Method for automatic distributed load tripping to reduce active power flow through energy system elements when latter are overloaded. RU patent 2476969.

FIELD: electrical engineering.

SUBSTANCE: by way of preliminary analysis, a complex power pool grid is split into subsystems with minimum mutual impact. For each subsystem being controlled, based on the assumed splitting, one forms databases of sensibility factors and maximally allowed active power flows through links for different types of topology and accidental agitations. All the load lines and nodes of each subsystem being controlled are equipped with link agents and load agents. When any link is overloaded within the subsystem being controlled, such link agent, proceeding from information on the grid current topology and by way of linear optimisation tasks solution and by way of exchange of massages between link agents and load agents, calculates and implements the optimal control action vector. The algorithm keeps being executed until elimination of overloads with all the links within the subsystem being controlled or until any optimisation task being solved within the automation means operation cycle has a solution.

EFFECT: reliability improvement.

2 dwg

The technical field to which the invention relates.

The invention relates to the electrical engineering, namely to emergency management. The invention can be used to create a distributed automatic system that performs the functions of the load off of consumers for the purpose of reducing the active power flows through the elements of the power grid overloaded. The complex network interconnection must be broken down into subsystems that have the least interference. Automation each subsystem controls the current mode and prevents overload of the elements by issuing control actions (CA) on customer load shedding.

The level of technology.

There are a large number of ways local unloading interconnection of network elements by limiting the power flow [Russia. Patent №RU 2023337, cl. 02J 3/06, 1994]; [Russia. Patent №RU 2011263, cl. 02J 3/06, 1994]; [Russia. Patent №RU 2015601, cl. 02J 3/06, 1994]; [Russia. Patent №RU 2069437, cl. 02J 3/24, 1994]. These methods are used for the local flow control by one or several elements (as a rule, included in parallel) without accounting for the impact of changes spilling on the other links on the interconnection. Such a local approach when managing loading associations that are in the same section, or relationships, the change in power flow which will not have a significant impact on the loading other system elements. However, for systems with a complex structure in which you cannot clearly define the starting and the receiving part, ways of local control of power flow can cause invalid transshipment of other network elements.

Also known methods of centralized control power flows [«guidelines for the stability of power systems», Approved by the Order of the Russian Ministry of energy 30.06.2003 №277]. If you are currently running centralized emergency control (PAH) power pool is divided into areas of PAU. The regulated flow loft and partial cross sections with the necessary reserves. Usually the borders of the districts of PAH are defined as technological peculiarities of the regions and the possibilities of organizing a reliable and fast television necessary information on patterns and modes controlled network. Various emergencies lead to line power section, and to prevent violations of the stability on the loft emergency control unloads by switching off the load, or by changing the generation of starting or receiving parts. Selection of settings emergency control is carried out taking into account the severity of the disturbance, the current modes and schemes district of PAU, in which there was imbalance of power. The described approach is currently being successfully implemented when PAH UES of Russia. However, the growth of electricity consumption and the subsequent network construction, as well as the accompanying restructuring of the power industry and the transition to the market have a significant impact on the functioning of the power system. Further complication of the structure of a network of large power districts, most likely, will lead to the necessity of creation in these areas a single centralized system of PAHs, which will control power flows on a separate loft, and operation of all energy district as a single object of management. But even with the possibility of using the existing ideology of PAH in relation to the management of major with a complex structure, algorithms centralized complex will undergo significant changes connected with the necessity of account mutual influence of flows on various cross-sections.

The main disadvantage of large centralized complexes PAHs include high cost of implementation and operation. In particular, the speed of implementation of HC for centralized complexes PAH set standards, however, an increasing number of communication channels to provide the required speed of implementation of hydrocarbons will be increasingly difficult due to the necessity of synchronization of measurement in time. Thus, the growth of the complexity of the centralized complex will considerably raise the cost of its creation and operation. In addition, centralized complexes PAH not take into account the possibility of the emergence of low-probability events, and as the practice, it is unlikely blend of failures are the reason of occurrence of major system crashes. The issues of reliability of functioning of the large centralized complexes PAH should be given special attention. The reliability of large centralized complexes PAU will essentially depend on the reliability of a large number of channels of data transmission. Of course, we can talk about reservation of communication channels and the development of some adaptive centralized algorithms that must only part of the information about the current state of the system. However, it is important to note two aspects. First, the channel reservation itself is pretty costly exercise. Secondly, if algorithms centralized systems will require the full set of data, such algorithms are unreliable nature. If develop some centralized adaptive algorithms running on a random amount of information, these algorithms are not only complex in themselves, but will create significant difficulties in implementation of interaction with the terminal (grassroots) devices PA.

The above arguments are not calling for the abandonment of the existing system of centralized PAH, the effectiveness of which has been tested by time. The proposed method is automatically distributed load disconnection is designed to raise the intellectual level of high-speed devices, local automatics, which controls the loading of the intersystem ties. Local devices PAH always worked in conjunction with the centralized complexes PAH. Thus, implementation of the proposed method will lead to higher levels of system reliability PAH in General. In addition, the proposed method will allow to reduce the costs of switching off the load at the expense of realization of optimum HC.

