts up into two lines, Rn and Rn' . Thereby pressure drop along the relatively long return line is reduced, helping keep down the pressure level of the radiator system of the far-away building. At the first building the pressure differential between FI and Rl lines is relatively big; differential pressure controller DIFP reduces the pressure drop across the radiators in the connected building heating system. Lines HWF1 and RW1 provide hot service water for the first building, and lines HWFn and RWn provide hot service water for the nth building. When there are no or only small hot water tappings, a circulation is maintained by hot water circulation pump PUHW, i.e. in lines RW1 and RWn there flows are being returned to STA for heating. At greater tappings the flow direction in lines RW1 and RWn are reversed, to supplement the outgoing hot water flows in lines HWF1 and HWFn. The dual mode operation of the hot water systems is obtained by the connection arrangement with pipes comprising non-return valves VANR1 and VANR2. If existing buildings comprising hydronic systems are being connected to the district heating system, they will commonly comprise existing expansion tanks EXP and EXPn . Depending on their size and the total water volume of the radiator systems, including water in conduits outside the buildings, these expansion tanks may or may not suffice for proving sufficient thermal expansion of water. In the latter case the expansion capacity can be extended by a dynamic pressure control system, by which water can be supplied from the primary water loop, via valve Vapc, and water can be drawn out via pump PUpc. The local expansion tanks will provide buffer capacity, so that the frequency of automatic switching between operation of Vapc and Pupc will not be excessively high. A central circulation pump, PUrad maintains circulation of water in the extended radiator water loops . A central METER monitors the total amount of heat being transferred. Local meters in buildings, METER1 and METERn monitor the amount of heat supplied for building heating and for supply of hot water. These meters can be equipped such that the customer can read the two amounts of thermal commodities separately, which provides a more detailed information to the customer than is normally provided for. The sums of individual heat deliveries can be compared with readings of METER, the difference being a measure of heat losses in the distribution system. Pumps, valves, control equipment, and metering equipment arranged within or adjacent to a branching station may be driven by electric power supply via cables from the outside. Another or supplementary option is to supply electrical power and / or mechanical power to drive the mentioned types of elements from a turbine, in turn driven by a pressure drop or enthalpy drop taking place in one or more conduits leading fluid flow to a branching station. In the case of electrical power such power can be generated by a dynamo driven by the turbine. Such an arrangement may reduce or completely eliminate the need for electrical power to be supplied from the outside. Fig. 11, which is a drawing of a branching element of any embodiment of the invention, illustrates how such a concentration of metering can materialise. The total branching station, STA, is understood to comprise an appropriate number of branching elements, all connected to the same signal handling unit, SIG (a number of arrows pointing towards SIG in fig. 11 indicate signals to and from various elements not shown, including elements the branching element (s) not shown) . The branching element, BRA2, shown in fig. 11, can be almost identical for lines R and WR (cf. fig. 10), while branching elements serving lines F and HWF can be modified versions of BRA2 , for instance to comprise no flow meters, but instead each to comprise a common thermal sensor (upstream of branching) and valves fitted into each branched line. In fact, branching elements used in any of the previously shown embodiments of the invention can be designed as variations of BRA2 shown in fig. 11. The entire branching element, BRA2 , is shown to include thermal insulation, INS, of all parts. 3 return line channels, RCHA1, RCHA2 , and RCHA3 are shown in the figure, and more such lines can be imagined. Each chan- nel includes a classical venturi type piping element,

