DE19530826C1 - Power output optimisation system for energy distribution system - Google Patents

Abstract

The power output optimisation system is effected by on-off switching of at least one of the loads (1,2,3) connected to the energy distribution system having an energy storage characteristic. The switching of the loads is effected in accordance with respective load priority values which are altered according to a function which simulates the energy storage time characteristic.

Description

The invention relates to a method for optimizing the Power consumption in a multiple power distribution system Consumers by switching at least one consumer on and off Energy storage behavior, in which a priority parameter for each of the at least one consumer for the time that it is switched off or on, is won and if necessary the consumer with the highest Priority parameter is switched on or off. The invention further relates to a device for performing the method, in which at least one controllable switching device for switching on and off at least one Consumer with energy storage behavior and a control device for To control this switching device are provided, which is constructed so that a Priority parameter for each of the at least one consumer for the time in which this is switched off or on, is obtained and the control device generates signals based on the priority parameter, with which one can if necessary Consumer with the highest priority parameter is switched on or off.  

Such energy distribution systems are known. You have become necessary because the energy supply companies not only used the tariffs Total energy but also peak performance taken from the grid consider. A uniform withdrawal of benefits is therefore subject to tariffs cheaper than a total energy consumption in a single high power peak once a day. That is why it is important to consider the energy requirements of a system as much as possible keep even.

This can be done, for example, in a hotel with a restaurant achieve that during lunchtime or in the evening, in which a lot of energy for Cooking is needed, the sauna is switched off. In this simplified For example, this can be done by hand. More complex systems of different However, consumers can only be sufficiently optimized using an automatic system. However, this simplified example demonstrates how these work Power distribution systems.

Much more effective measures than decommissioning whole Power groups consist of the mean power draw, for example also limit when cooking. A hotplate that has already heated up can be switched off over a shorter period of time, because of this their energy storage behavior only cools slowly.

Such short-term shutdowns are usually by special designed control devices controlled by a consumer message for receive a performance request, and the necessary energy supply due to the Optimize the connected loads of the consumers by switching them on and off. Here the control device optimizes the current power consumption by switching this consumer with energy storage behavior entered accordingly Connection values. When designing such a control device can also take into account that, for example, there are consumers who do not should be switched off, such as a bread or Sausage slicer in a hotel kitchen, where a sudden for a cook unexpected switching on could lead to an increased risk of accidents.

Such power distribution systems are now used in restaurants and hotels, as well as in factory canteens in use, since a clear saving effect is achieved. The control units are also not very expensive, so that they are quick amortize.  

Basic information about such systems can be found in the manuals for these systems, for example in the technical manuals "General Information ", 1994, from the company Ecotron Systemtechnik, Föhrenstraße 53a, 83052 Bruckmühl.

EP 0 053 383 B1 also describes such a system. With this one a priority parameter is assigned to each consumer, based on which with the help a control device decides which consumer should or is switched off. The following can be included in this priority parameter Enter current parameters:

• 1. The power consumption in the heating phase,
• 2. The total amount of time since a consumer switched from one Switching state to another switching state,
• 3. The total amount of time that a consumer is currently in the current switching status,
• 4. The ratio of the total time span of an electricity consumer in Switching status at the total time of the last cycle.

As can be seen from these parameters, these values only have something indirectly to do with energy storage behavior. The appropriate selection of Priority metrics are essentially based on experience to make this out to form such parameters. This is unsatisfactory because it is never guaranteed that the reduced power consumption due to these values is really optimal.

In principle, the temperature could, for example, be measured on a hotplate and at with the definition of the priority parameter. This is expensive because then additional lines for transmitting the temperature to the Control device are required. Next, none could be standard for Available kitchens and large appliances are used. A conversion would be always required.

The object of the invention is a method and an associated device with which to find an energy distribution system with the least possible Power consumption with little additional effort compared to existing ones Methods and devices can be improved.  

The object is achieved by the characterizing Features of claims 1 and 4 solved.

The above-mentioned principle of temperature determination can be expected unexpectedly use according to the invention with reduced effort, but by Temperature is not measured, but the change in temperature is at least approximately modeled. The additional effort compared to existing methods and devices according to the prior art is therefore small. Additional temperature sensors and a modification of existing ones Consumers are not required. In principle, for example when the procedure with the help of a computer, the additional effort is reduced essentially on a software change to determine the priority parameter, So on pure development costs, the price for the corresponding quantities of such a computer system in a negligible manner.