As a prototype method, you can bring the [Author's certificate of the USSR № SU 1785063, cl. 02J 3/24, 1992]. The copyright certificate describes how to reduce damage from disconnecting the load in the power system with the chain structure. Authors pay special attention to the issues of mutual influence of subsystems and reduce damage from disconnecting the load. As disadvantages of this method, you can specify the possibility of its application only to power supply systems with chain structure.

The new method is applicable for power systems with a complex structure, it is able to adapt quickly to changing conditions of operation of a power system that can adequately react to unforeseen emergencies. The proposed method has high availability and uses the local information about the state of the regime and a minimum amount of information obtained from nearby measuring devices. To minimize the amount of information to control loading of network elements will enhance the reliability of the system PAH, as, for example, failures of communication channels transmitting the information to the centralized system of PAH, automation, based on the proposed method, will remain in work. In addition, the proposed method implements optimal or suboptimal of hydrocarbons, which will minimize the amount of load.

Disclosure of the invention.

The proposed way to address the challenge of distributed control of loading of the network elements in a complex interconnection. The main technical result of the proposed method is to reduce the download overloaded elements. Decrease in loading overloaded elements is done by switching off the load of consumers.

The essence of the invention disclosed on the example of the power system, the scheme is shown on figure 1. Power flows on cross-section of 1 can be controlled using well-known methods, including using the prototype method [Author's certificate of the USSR №SU 1785063, cl. 02J 3/24, 1992]. However, using known methods are difficult to control the loading of the elements inside the subsystem 2 and subsystem 3. In this particular case, the proposed method can be aimed at solving the tasks of control of startup items in the subsystem 2. Conditions of the method include the following preliminary set of actions:

5. All line and load centres of each of the controlled subsystems involved in the automation, based on the proposed method, shall be equipped with agents. Agents relations (4) and agents load (5) is an intelligent electronic devices, each of which has a unique identification number (agent). Using a single information space, agents are able to transmit to each other informational messages. Software and hardware implementation of the agent shall be sufficient for the implementation of this approach.

6. Inside each controlled subsystem should be implemented information system that is able to organise the transfer of messages between agents. Any agent of the line at any time should receive information about the current network topology, which contains information about the fact disable the subsystem. Furthermore, the agent connection with certain intervals, depending on the level of critical communication channels loading, should receive information about the current state of load nodes that have the greatest impact on the download links controlled by the agent. Information about the current state of load nodes should include information about the current number of working connections, loading the data connections, as well as the price that you will pay in the event of a power of each of the connections.

7. All agents relations in each of the controlled subsystems receive an appropriate factor database sensitivity and maximum permissible flows active power relations controlled subsystem.

The proposed method PAH includes the following set of actions (points), performed in a cycle:

1. Based on information on the current network topology, agents all the overloaded relations choose the appropriate table of coefficients of influence. On the basis of the table of coefficients of influence and information on the current load nodes that have a significant impact on the download overloaded connection, make preliminary calculations of optimum volume of the load. For each j-th overloaded due this calculation is the solution of the following linear optimization problem:

where a lot of load nodes in the subsystem, participating in the work of automatics; S, and P i - respectively, the value and volume of off the i-th loading; - respectively, the maximum and the current flow of active power on the j-th overloaded communication; k ij - factor of influence, which determines the magnitude of the impact of changes of the i-th load the flow of the j-th of communication; α - minimum allowable value of the coefficient of influence, a & GE 0; - the current value of the i-th loading. As a result of the solution of optimization problem must be received by the vector control actions (CA)that contains the optimal amount of shut-off for the load nodes in a controlled subsystem. If the task of optimization has no solution, the work of the algorithm terminates.

2. Each agent overloaded communication on the basis of the current vector HC generates a list of unique relationships with negative coefficients of influence. The formation of this list is the following : if disabling load in a certain node n belongs to G leads to increased flows in some relation n belongs to L (here a lot of load nodes included in the current vector of hydrocarbons; L is the set of controlled ties in the subsystem), this communication must be once included in the list of unique relationships with negative coefficients of influence.

3. Agent overloaded requests information on the current flow of active power relations, included in the list of unique relationships with negative coefficients of influence. Usually this list with too many links, because, as a rule, load leads to a decrease in load most of the elements in a controlled subsystem.