VI, V2, V3 ..., in which the flow is narrowed down to a smaller pipe diameter from which the diameter in a diffuser part gradually expands back to the original diameter at flow outlet into a BOX, where mixing takes place, and from which a bigger pipe, CHA, leads the aggregate fluid flow further from the box, for heating in one the heat exchangers (not shown in fig. 11) of the branching station, STA. Flow meter FM2 can be of a type commonly used in district heating practice, such as e.g. an ultra-sonic flow meter. A flow straightener, STR2, is arranged upstream of the meter, to even out skewness and / or rotation set up in the flow profile at inlet to pipe CHA. Each venturi element is fitted with two pressure sensors to record the pressure dif- ferential, Δpl ' ' , Δp2 ' ' , Δp3 ' ' , set up in the converging part of the venturi, the size of this pressure differential being a measure of the flowrate, i.e. the venturies fitted with pressure differential sensing are in fact flow meters, FMl'', FM2 ' ' , FM3 ' ' .. The pressure sensing elements can be of the piezoelectric type. As can be seen, each of the branched lines also includes a temperature sensor, TS1'', TS2 ' ' , TS3'', which can be of the resistance type or of the thermo-couple type. As the big flow meter, FM2 , all the small venture type flow meters are supplied with a flow straightener, STR''l, STR''2, STR''3, ... upstream of the meter, to reduce the effects on metering of flow profile skewness, which will be caused by bends and other deviations from a straight pipes upstream of the meters . A number of advantages of concentrating metering equipment in the branching station have been discussed already, such as simplification of procedures for tak- ing out meters for calibration as assembly from the branching station, instead of taking out meters from each building. By discussion of fig. 11, several further, appealing and distinctive features made possible by the invention can be pointed out: (1) : The branching element, including metering / sensing equipment is compact, (2) : By adopting an appropriate calibration procedure, flows and temperatures can measured relatively independent of installation effects. (3) : The type of flow and temperature sensing equipment selected can be relatively cheap. Ad (1) : The compactness is attributable to 2 features: Use of flow straighteners to eliminate long, straight pipes, and to the use of small sensors for pressure and temperature on individual channels. It is true that a flow straightener arranged rather close to a flow meter, as indicated in fig. 11, will influence the calibration curve (reading of electrical signal vs. flow rate) of the meter, but this effect can be eliminated by calibrating the meter to- gether with the flow straightener. In fact, meters and sensors of the branching element should be calibrated as they are arranged in the branching element. Ad (2): Due straighteners STR''l, STR''2, STR''3, ... effects on measurement of skewed flow profiles caused by bends etc. upstream of the branching element are reduced. However, since straighteners do not function perfectly, if bends are arranged close to the branching element, measurement accuracy can be improved by including such bends in the branching element, i.e. calibrating with such bends not being dismantled prior to calibration. Ad (3) : The cheapness to some extent is related to the fact that point measurements are made instead of bulk flow measurement. Thermal sensors applied to a point, as indicated in fig. 11, can make temperature recording sensible to any thermal stratification of flow remaining downstream of the flow straightener. To a great extent, without increasing the dimensions of the branching element, this can be compensated for by designing the resistive sensor as a ring to extend all, or almost 360 degrees round the periphery of the pipe. One of the requirements on flow meters for installation in district heating substations that generally tends to call for expensive equipments is that the meter should measure accurately within a great span of flow rates, e.g. 1:100, a requirement which is particularly demanding in the case of single family houses equipped with instantaneous hot water heaters. In the scheme of fig. 10, due to the presence of tanks, there is no great variation of flow rates . This corresponds well with the selection of venturi flow meters, since they rely on pressure differential coupled to differences in dynamic pressure, i.e. a quadratic type of re- lation to flow rate. With such a relationship, signals become very weak at low flow rates. Other types of compact, low-price flow sensors than piezo-electric pressure differential sensors are available in the market, such as e.g. thermal flow sensors, which may be better suited in branching stations of system configurations where hot water tappings are directly reflected in flow rate variations, to be measured in the branching station. Fig. 12 shows a ninth embodiment of the invention in which forward, F, and return, R, conduits, CON, defined as the inner pipes (not insulation and casings), serving the same building (not shown) run in different cavities, CAVa'and CAVa ' ' inside casing CASa, along their CONa portions and merge into a common casing CASc of their CASc portions. The CONc portions may be bonded to insulation INSc, or there may be a small spacing in- between, so that conduit parts CONc can slide axially inside insulation INSc. Conduits, as they are definied according to this wording, do not themselves comprise any significant insulation. An alternative description would be to interpret the whole assembly: 2 times CONc, INSc, and CASc as a conduit whose central parts split up in the direction backwards to the branching station (not shown in fig. 12) . If CONb portions are short, it may be permissible to leave out insulation around these conduit parts. Fig. 12 shows instead an embodiment where additional insulation, INSb, has been arranged around curved portions CONb, inside a casing, CASb . Fig. 13a, b and 14 show a tenth embodiment of the invention where forward, F, and return, R, lines, serving the same building (not shown) are completely separated. Also, this embodiment shows how a system according to the invention can based on very simple conduits, keeping sizes of casings, CAS' and CAS'', rather small, those casings containing forward, F, and return, R, channel conduit parts CONa' and CONa'', respectively, inside cavities CAVa' and CAVa'', respectively. Fig. 13a is a cross-sectional side view, and fig. 14 is a sectional view from above. Fig. 13b shows an enlargement of one (forward- line) conduit CONa'. It can be seen that the conduit comprises an inner metal (e.g. copper) pipe, MET, and an outer, polymeric (e.g. high-density polyethylene) coating layer. This coating layer breaks any galvanic currents that might otherwise be set up outside the conduit and provides mechanical protection of MET. In addition, POL provides some 'residual' heat transfer resistance, i.e. a thermal resistance that on the one hand is not sufficient for generally providing insulation of conduits, but on the other hand, since polymers generally exhibit lower thermal conductivity than do metals, provides much better resistance to heat losses than do naked metal pipes; this can be of advantage with locally lowered heat insulation along conduits, e.g. at conduit portions CONb (curved parts leaving the assembly of co-extending conduits) . In fact all conduits parts are provided with added heat insulation: INSa' to CONa'. INSa'' to CONa'', INSb' to CONb', INSb'' to CONb'', INSc' to CONc', and