The method has compared to the temperature measurement mentioned, for example also the advantage that the real physical parameter to be optimized, namely the performance that can be simulated. The temperature, on the other hand, reflects only indirectly the performance again because a temperature drop also from the ambient temperature, depends on the heat capacity of the heated goods, etc. Therefore, with this Simulation a much better optimization can be performed.

According to a preferred development of the invention, the Priority parameter at least one summand with exponential Timing behavior.

Due to the training, a simplified approximation for the Energy storage behavior of the consumer used. An exponential Behavior after switching on or switching off a consumer provides a simple and almost realistic model for the Increase and decrease in performance of a consumer. This exponential Behavior can not take into account that the heat capacity of the Cooking food, for example, while boiling water Evaporation changes over time, however practice has shown that such effects make an insignificant contribution to effective optimization. The effort  remains with the exponential power simulation, the input to the Priority parameter finds, minimal.

According to an advantageous development of the invention, a characteristic Time for the exponential time behavior measured.

This can significantly improve the aforementioned training will. With a measurement, changes in heat capacity can also be made are taken into account, as the following example makes clear.

The time constant is reflected, for example, in the case of a thermostat-controlled one Consumers in the on and off time of the thermostat again. This is over the time recording of the performance requirement by the individual consumer accessible, so that for such a development of the invention only a small one Effort, such as a software change, is required.

The thermostat time reflects the power consumption in the same way as they are also expressed in the exponential behavior for the priority parameter is coming. If the stored in the priority parameter to simulate Energy in the consumer takes into account the time constant for exponential decay If this thermostat time is set proportionally, the effects become more variable Heat capacities are taken into account at the same time.

A device according to the invention for performing the method starts from state of the art mentioned at the beginning and is by one device to generate one from the time the consumer is switched on and off output variable that changes over time to simulate energy loss and Energy increase due to the energy storage behavior for each of the at least a consumer who helps determine the value of the priority parameter, featured.

This device is particularly suitable for its low expenditure Implementation of the method according to the invention. Compared to devices according to the prior art is only a device for generating the shown output size necessary. This facility can be used for Presentation of a very complicated situation, very complex, for example as Analog computer, can be executed or by means of a microprocessor or obtained within a computer program by a digital program  be, however, practice has shown that little complex equipment for Approximate simulation is sufficient.

A major simplification is with a preferential further education of the Invention achieved in which the device changes over time Output variable generated exponentially increasing or decreasing. At a digital control this can be achieved very easily because Exponential functions are usually more common to the software scope Computers belong.

In particular, the low effort required to simulate the physical facts becomes clear when compared to a simple analog circuit:
It is known that voltages that are exponential over time can be realized on a capacitor that is charged or discharged via a resistor. If the voltage that charges the capacitor via the resistor is switched on and off at the same time as the consumer is switched on and off, the resistor and capacitor are dimensioned accordingly, and a voltage is present across the capacitor that simulates the stored energy content.

This extremely simple circuit offers one in the way Analog computer a physical model for the energy storage behavior, whereby the capacitance of the capacitor model the behavior of the energy storage Cooking plate simulated. If instead of one resistor two resistors to Loading and unloading can be selected when the consumer is switched on or off, can also be the characteristic quantities for the energy increase and the Energy consumption of the consumer can be easily simulated.

The example shows that the power consumption when using Exponential functions in an extremely simple manner using one here as Voltage present output variable can be simulated. Due to the Similarities in the storage behavior of a capacitor and one energy-storing consumer, such as a hotplate, is one almost realistic simulation of everyone's energy storage behavior Consumers possible.  

In an advantageous development of the invention is a measuring device provided with which the time constant for the exponential course can be determined is. With the help of this device, the method shown in above perform a simple way by measuring the time constant. The at The advantages mentioned in the method also relate to the device.

In particular, in a preferred further development, in which at least one Consumer is controlled by means of a thermostat, the measuring device is a clock and / or a clock generator with which one can be detected via the measuring device Switch-off or switch-on time of the thermostat can be recorded and the time constant proportional to the switch-off or switch-on time of the thermostat.