4. After receiving information about the current active power flows for relations from the list of unique relationships with negative coefficients of the influence of the agent each j-th overloaded, performs the re-calculation of optimum volume of the load. This takes into account the impact of switching off the load of communication with negative coefficients of influence:

where a lot of load nodes in the subsystem, participating in the work of automatics; G - a lot of load nodes included in the current vector of ETA, G ⊂ B; T - multiple links, included in the list of unique links; i and P i - respectively, the value and volume of off the i-th loading; - respectively, the maximum and the current flow of active power on the j-th overloaded communication; - respectively, the maximum and the current flow of the active power of the m-th communication, m belongs to the interval T; k ij - factor of influence, which determines the value the impact of changes of the i-th loading on overflow of the j communication; and β - minimum allowable values of factors of influence, a & GE 0, b & GE 0; - the current value of the i-th loading. As a result of the solution of optimization problem must be received by the vector control actions (CA)that contains the optimal amount of load shedding in a controlled subsystem and its impact on communication with negative coefficients of influence. The elements in the vector control actions (CA)received as a result of optimization, rounded on the basis of previously received information about downloading accessions load nodes.

5. For each element of the vector HC produces a list of links that were previously not included in the list of unique relationships with negative coefficients of influence. The list of links that were previously not included in the list of unique relationships with negative coefficients, is formulated as follows. If disabling load in a certain site i let In, within the vector of hydrocarbons leads to the increase of cross-flow some connection m belongs to L (here L is the set of controlled ties in the subsystem) and if the connection m not previously included in the list of unique relationships with negative coefficients of influence, this relationship should be included in the list of links with negative coefficients of influence. If the generated list of links appeared to be empty, then one proceeds to step 6, otherwise move to step 2.

6. Each element of the vector of hydrocarbons obtained in step 5 is included in the information request message and transmitted to the relevant agent load. Agent load handles requests from agents relations, and implement the requested HC. If the agent loads are not able to execute disable the requested accession or if the amount of load on the accession of a substantially smaller than requested, it sends the agent connection information message about the impossibility for the shutdown of the requested connection. This report should contain the information about the current status of agent load: number of connections are in operation, as well as the load on each accession. If the load is possible to realize the requested HC, he sends communication message about the successful implementation of the requested HC.

7. If after the implementation of the hydrocarbon flow of active power in at least one of controlled ties in the subsystem exceeds the maximum, then, given these changes, go to step 1. Otherwise we finish execution of the algorithm.

At any step of the algorithm communications agent terminates the execution of the algorithm provided that the flow of active power for communications controlled by the agent, less than maximum allowable. At any step of the algorithm is efficient communications agent must respond to requests from other agents of communication on the current flow of active power for communications controlled by the agent.

Comparative analysis of the proposed method with the well-known not identified the ways of containing the characteristics identical or equivalent distinctive feature of the proposed method. So, we can make a conclusion on compliance of the proposed solutions to the criterion of «substantial differences».

Brief description of drawings

Figure 1 shows a scheme of energy, which reveals the essence of the proposed method.

Figure 2 is a block diagram of the device, which can be described method.

Realization of the invention

The implementation of how it is possible with the help of the device of the block diagram shown in figure 2. The schematic image reflects the features of the local device, PAH, which implements the functions of agent communication. Device PAH consists of the block of measurement collection of 6 and a unit of issue of control actions 16. The unit of measurement collection 6 receives a set of information about the current network topology and information about the current state of load nodes that have the greatest impact on the download links, controlled by the agent. Information about the current boot controlled connection is passed to the block 7, which is checked Boolean condition of having an overload. In case of absence of overload control is again transferred in block 6 in the case of overload control is passed to block 8. In block 8 based on the information about the current network topology select the table of coefficients of influence, after which the control is passed to block 9. In block 9 based on current information available to the agent connection is made preliminary calculation of the optimum volume of the load. In case of absence of the decision of problems of linear optimization, formed in block 9, the algorithm terminates and control is passed to block 10, which does not refer to a given algorithm and implements other algorithms of PAH. In case of existence of the solution of the problem, formed in block 9, the resulting vector control actions (CA) is passed to the block 11. In block 11 is forming a list of all the unique relations with negative coefficients of influence. Then this list is passed to the block 12. Block 12 performs the request for information on the current flow of active power relations, included in the list of unique relationships with negative coefficients, after receiving this information is transmitted in block 13. In block 13 happens re-calculation of optimum volume of loads with account of the effects of HC on communication with negative coefficients of influence. In case of absence of the decision of problems of linear optimization, formed in block 13, the algorithm terminates and control is passed to block 10, which does not refer to a given algorithm and implements other algorithms of PAH. In case of existence of the solution of the problem, formed in block 13, the resulting vector HC is passed to block 14, which generates a list of links that were previously not included in the list of unique relationships with negative coefficients. The generated list is passed to the block 15. Block 15 is the verification logic. If the list is passed to the block 15, is not empty, then there is a transfer of the current vector of hydrocarbons in block 11, otherwise vector HC is passed to block 16. Block 16 is responsible for sending information messages to the agents of the load with the request of control realization. The requested agents loads can answer, or a rejection, a reply message will contain information about the current status of agent load answering waiver or consent with agents, who answered consent, to implement the requested HC. After receiving the responses from all the requested load agents or after a fixed timeout, which depends on the level of current overload controlled line, time of the maximum possible waiting for a response etc., communications agent passes control in block 6. The transfer of control in block 6 may also be performed from any part of the algorithm on the external interrupt in case of removal of overload-controlled communication.