INSc'' to CONc''. Insulations CONb' and CONb'' are provided by re-fill soil, SOILins, having a lower heat conductivity than has other SOIL surrounding the embedded system. It can be seen that the system has been system designed that water can in fact percolate into cavities CAVa' and CAV ' ' as well as into annular clearings between insulations INSc' and INSc'' and conduit parts CONc' and CONc'', respectively. This can be acceptable under the premise that such water does not compromise the thermal integrity of any system parts being exposed to water. Insulations can be made such that they will not become soaked with water. After all, water which stands still does not conduct heat very efficiently. Some designs according to fig. 13 may be restricted in use to locations where they will be well above the water table in the ground. It is true that the only moderately insulated conduits CON provide possibilities for heat flows between individual conduit parts CONa' and CONa'' inside cavities CAVa' and CAVa'', respectively. However, in the forward line case, fluid flows will be of almost the same temperature. Return flows will in general not be of the same temperatures, i.e. heat flows will be set up between individual conduit portions CONa''. In general this will not be any problem, since return flow will be mixed anyhow, once they reach the return flow branching element (BRA2 in fig. 11) . In special cases, one may single out a group of buildings with especially low return temperatures to be utilised for a good thermodynamic performance of a heat pump connected to these special return pipes. In that case, heat transfer between individual return conduit portions CONa' ' may be unwanted and should be taken care of. Figs. 15 - 17a-c show an eleventh embodiment of the invention. Fig. 15 is a longitudinal, sectional view; fig. 16 is a view from above, earth on top of the embodiment having been removed; figs. 17a-c are three cross-sectional views, as indicated in the two preceding figures. In this embodiment, in total 7 conduits, CON, are arranged inside a casing, CASa, departing from the casing by penetrating the top of the casing by elastic de- formation of the casing. As fig. 17c shows, the casing all along its longitudinal extension comprises a separation plane, SEP. In a preferred design of the invention the casing is such fabricated that the two meeting surfaces, SURF, at SEP are pressed together and/or a covering TAPE is fastened to cover the separation, so that it is water-tight from the outset. All surfaces of the casing, comprising: the outer surface in contact with soil, the inner surface circumscribing the cavity, CAVa, and the assembly of conduits, as well as the two surfaces, SURF, in contact with each other at SEP, all the way round are compact, rugged and substantially impermeable to water. The casing together with its interior, insulating part, INSa, is deformable, preferably completely or at least partly elastically deformable, so that when separating, horizontal forces, FO (cf. fig. 17b) are applied to a cross-section at any location along the casing, and TAPE has been removed along an appropriate length of the casing, the casing will tend to reverse to its initial, closed shape or, when a body is inserted, will exert forces upon this body. As can be seen from the figures, this design of the casing has been utilised for permitting conduits to penetrate it, the figures showing by example the first conduit leaving the casing by deformation of the con- duit in two planes along its portion CONb: First the conduit bends mainly upwardly, combined with a slight bending towards the middle of the cross-section, to penetrate the casing through SEP. Then, having passed all the way through the casing, the conduit bends 90 degrees in the horizontal plane, to leave the casing in a direction substantially perpendicular to the casing by conduit portion CONc, which is continued up to a building, BUILD (not shown) . Openings are thus created by deformation of the casing, not by taking away material from the casing by cutting, grinding etc. This speeds up installation, in particular when conduits are to be inserted at a later stage of network development. Openings are created in situ, where needed, either by simply drawing or press- ing the conduit through the casing, or with the assistance of one or more appropriate tools. By appropriate design of casing and conduit cross- sectional shapes and surface finishes, it becomes possible to create a structure by which the casing by it- self closes completely around the penetrating conduit, so that at all cross-sections, including all those not shown in figs. 16a-c, either two surfaces of the casing itself are pressed together, or there are contact zones between a surface, SURF, of the casing and a surface of the conduit. Still, ageing of material, settings in the ground, etc. after shorter or longer time of operation of the system could create openings around a penetrating conduit. In order that water-tightness, if needed, be maintained, one might apply GLUE (cf. fig. 16a and b) around the conduit, wherever TAPE has been locally removed. Another alternative could be to design the system such that water-tightness at conduit penetrations is not a pre-requisite, as discussed here above in re- lation to fig. 14. A classical district heating culvert principle, which has been used in old designs where conduits were placed inside concrete culverts and could be applied here as well, is to arrange the culvert / casing CASa with a slight longitudinal slope towards a location where drainage is provided for. In any case, insulations, INSa and INS of casing and conduits, respectively, preferably are made such that they will not soaked if exposed to water. This can be achieved by using closed-cell foam as thermal insulation. As can be seen from fig. 15, conduit portions CONc can be arranged rather close to the ground surface GS . This, in combination with a conduit profile of small horizontal extension, allows for rather little earth (trench, TR) to be removed if conduits up to buildings are arranged underground by digging, carving etc in the ground. An alternative to the arrangement shown could be to turn the cross-section 90 or 180 degrees, so that conduits leave the casing from a side of the casing or downwards . Figs. 18a-g show a twevlth embodiment of the invention. Fig. 18a is a top view of an arrangement showing essentially un-insulated conduits being arranged mainly within casings that take the form of insulating blocks. BLOCKn-2', BLOCKn-1' , BLOCKn' , and BLOCKn+1' are arranged adjacent to each other, to acommodate portions CONa of conduits. It is understood that there are further blocks, both to the left of BLOCKn-2' and to the right of BLOCKn+1' , so that only parts of portions CONa' are shown in the figure. BLOCKbrl' is the uppermost of blocks providing a first part of a casing comprising conduit portion CONc' leading up to a building, not being shown in the figure. The transitional, curved conduit portion CONb' is contained within an insulation, INS, that be a loose fill-in material, contained within a SHELL; alternatively, also portion CONb' could be contained between blocks of insulating material, with shapes adapted to the curvatures of conduits. Fig. 18b is a side view of assemblies of building blocks, BLOCKn, BLOCKn', BLOCKn", BLOCKn'", and BLOCKn' ' ' ' constituting the assembly shown in their full extensions, and placed on top of each other. To the left and right parts of further assemblies of blocks are shown. At the interfaces, for instance between BLOCKn' and BLOCKn+1', blocks are shown to have convexly and concavely curved shaped ENDs, respectively, where the blocks touch each other. They could also be glued together or, for instance, fastened together by means of not shown metallic clamps. As can be seen, the curved ENDs serve a double purpose: First, they provide some coherence of blocks, preventing relative moments in the vertical direction. Second, they permit some angular movement, such that blocks to some extent can follow geodetical variations of the lines of conduits in the vertical plane, as illustrated by the small angle indicated in the figure. Below the assemblies of blocks, there is a layer, RU, of rubbles, which provides interface with the ground below and allows for drainage. Figure 18c shows a cross-sectional view of the assembly of blocks BLOCKn, BLOCKn' ... BLOCKn"" with conduit portions, as well as a section of the transitional portions CONb of conduits with SHELL and in-fil insulation INS, as well as a longitudinal section of a first part of