As previously mentioned, the thermostat time can be calculated from the Time points for the energy request after switching on can be obtained. if the Thermostat circuit itself is not accessible, the thermostat time can also be off the load steps when determining the total power are determined: If one Microprocessor control is used, which is carried out with the help of all Switching cycles an n-parametric curve adaptation to the time course of the Total power consumption taking into account the switching on and off of the individual consumers through the control device of individual consumers in which the n parameters determine the thermostat times mean. Mathematical methods for curve fitting are known so that not to be discussed here in more detail.

As has already been mentioned several times, the device can be used in particular realize in a simple manner if the control device is a PC or a Has freely programmable microprocessor circuit in which the device to generate the time-variable output variable and / or the Measuring device for measuring the thermostat time as part of a program are realized.

The invention is described below using an exemplary embodiment further illustrated with the drawing. Show it:

Figure 1 is a schematic representation of a device according to the invention for performing the method according to the invention.

Fig. 2 of a priority parameter, a time dependence used in the embodiment.

In Fig. 1, the basic structure of a device is shown for energy optimization. Consumers optimized with regard to their total power consumption are provided with the reference numerals 1 , 2 , 3 . Electricity is supplied via the connections shown on the right side of the drawing. The electrical power is fed via a measuring device 4 to a common power supply 5 , 6 , 7 , from which it is distributed to the consumers 1 , 2 , 3 .

With the help of the measuring device 4 , the total power of all consumers 1 , 2 and 3 is measured. The power was recorded in the exemplary embodiment on the basis of a current measurement via an inductive coupling to the currents introduced via the connections and led to the consumers 1 , 2 , 3 . Furthermore, a phase factor of the current relative to the applied voltage was also taken into account in a manner known from the prior art.

The measured value for the total power consumption determined in the measuring device 4 is transmitted to a control device 8 which , on the basis of the measured total power, decides whether the total power is above a predetermined limit value and therefore the power consumption is reduced by switching off at least one of the consumers 1 , 2 , 3 got to.

For switching on and off, a switching device 9 , 10 , 11 is provided for each of the consumers 1 , 2 , 3 . The switching devices 9 , 10 , 11 are connected to the control device 8 via a bidirectional bus 12 . These switching devices 9 , 10 , 11 detect the power requirement, for example the switching on of a stove in a kitchen, of the consumer 1 , 2 , 3 to which they are connected. The information that a certain consumer 1 , 2 or 3 requires power is transmitted to the control device 8 by the switching devices 9 , 10 , 11 via the bidirectional bus 12 .

The bidirectional bus 12 , further serves in the opposite direction, from the control device 8 to the switching devices 9 , 10 , 11 , to address a specific one of the switching devices 9 , 10 , 11 and to signal to it whether the consumer addressed via the addressing 1 , 2 , 3 should be switched on or off. The switch-on and switch-off signal is stored in the switching device 9 , 10 or 11 , so that the bus 12 is free for further signals after the signal has been transmitted.

With the stored on or off signal, relays are switched, the normally open contacts of which switch the power from the power supply 5 , 6 , 7 to a power section of the selected consumer 1 , 2 , 3 .

If the total power is below the predetermined limit value, the control device 8 switches on the consumer 1 , 2 or 3 from which it has received a power request via the bus 12 .

Conversely, if the total power limit is exceeded, the energy optimization process begins. Consumers 1 , 2 , 3 with energy storage behavior are then switched off in the meantime. In order to decide which of several consumers 1 , 2 , 3 can be switched on again when the power drops, the control device 8 sets priority parameters for each consumer 1 , 2 , 3 , on the basis of which consumer 1 , 2 , 3 is determined, which must be switched on first, or it is determined which other consumer 1 , 2 , 3 can be switched off in the meantime because it has already stored enough energy.

As already described in the introduction using several examples, the control device 8 can be implemented with conventional circuit technology both analog and digital. In the exemplary embodiment according to FIG. 1, however, a computer was used which was programmed to calculate the priority parameter and which controlled the bidirectional bus 12 .

This computer also includes an internal clock generator that controls the timing of switching on and off. In other control devices 8 , which are constructed digitally or analogously, for example in conventional technology, such a clock generator or a clock with an oscillator and / or a counter can be constructed in a known manner.

In addition to the control function, the clock generator or the clock can also be used Inclusion of the time-dependent total power can be used, its temporal Course, for example, as described later, to determine the process used time constants can be used.  