conduit portions CONc arranged within and between further insulation blocks. Fig. 18d shows a cross-sectional view of these latter components . On top of BLOCKn, INS, and BLOCKbr a SHIELD has been arranged which serves the double purpose of, first: stopping water seeping through the soil on top of the insulation arrangement from entering these elements by vertical motion downwards, and second: shielding them from being damaged if somebody digs in the ground, for instance when a building owner prepares for plantations by using a shovel . BLOCKn' to ^Λ" have a shape that has been adapted for branching off of a conduits . Thus, fig. 18a indicates that there is a mid-section 2 that deviates from the identical outer sections 1 and 3, whose shapes are shown in fig. 18c. Here it can be seen that the three mid-blocks BLOCKn' , BLOCKn", and BLOCKn'" are identical, while the top BLOCKn and the bottom BLOCKn'"' are different. Blockn' has a mid- section, 5, that lies deeper than the identical outer sections 4 and 6. The shapes of the blocks create cavities above mid-sections 5, where conduits are arranged. Fig. 18c shows an example where in total 10 buildings are served by identical conduits, each being served by 4 conduits: HWF = Hot Water Forward, WR = (hot) Water Return, F = Forward for building heating, and R = Return from a building heating system, as indicated in fig. 18b. It can be seen that conduits HWF and WR have been placed on top of each other in the same cavity. As long as WR is only being used as a return hot water line or as a temporary additional hot water supply line (according to fig. 10a), there is no need for separating HWR and WR, and the type of arrangement shown serves the purpose of reducing the height Ha of the assembly of blocks. On top of conduits in mid- section 4 an elastic disc, ELA, has been arranged which helps keep conduits fixed, so that thermal expansion in axial direction can be prevented, provided conduits are designed for a relatively low modulus of elasticity, E. This will be the case if conduits are made of polymeric material. Even so, there may be a need for fixing conduits while arranging them in their respective places. This could be done by using glue, adhesive tape, or, as indicated in fig. 18a, by placing metallic strips, ME that are such shaped that they do not prevent conduits form being laid down from above but at the same time provides some inward pressure on the sides of the conduits. Thereby, they are being strongly positioned sidewise and are being prevented from bouncing elastically back in upward direction after having been laid down. This is due to friction between the sides of conduits and small, vertical walls of ME (not shown but easily imagined) . Similar arrangements are known from the art of designing under-floor heating systems where, as with the present embodiment of the invention, there is a need of fixing polymeric conduits onto horizontal plates. For simplicity, the top surface of mid-sections 2 (fig. 18a) of blocks can be shaped to provide a plane surface, i.e. the top surface of mid-sections 4 (fig. 18c) will continue to the left and right, ^cutting' sections of 4 and 6 away, such the assembly of blocks will show in total three rectangular openings, as indicated in fig. 18b. To the left in fig. 18a it can be seen how a curved assembly of blocks (BLOCKn-1' etc) can be used for letting the assembly of conduit portions CONa deviate from a straight line, as seen from above. Thus, the twelvth embodiment of the invention can be adapted to follow a street which curves. It can be seen that a high degree of standardisation can be adopted in fabricating elements that can be combined to adapt to various local deviations from straightness and regularity. For instance, curved blocks like BLOCKn-1' can be fabricated from circular elements from which segments, covering varying degrees for various blocks, to be cut out. Another possibility is to use more adjacent conduits to increase the effective cross-sectional area of all conduits serving as HFW, WR, F, and R-lines, respectively. A further example, which has been indicated in a schematic way in fig. 10a, is to double the number of R-lines only, in order that the pressure drop in the this line be reduced. This can help reduce the pressure inside radiators of connected buildings . From fig. 18c and d it can be seen that by having F and R lines of conduit portions CONc ( ^Λbranch' line portions) at the same level, the height He of this assembly can be made smaller than the height Ha of the assembly containing conduit portions CONa. Also, width Wc is smaller than width Wa. All conduit portions CONc of lines FHW, HW, F, and R have been arranged in grooves cut out in BLOCKbrl" and BLOCKbrl" ' . Grooves provide a simple and efficient means for immediately positioning conduits solidly. As has been mentioned already, positioning of conduit parts CONa is being assisted by metal strips ME. The reason for not using grooves in BLOCKn' etc. can be that a relatively small width Wa is important for minimising interference with other types of conduits (water, sewage, etc.) already in place when digging for district heating is performed. It can be seen that channels dug out for CONc block assemblies have a depth He' ' that is significantly smaller than depth Ha'' below the pavement. There are two reasons for this: First, as has already been pointed out, height He is smaller than height Ha; sceond, the overhead soil layers are of differing heigths Ha' and He' , respectively. This reflects a common practice already used in district heating practice. On the one hand there is always an installation cost incentive for using as shallow ditches as possible. On the other hand, where mains are installed within a street area, including arrangements under pavements, local regulations, securing the solidity of streets against loading from vehicles etc. are manifested in requirements regarding minimum heights of overhead layers. For conduits on private property much lighter traffic loads can often be assumed, especially where conduits are led below grass-beds or naked soil. In many cases insulation blocks can preferably be made of an open-structured insulation material, such as Expanded Polystyrene, EPS, which is cheap and has a good record of use for heat insulation and mechanical support of lighter building structures, as well as other applications. EPS is amply available from many suppliers who are already tailoring element in this material in many different shapes. Also, it is easy to form EPS elements to various shapes in the field where it can be sawed very easily, or it can be cut by a hot metal strip which is resistance heated by having an electrical current run through the strip. The water permeability of the open structure EPS admittedly leads to a higher thermal conductivity than with closed-cell foams, such PolyURethane, PUR, which is nowadays the insulation material of choice for must prefabricated district heating pipes. Thus, to achieve comparable heat losses, in systems using EPS for insulation the amount of material will have to be greater. However, moderate amounts of water will only increase the heat conductivity by a factor of the order of 2. Furthermore, the rectangular shapes of blocks, together with the fact that there is no need for extra space for fitting or welding means that ditches may be of the same size as with conventional, prefabricated district heating mains. EPS blocks are rather robust; thus, workers can walk on them without causing damage to them which their use in practice convenient . In various circumstances it can be preferable to provide some, or all blocks in another insulating material, such as for instance PUR or flexible, cross-bounded (PEX) foam, or another foam with closed insulating cells. Such materials that are more expensive, can preferably be chosen where insulation blocks are arranged below a high ground water table, which would cause EPS blocks to become fully soaked with water. Another case is when F and R conduits in the summer season are used for circulating cold water for providing building cooling, since on the surface of the then cold F-conduit there will be a tendency for water vapour present in the insulation to condense. On the other hand, if F and R conduits, when used, always run in a heating mode, the assembly of blocks will be characterised by a thermal gradient towards lower temperature at all point of the boundary surface with the surrounding soil, a fact that counteracts excessive water accumulation within EPS blocks . Insulation blocks made of PEX-foam can be used for a few, strategically selected blocks, to allow for increased adaptability to local geometric conditions. As an example, if flexible PEX-foam is selected for curved blocks BLOCKn- 1' etc, any angle of deviation from a straight line, below a certain maximum angle, could be realised with a selection of a limited number of standardized elements of, say 5, 10, 15, 20, 25, and 30 degrees of curvature. In this context it can be noted that curved ENDs automatically provide a similar adaptability in the vertical plane. Thus, a high degree of standardisation of elements can be combined with an adaptability to various demands on the geometry in a way that does not require time consuming accuracy when arranging blocks and conduits. This helps speed up the installation process. When blocks are made of a foamy material, such as EPS, that possesses some, but not very great mechanical strength, conduits may preferably by made of a predominantly polymeric material that has a sufficient creep rupture strength at elevated temperature. Polymeric materials are characterised by a much lower modulus of elasticity than metals, whereby obstructed thermal expansion of conduits will only create relatively small stresses, both in the material itself and in the insultaion blocks in which the conduits are embedded. PEX is currently the standard polymeric material used in district heating mains and can also conveniently be used in systems according to the invention. When F and R conduits circulate water that passes through a radiator system with steel components, it is essential to provide the pipe with a membrane that prevents oxygen from the ambient to become dissolved in the water. PEX is not weldable, but with systems according to the invention the number of fittings needed is limited. Above all, where a conduit ^branches' off onto its route up to a building there is no need for a T-branch, which is required in a comparable, conventional system. Rather recently it has been discovered that currently used membranes that provide efficient blocking of oxygen diffusion unfortunately do not stop diffusion out from the medium pipe of water vapour, especially at elevated temperature. In PUR-foam, surrounded by a shield pipe, provided with its own membrane to counteract out-diffusion of insulation gas, for instance carbon-dioxide, the thermal gradient tends to cause such water vapour to condense on the inside of the shield pipe, which increases heat losses and may cause long-term degradation of the foam. When the construction surrounding a polymeric pipe is EPS instead, there will already be some water vapour from the surroundings, so that a small increase in water vapour from the water-carrying pipes will only make a minimal difference. Thus, the water diffusion problem is not a serious one when using EPS insulation blocks. Polybutene is a well-known material that could be used instead of PEX as a material for conduits. A recently marketed, further material is polyethylene that has been treated in other ways than the classical cross-binding, used for PEX to achieve increased strength at elevated temperatures. This new group a materials, termed PERT, in addition to being competitive with PEX in terms of price, possess several properties that are attractive in the context of the invention: Foremost, PERT is softer than PEX, which will make bending more easy. Also, PERT is weldable, a property that, though needed in most parts of the system, will be an advantage where conduits are to be united with components at the ends of the conduits, for instance at a branching station, at the building, and where a dummy conduit at a later stage is being extended to connect a new customer building to a system according to the twelvth embodiment of the invention. Figs. 18e, f, and g are: a side-view (A-A) , a front- end view (B-B) , and a top view (C-C) , respectively, of a branching station, STAbr, together a view of a right-hand part of a transfer station, STAtr and a left-hand part of a first assembly of insulation blocks BLOCK1, BLOCKl' , BLOCK1", BLOCKl'", and BLOCKl"". STAtr has an outer, casing part STAtrcas, providing mechanical and moisture protection, as well as heat insulation of the interior that can be heat exchangers, control equipment, etc. The assembly of insulation blocks mentioned constitute part of the MAIN conduit line, viz. the first of the series of assemblies of blocks providing a casing for co-extending conduit first portions, CONa, a selection of which blocks and conduit first portions are shown in figs. 18a, b, and c. The branching station interior contains four branching boxes, BOX', BOX", BOX'", and BOX"". Each box comprises a smaller primary chamber and a bigger secondary chamber, for instance CHAp' and CHAs' , respectively, constituting the inner parts of BOX' at the top. Each primary chamber is connected to a pipe, for instance pipe PIPEhwf that leads domestic Hot Water Forward (HWF) into BOX' from the transfer station and extends backwardly into the transfer station, through STAtrcas. Inside the transfer station PIPEhwf is provided with a shot-off valve, VAhwf . To the right of each box, ten conduits, CONl-10 are connected in such a way that they inside converging chamber part CHAcon of STAbrcas converge towards a certain of the four layers of conduit portions CONa. Inside loose insulation material, INS, provides thermal insulation between the various layers of converging conduits. Boxes ' and right-hand part of pipes are arranged inside an assembly of five insulating blocks, viz. BLOCKst, BLOCKst' , BLOCKst' ' , BLOCKst' ' ' and BLOCKst' " ' , arranged on top of each other; they can be made of the same material as insulation blocks comprising conduit parts CONa, for instance EPS. The top view of fig. 18g is a view of a section (indicated by arrows C-C in fig. 18f) that appears if one imagines the top BLOCKst having being taken away. In addition to the four pipes connected to boxes, a fifth pipe, PIPEcw, leads cold town's water into STAtr. It can be seen that by this arrangement each of the four flows HWF, RW, F, and R are divided into individual conduits outside STAtrcas in a compact way that forms a close-fit arrangement with STAbr. Boxes and other system parts inside STAbr can be made of corrosion resistant materials. Thus, it can be tolerated that insulation blocks of STAbr are made of a material that allows moderate amounts of water to enter the insulation blocks. Control equipment and possible metering equipment that tends to be sensitive to moisture are contained within STAtr, whose casing can be made of, for instance outer and inner layers of polyethylene, with an insulating polyurethane layer in between. By providing STAtr with a lock that can be demounted, such equipment can rather easily be inspected, and by suitable arrangement of pipe connections, greater or bigger portions of the interior of STAbr can be exchanged rather swiftly for control and, if necessary, repair or component replacement in a laboratory in suitable testing and working conditions. Arranging STAbr outside STAtrcas has the advantage of minimising the size of STAtr. Sometimes, when available space underground is sparse, it may even be advantageous to make PIPEs longer and separate STAbr from STAtracas, i.e. not arranging the two stations adjacent to each other. Figs. 19a, b, c, d, e, and f show an apparatus intended for rapid and precise mounting of conduits according to the twelvth embodiment of the invention, as shown in figs. 18a-g. Fig. 19b is a top view from a horizontal plane section of the apparatus, as indicated in the side view 19c. Fig. 19a is a view from the rear of the apparatus, together with a cross-section of insulation blocks BLOCKn', BLOCKn", BLOCKn'", and BLOCKn" " , and with all four layers of adjacently conduit parts CONa, constituting the MAIN assembly for leading fluids HWF, CW, F, and R. As shown in figs 18a-d conduit lines CONb and CONc can branch off to lead up to individual buildings. In figs. 19a, b and c the apparatus and MAIN are shown at a stage when the upper-most conduit layer, that of HWF, is being arranged, before the upper-most insulation BLOCKn has been laid down, with an elastic disc, ELA, (cf . Fig. 18c) to be arranged between this insulation block and the top layer of conduits. Variations of the design of the tool can be tailored to other embodiments of the invention. That is, the tool illustrates some basic ideas that are not restricted to this particular embodiment. The apparatus is essentially a roll with conduits that is intended to be drawn along an already dug out DITCH for arranging conduit portions CONa