The priority parameter calculated in the exemplary embodiment according to FIG. 1 was recalculated in predetermined time intervals for each consumer 1 , 2 , 3 . Depending on the time elapsed between the switching on or switching off of a consumer 1 , 2 , 3 , the energy increase or decrease in the consumer 1 , 2 , 3 was formed according to an approximate model for each consumer 1 , 2 , 3 .

As suitable for the approximate calculation of the energy storage behavior or the Energy loss have been shown to be exponential functions, since with this heating and cooling behavior mathematically simple and let it describe real conditions very closely.

In particular, the expression was used to calculate the priority code for reconnecting a consumer 1 , 2 , 3

100 _* [exp (t / tref) -1 + W / S]

used, in which t is the time elapsed from the time the consumer 1 , 2 , 3 was switched off, hit a time constant for the power loss of the consumer 1 , 2 , 3 in question and W or S are two constants which allow an individual adaptation to different ones Enable consumer types and which will only be discussed in more detail later. For the following considerations, it should first be assumed that W = 0.

A factor of 100 is also contained in the printout. This is due to the fact that the priority parameters are shown in percent when they are displayed on the computer screen. Such a factor is irrelevant for deciding which consumer 1 , 2 , 3 should be switched on based on the respective priority parameters if all consumers 1 , 2 , 3 are acted upon by the same factor.

However, such a factor has advantages in the case of a partially digital and partially analog circuit for the control device 8 , since the priority parameters can be represented as integers over a wide range and a circuit-like combination of digitally represented real numbers can then be dispensed with.

The expression given shows an exponential function that increases in value over time. This would not be expected a priori, since the energy loss is a function that decreases over time. Therefore, the power loss for the simplest realization of the solution of the task of the switched off consumer 1 , 2 , 3 according to the invention would be determined and the query for the highest power demand of a consumer 1 , 2 , 3 would be directed to a smallest parameter with exponential drop. With such a solution, different types of consumers 1 , 2 , 3 could also be taken into account by different scale factors.

Since the priority parameter was defined so that the decision about which consumer should be switched on or off to the highest Priority parameter is directed, the one used in this example Take the priority parameter, for example, as the reciprocal of this parameter.

In the present case of the exemplary embodiment according to claim 1, an already existing complex program was used, which controls the connection and disconnection of a consumer 1 , 2 , 3 via a query to the maximum of the priority parameter. It is therefore expedient, as in the expression for the priority parameter given above, to take the reciprocal of the exponential curve straight away.

Since the query for switching on and off according to the priority parameter is independent of a constant, any number can be subtracted from all consumers 1 , 2 , 3 when the priority parameter is formed. This was also used in the expression given by subtracting a 1 from the exponential function. With W / S = 0, this results in a representation for the priority parameter, the value of which disappears at t = 0. This behavior makes mathematically meaningful that the priority parameter for switching on has a vanishing value immediately after switching off.

With the additional size W / S in the expression given above, it is taken into account that for different consumers 1 , 2 , 3 different requirements when switching off and on may be appropriate. A deep fryer stores the energy supplied essentially in deep-frying oil, so that the type and age of the energy storage behavior are also important, while the energy storage behavior of a hotplate is essentially unaffected by such factors.

In order to be able to take such differences into account, the expression given above contains the summand W / S. With S, a so-called steepness S was introduced to differentiate the consumers, which is also used in other program parts for the control of consumers 1 , 2 , 3 . In the exemplary embodiment, the slope S was essentially inversely proportional to Tref and thus also means a weighting according to the energy storage capacity of the consumer 1 , 2 , 3 .

The factor W, on the other hand, is a freely selectable weight that can be entered into the control device 8 in order to adapt the priority parameters again specifically to different consumer types. In an experimental application of the exemplary embodiment, the value W = 9 was assigned to a deep fryer, for example, the value 2 to the hotplates.

By comparing the summands in the expression given it becomes clear that the exponential function with very large values of W / S and small times t contributes significantly less to the priority parameter than the size W / S. With the higher value for W in the deep fryer is therefore taken into account that due to poorly understood conditions, such as the condition of the frying oil, with a possibly incorrectly assumed exponential behavior not too great Errors can occur.

However, errors of this kind can also be avoided if the characteristic time Tref is determined via a measurement. In the case of thermostat-controlled consumers 1 , 2 , 3 , for example, a characteristic time for the energy storage behavior is given by the thermostat switching time, so that when measured, the time Tref can be set proportionally. The proportionality constant depends on the temperature at which the thermostat is switched on and off, and must be entered accordingly in the control device 8 .