underground, completed by a simple twin-roll tool for pressing conduits down to their final positions, as indicated in previous figures 18a-c of the last described embodiment of the invention. All conduit parts CONa belonging to a certain layer are rolled out simultaneously, which speeds up the procedure. The apparatus can be moved, either by hand or by a (not shown) remote control, while three wheels, WH1, WH2, and WH3 run on the surface of the ground, two wheels on one side of the ditch and the third wheel on the other side. To facilitate precise movement of the apparatus it may be supplemented by a (not shown) positioning system that (for instance by use of laser beams) can guide the apparatus to closely follow the course of tracks along insulation blocks arranged underground. Half of the roll (the right-hand part in fig. 19c) is being heat insulated from the surroundings by an insulating shield, INS; the apparatus may be provided with (not shown) a secondary, movable (by turning around the axis of the roll) shield that can insulate most of the left-hand part of the roll. This helps keep conduits warm while being mounted. The bundle of conduits CON are being pressed down by the weights of rolls ROl and R02, that may be supplemented by (not shown) an extra weight (or a manually operated stick) attached to connecting ARMl . The assembly of ROl, R02, and ARMl are attached to bottom FRAME of the apparatus by connecting ARM2. The ten conduits are rolled out from an assembly of ten DISCs, all shown in fig. 19d. In fig. 19e is an enlarged view upper half part section of the right-hand outermost DISC (with windings of one conduit) is shown. It can be seen that all discs are of equal width, but that there is a variation of from one disc to another of the radius of transition from a solid, inner part to an outer, thin part providing side-wise support of conduit windings. The inner-most discs support more windings. Each disc supports an appropriate length of a conduit, a length that initially (before movement of the roll has started) normally will comprise an entire conduit, from its connection to branching station STAbr to the particular building up to which the conduit extends when all parts of the conduit have been arranged underground. By the example shown in fig. 19d DISCs and conduit lengths have been sized and arranged such that conduit lengths increase towards the centre, both from the right and from the left. Such an arrangement will be suitable when an equal number of conduits are supposed to ^branch off to the right and left of the main trench. Each conduit will be sized with some extra length, so that the final length of the conduit can be determined in the field when most of the conduit has been laid down. Fig. 19f shows a enlarged view of the central part of the assembly of DISCs, mounted onto the apparatus. From the right-hand and left-hand sides, respectively, the discs are pressed together by way of two cylindrical members, CYL1 and CY12, respectively, that are screwed inwardly. In addition, SCREWs hold the DISCs together, between two further, outer DlSCoutl and DISCout2. CYL1 and CYL2 are each arranged within bearings, BEARl and BEAR2 , respectively. Intermediate disks DISKintl and DISKint2 are inserted between bearings and the outer discs of the rotating assembley of DISCs. Finnally, nine, small DISCsmall are arranged between DISCs to assist in securing a concentric assembly of DISCs of the roll. The apparatus is intented to be operated in the following way: An assembly of DISCs with conduits CON, as shown in fig. 19d, will have been mounted in a shop, to arrive at the building site. An appropriate diameter of DISCs can for instance be 2.4 m, which in some countries is a common standard for rolls of many sorts of conduits (district heating flexible pipes, electric conduits, etc), permitting easy loading on standard trucks for road transportation. If, for example, conduits are sized to an outer diameter of 20 mm, this will allow for a conduit to have a length of up to around 300 meters, which in many cases will be sufficient for an unbroken conduit length from STAbr up to a farthest away building to be connected. If the distance is significantly greater, an extension conduit that can be connected onto a rolled-out conduit from DISCs will be added. The apparatus is taken to a MAIN trench where all BLOCKS, appropriate for positioning a certain layer of conduits, have been laid down into the trench. The apparatus is positioned on top of the TRENCH, close to a position where STAbr is intended to be situated or has been arranged already. An appropriate amount of MAIN elements, corresponding to the particular layer of conduits to be arranged will have been put in place. All conduits are rolled out a bit, so that ends of conduits can be fitted onto the particular box of STAbr. Then the entire apparatus is moved along the trench, by which ever longer lengths of CONa parts of all conduits of the layer are being arranged underground. When the first point of departure ( ^branching off ) of a conduit has been reached, the particular DISC comprising the rest of the conduit to depart is de-mounted from the apparatus. Thereafter the apparatus, with the rest of the DISCs is re-assembled, one of the intermediate discs (DlSCintl or DISC2) being replaced by a thicker one, so that there will be no need for moving the entire apparatus, including its wheels, sidewise. The apparatus is then moved further down the trench by one man, while the de-mounted disc remains behind. A second man can take care of this disc, forming the bend CONb of the conduit, rolling out the rest CONc of the conduit, arranging in its appropriate place, i.e. within an insulation block groove, as shown in fig. 18d. The conduit may be rolled out at once, or - especially if the rest of the conduit is relatively long - the disc may be mounted on a second apparatus with wheels, running either along the branch-off ditch, or on insulation blocks, i.e. close to where portion CONc of the conduit is to be arranged on its route up the building. When a group of houses are connected to a district heating system, they will usually comprise the majority of houses within a certain area to be served by district heating. Yet there will often be a rest of houses not being connected initially but maybe at a later stage, for instance when the ownership of the building changes. With the embodiment of the invention shown by figs. 18a - e, easy later connection can be prepared by initially laying down a dummy conduit, comprising portion CONa and a little more conduit length, supplied with an end fitting that makes the conduit end blindly, the part making this end fitting later to be exchanged by a fitting of an extension conduit that may later be provided to connect the further building to the district heating system. Rolls with an assembly of conduits can be sized to match the individual lengths of conduits needed for a certain supply area, served by a given branching station. If this approach is adapted, the design of the discs could be modified, or they could be replaced by other elements that could be more or less similar to per se known types of rolls used to wind up electrical cables and other conduits on rolls. For instance, a conduit need not be supported at all points round the innermost winding. Round bars or pins supporting at, for instance six points round the periphery, may be sufficient. The roll can be designed such that individual length assembly packages can be arranged onto the roll by inserting pins at various positions in arrays of holes in appropriately designed discs or other mechanical elements of a roll. An alternative or complementary approach is to manufacture a few selections of standardised package rolls, such that there in most cases will be shorter or longer excess lengths of conduits. If conduits are rather cheap, a loss of, say 20% conduit length, due to such standardisation, may be fully acceptable, since the concept will save time when preparing rolls in a manufacturing plant. It has already been pointed out that conduits may be pre-heated and that the heat insulation of rolls mounted in the apparatus can be complemented. A further way of controlling the temperature of the conduits is to provide the apparatus with heat storage and / or heating means, such as electrical resistance heating or an oil burner heating element. Such an element could be arranged centrally in the apparatus, and it could be arranged to blow hot air through each of the conduits. A careful control of conduit temperature when conduits are being arranged underground can help prevent damages, especially to oxygen diffusion barriers that may suffer damage if conduits are being bent in a cold condition. Regarding bending radii, especially of conduit portions CONb, where individual conduits ^branch off from MAIN, differing policies can be adopted. Figs. 18a and d indicate some bending radii in a horizontal and vertical plane, respectively. Considering the entire 3-dimensional curves of conduits the effective, minimal bending radius of a conduit bend can be calculated. The embodiment shown adopts minimal radii that are of the same order of magnitude as is commonly used in bends of meander-like patterns of polymeric conduits arranged in under-floor heating systems. Here, bending radii are typically in the order of 10 times the diameter of the conduits, i.e. r/d = 10. If minimal bending of pipes is considered essential to cause minimal damage to diffusion barriers, the embodiments of the invention can easily be designed such that r/d is significantly greater than this value. A first, simple modification could be to arrange transitional portions CONb of conduits to bend only in the horizontal plane, so that essentially all parts of a conduit (with the exception of end parts) extend at roughly the same depth below the surface on the ground. Another approach will instead rely on utilising a good control of conduit temperature when the conduit is being arranged underground, to allow for smaller bending radii, i. such that r/d ≤ 10. This could be used to dispense with the rather bulky element containing loose insulation material, INS that is shown in figs. 18a and c. It is even possible to have all bending of a conduit, i.e. the entire conduit part CONb, arranged within the boundaries of MAIN. Fig. 20 a, b, c, d, and e show a thirteenth embodiment of the invention where such an approach has been adopted. Fig. 20a is a cross-section of two insulation blocks and six conduits, each one arranged in a groove of the lower insulation block. Two grooves are empty; it can be imagined that two conduits have already ^Λbranched off at positions closer to a branching station than where conduits are shown in the figures . More insulation blocks and conduits (not shown) may be arranged beneath and/or above the two blocks shown in the figure. Fig. 20b is a top view of one insulation block and parts of adjacent blocks, as well as the six conduits. Conduit no. six (counting from the left in fig. 21a) is bent to ^xbranch off , downwards in fig. 20a. It can be seen that the bending radius r is rather small, compared to the conduit diameter d. Fig. 20c is a side view, showing a cross-section of the branched-off conduit, positioned within a flat element made of a bent steel plate, with the 180 degrees bend at the far left end of the element, and a polymeric, flat plate sandwiched between the upper and lower flat parts of the steel plate. The right-hand end of the plate, as shown in fig. 20c is rounded inwardly by a groove, to support part of the segment (a part to the left) of the conduit, so that there is more than a point contact (viz. what may be termed a ^Λsegmental contact') between the plate and the conduit. From fig. 20b it can be seen that the groove follows the conduit backwards inwardly, even to provide a segmental contact between the plate and the last conduit part CONa, leading up to the transitional part, CONb. Thus, the conduit in its curvature is supported by double-curved contour of the plate on an inner segment of the contour of the conduit. Fig. 20c shows the plate in two positions: Its final position, drawn in full lines, and a position where the conduit to be bent is being temporarily bent upwardly, supplemented by a bending tool that in fig. 20d is shown as seen from above. A hole in the sandwiched plate is used as a pivot for a bar with a wheel that touches an outer segment contour of the conduit. When the bar and the wheel are moved (downwardly in fig. 20d) the conduit is being bent. If necessary to prevent elastic rebounding of the conduit after having being bent, a clips (not shown) can be fastened round the conduit and pressed into polymeric plate. When the bending has been completed, to tool is removed, and the sandwiched plate, together with the bent conduit can be laid down to their final position. When a circular pipe is being bent its cross-section will tend to become ovalised, with a bigger diameter in the plane perpendicular to the curvature of the pipe and a smaller diameter within the plane. The arrangement shown in the figures could be designed to counteract such ovalisation. However, from the point of view of minimising stresses built into the conduit by bending it, a better strategy is to allow for such ovalisation. On the other hand, ovalisation may in the end lead to a local collapse of the conduit cross-section, a collapse that may take place at a point where the conduit happens to be relatively less stiff, due to geometric tolerances and / or a locally higher temperature. By supplementing the tool by heating means (not shown) one can ensure that the temperature distribution along the pipe where it is being bent is rather homogeneous . Fig. 20e shows a detailed geometry that can be used, either to dispense with such homogeneous heating, or in combination with it. Shown is an enlarged view of the conduit between the sandwich plate arrangement, the conduit cross-section being slightly oval, as is emphasized by comparing it with the initial, circular cross-section shown to the right of the oval. The thickness of the polymeric plate is somewhat bigger than the diameter of the circular conduit, allowing for ovalisation of the conduit being bent. On the other hand, a maximum is imposed on the vertical and bigger diameter of the oval. Thus, summarizing, the sandwiched plate has been designed to prevent two phenomena during pipe bending: A local, smaller radius of curvature than r and a local over- ovalisation. Both phenomena are potentially dangerous from the point of view of harming the conduit, especially any oxygen-diffusion stopping barrier, such as an aluminium membrane, built into the conduit. On the other hand, a minimum of stresses are being built into the conduit. Both in the embodiment of the invention shown in figs. 20a-e and in the preceding embodiment, heating of conduits has been foreseen, as has been explained. There is no doubt that whenever polymeric conduits are used, mounting them at a material temperature corresponding to at least room temperature is better than if they are laid down at cold ambient temperature. In the winter season of a city situated in a temperate climate zone, air temperatures are often drop below zero degrees Centigrade. It is well known that under such conditions polymeric conduits can be difficult to handle. It seems probable that an optimal material temperature is even higher than normal room temperature. Laboratory experiments can clarify what temperature, depending on method on mounting, bending radius etc. is optimal. At such experiments de-aerated water can be circulated within conduits arranged in a test rig. The concentration of oxygen can be monitored, to detect any rise that may be attributable to damage to a membrane, caused by bending. A bent conduit can be monitored in parallel with a straight conduit. In such experiments conduits can be exposed to various types of overload, such as high pressure, pressure transients, excessive water temperature, etc. Very small bending radii may be used as a further way of accelerating such tests. In both embodiments nos. twelve and thirteen, conduits are shown to have a circular cross-sections, thus being conduits that can be supplied directly from stock from manufacturers delivering conduits for under-floor heating systems and other applications. It other embodiments of the invention the cross-section of conduits will be tailored to achieve particular features. Thus, conduits can be provided to have a squared outer contour, which increases the surface of contact between adjacent conduits arranged, as shown in fig. 2. Also, the surface may be conditioned, for instance roughened, to increase the coefficient of friction between a conduit and the contact surface of insulation blocks, to improve prevention of thermal expansion of conduits. The invention has been described above with reference to several embodiments given for illucidating the invention. However, the different features mentioned may be combined in different manners than described in the embodiments and all such combinations should be included within the scope of the invention, which is only defined by the appended patent claims .