The thermostat time can be determined, for example, with the help of the times for the power request via the switching devices 9 , 10 , 11 .

Another possibility for measuring the thermostat times arises when the control device 8 is, for example, a PC. Then the expected total power can also be simulated internally on the computer due to the switching on and off by means of the switching devices 9 , 10 , 11 and the desired parameters, like the thermostat times, can be determined by adapting the curve to the total power consumed by the measuring device 4 at defined time intervals.

The time-variable value of the priority parameter according to the expression given with W / S = 0 is shown in FIG. 2. The abscissa is shown in units of 100 _* t / tref, i.e. denotes the time in percent of the time elapsed with respect to the reference time tref. It can be clearly seen that the priority parameter varies considerably less with shorter switch-off times than in the vicinity of the characteristic time. In prior art devices, the priority parameter would increase in proportion to the time so that the priority parameter would be 50% at 50% of the characteristic time. A priority parameter of 50% is achieved in the exponential behavior of FIG. 2 only after approximately 85% of the time tref. This comparison clearly shows that the reactivation of a consumer 1 , 2 , 3 takes place much later in the method according to the invention than in the prior art. This saves energy much more effectively than in the prior art.

The switch-on is also initiated via the control device 8 due to the approximate simulation of the physical conditions with the help of the exponential function at a physically meaningful point in time, so that overall an energy optimization system in which the method according to the invention is used leads to a significantly better minimization of the overall performance of the consumers 1 , 2 , 3 can be used.

The method according to the invention has been shown above in particular for switching on a consumer 1 , 2 or 3 after it has been switched off. The same or similar expressions to be formed on the basis of the considerations presented above can also be used to obtain priority indicators for switching off consumers 1 , 2 , 3 .

Claims (8)

1.Method for optimizing the power consumption in an energy distribution system with several consumers ( 1 , 2 , 3 ) by switching on and off at least one consumer ( 1 , 2 , 3 ) with energy storage behavior, in which a priority parameter for each of the at least one consumer ( 1 , 2 , 3 ) for the time in which it is switched on or off, is obtained and, if required, that consumer ( 1 , 2 , 3 ) with the highest priority parameter is switched on or off, characterized in that the priority parameter is temporal is changed according to a function, the course of which, depending on the time, at least approximately simulates the energy storage behavior of the at least one consumer ( 1 , 2 , 3 ). 2. The method according to claim 1, characterized in that that the priority parameter has at least one summand with a exhibits exponential time behavior. 3. The method according to claim 2, characterized in that that a characteristic time for the exponential time behavior is measured. 4. Device for performing at least one of the methods according to claim 1 to 3, in which at least one controllable switching device ( 9 , 10 , 11 ) for switching on and off at least one consumer ( 1 , 2 , 3 ) with energy storage behavior and a control device ( 8 ) are provided for controlling this switching device ( 9 , 10 , 11 ), which is constructed in such a way that a priority parameter for each of the at least one consumer ( 1 , 2 , 3 ) for the time in which it is switched off or on, is obtained and the control device ( 8 ) generates signals based on the priority parameter, with which the consumer ( 1 , 2 , 3 ) with the highest priority parameter is switched off or on, characterized by a device for generating one from the time of closing - and switching off the consumer ( 1 , 2 , 3 ) temporally changing output variable to simulate the energy loss and the energy increase due to the E energy storage behavior for each of the at least one consumer ( 1 , 2 , 3 ), which also determines the value of the priority parameter. 5. The device according to claim 4, characterized in that the device exponentially changes the time-varying output variable generated increasing or decreasing. 6. The device according to claim 5, characterized in that a measuring device ( 4 ) is provided with which the time constant for the exponential course can be determined. 7. The device according to claim 6, characterized in that at least one consumer ( 1 , 2 , 3 ) is controlled by means of a thermostat and the measuring device ( 4 ) contains a clock and / or a clock generator, with which or via the measuring device ( 4 ) detectable switch-off or switch-on time of the thermostat is detectable and the time constant is formed proportional to the switch-off or switch-on time of the thermostat. 8. Device according to one of claims 4 to 7, characterized in that the control device ( 8 ) has a PC or a freely programmable microprocessor circuit in which the device for generating the time-variable output variable and / or the measuring device ( 4 ) for measuring the Thermostat time are realized as part of a program.