Claims

1. A thermal energy distribution system for supplying several buildings with thermal service characterized by at least two of said buildings being fully detached from each other, and in that said system comprises : a branching station; several conduits each forming an essentially unbroken connection from said branching station to a thermal connection in or close to each said building, each conduit being flexible and having essentially the same cross-section over the entire length hereof, and at least two of said conduits extending essentially adjacent to each other over a first portion (CONa) , and extending to each building over a third portion (CONc) and having a second portion (CONb) forming a transition between said first and third portion.

2. The system of claim 1, characterized in that at least one group of said first portions (CONa) of said conduits extend within or along a street.

3. The system of claims 1 or 2, characterized in that said conduits are essentially thermally un-insulated, thermal insulation being arranged around said conduits .

4. The system of claims 1 or 2, characterized in that said conduits comprise thermal insulation being arranged as at least one part of said conduits.

5. The system of any of the preceding claims characterized in that at least one of said conduit is made with one or more channels for leading fluid flow, said channel (s) being formed within said conduit that is either fully polymeric or mainly polymeric, in the latter case being provided with at least one metallic membrane arranged to prevent gas diffusion across a said pipe.

6. The system of any of the preceding claims characterized in that at least one said conduit is made with one or more channels for leading fluid flow, said channel (s) being formed within said conduit that is either fully metallic or mainly metallic, in the latter case with one or more polymeric layers on the inside and / or on the outside of the metallic part, the metallic part providing the main mechanical strength of said conduit.

7. The system of claim 5 characterized in that said at least one conduit is prefabricated in the form of an integrated polymeric structure being fabricated from the same base material and in a simultaneous manufacturing process.

8. The system of any of the preceding claims characterized in that the outer shape of said conduits is circular.

9. The system of any of claims 1 to 7 characterized in that the outer shape of said conduits is square, rectangular, or hexagonal, whereby voids between of the conduits can be minimized when they are arranged adjacent to each other.

10. The system of any of the preceding claims characterized in that said transitional portions (CONb) of said conduits are curved with a minimal radius of curvature that is at least 10 times the cross-sectional size of said conduit, taken in the direction of said radius, said size in the case of a circular shape of said conduit being the diameter of said conduit.

11. The system of claim 10, characterized in that said radius of curvature is instead less than ten times said cross-sectional size.

12. The system of any of the preceding claims characterized in that said first portion (CONa) of said adjacent conduits being arranged within a casing.

13. The system of claim 12 characterized in that said casing is formed mainly by blocks of insulating material.

14. The system of claim 13 characterized in that said insulation blocks are made of a polymeric material to form either of the following material structures: a permeable structure made up of coherent, small elements, or an impermeable, relatively stiff structure of mainly closed cells, or an impermeable, relatively flexible structure of mainly closed cells, such that either all said blocks are formed to the be of the same type of structure, or there is a combination of blocks made of differing structures.

15. The system of claims 13 or 14 characterized in that at least one part of said conduits is arranged within grooves made either directly as parts of said blocks or by addition of further system elements.

16. The system of any claims 13, 14, or 15 characterized in that said second portions (CONb) of conduits are either inserted into curved parts of said grooves according to claim 15 or are inserted into supportive structures that lead said conduits along predetermined curves .

17. The system of claim 16 characterized in that said second portions (CONb) of said conduits, when being arranged to be curved, are arranged within a structure provided with a solid boundary that allows for some deformation of the cross-section of the conduit, but prevents deformation beyond a certain, predetermined limit in order that local over- deformation that may compromise the structure of the conduit can be avoided.

18. The system of any of the preceding claims characterized in that thermal insulation of said conduits is made in the form of super-insulation comprising a vacuum between either gas impermeable foils and/or grains and/or open-celled insulation material contained within a structure that is substantially gas-tight towards the surroundings, said vacuum to be established, either prior to said conduits being arranged underground or afterwards, said structure either being of such stability that a vacuum can be maintained for at least five years without any additional measures, or said vacuum to be intermittently re-established and/or continuously upheld by having a vacuum pump in connection with said vacuum, and/or said vacuum to be continuously upheld by having a getter material being distributed throughout said vacuum, thereby removing from the vacuum molecules that will tend to otherwise degrade said vacuum.

19. The system of any of the preceding claims characterized in that buildings are connected to a forward line of said conduits, leading fluid flow to said building and a return line leading fluid flow from said building.

20. The system of claim 19 characterized in that said buildings are additionally connected by at least one further line of said conduits, leading hot service water to a said building, said hot water being prepared commonly for supply to all said buildings connected to a said branching station.

21. The system of claim 20 characterized in that said system is provided with circulation of hot service water taking place within at least two said lines for each said building, and that in at least one said line the flow can be reversed to temporarily flow in the opposite direction of re-circulation, i.e. to said building instead of from said building.

22. The system of claims 20 or 21 characterized in that four or more lines of fluid flow flowing in said conduits are connected to the said buildings.

23. The system according to any of the preceding claims charaterized in that building heating and / or cooling systems of said buildings are connected directly to said conduits, i.e. without heat exchangers for hydraulical separation.

24. The system of claim 23 characterized in that said branching station comprises a fluid flow leakage detecting system and valves for closing a hydraulical loop for transferring one or more fluids into a building heating and / or cooling system in case of a leakage of said loop being detected.

25. The system of any of the preceding claims characterized in that heat metering for accounting for the amount (s) of thermal supply to each building is made inside or adjacent to a said branching station.

26. The system of any of the preceding claims characterized in that one or more hydraulical lines leading fluid flow to said branching station is provided with a turbine that can supply mechanical energy and/or electrical energy to at least one pump and/or control equipment and/or metering equipment and/or further units, all arranged within or adjacent to a said branching station, said electrical energy, when provided for, being generated in a dynamo driven by said turbine.

27. The system of any of the preceding claims characterized in that said flexible conduits are supplied to a building site mounted on one or more rolls .

28. The system according to any of the preceding claims characterized in that said conduits, being flexible, when arranged underground are rolled out from at least one movable roll being part of said apparatus, such that the part of at least one said conduit being arranged at a given moment is situated relatively close to said at least one roll.

29. The system according to claim 28 characterized in that said at least one roll is adapted to be supplemented by one or more tools for arranging said conduits precisely into their final, intended positions underground.

30. The system according to claim 29 characterized in that a group of said conduit first portions (CONa) are rolled out simultaneously from a single said roll.

31. The system according to any of claims 28, 29, or 30 characterized in that said conduits, when being rolled out have been heated to and/or maintained at a material temperature of said conduit that is at least room temperature, i.e. at least around 20 degrees Centigrade .

32. A method for constructing a thermal energy distribution system according to claim 1, for supplying several buildings with thermal service, wherein at least two of said buildings being fully detached from each other, said system comprising a branching station; said method being characterized by arranging several conduits each forming an essentially unbroken connection from said branching station to a thermal connection in or close to each said building, each conduit being flexible and having essentially the same cross-section over the entire length hereof, and arranging at least two of said conduits for extending essentially adjacent to each other over a first portion (CONa) , and extending to each building over a third portion (CONc) and having a second portion (CONb) forming a transition between said first and third portion

33. The method of arranging said conduits according to claim 32 claims characterized in that said transitional conduit portions (CONb) , when being bent are heated to attain a material temperature of the said portion that is at least room temperature, i.e. at least around 20 degrees Centigrade.

34. The method of claim 32 or 33, characterized by arranging at least one group of said first portions (CONa) of said conduits for extending within or along a street.