Method, device and electronic equipment for coordinated control of multiple transformers

Abstract

This invention discloses a method, apparatus, and electronic device for the coordinated control of multiple transformers. The method includes: acquiring multiple control cycles; controlling the multiple transformers within their respective control cycles according to the execution order of the multiple control cycles, until the multiple control cycles are completed: acquiring operating data corresponding to each of the multiple transformers; determining grid state parameters based on the multiple operating data; determining the operation probability distribution corresponding to each of the multiple transformers based on the grid state parameters; determining the target control operation corresponding to each of the multiple transformers based on the multiple operation probability distributions; and controlling the multiple transformers based on the multiple target control operations. This invention solves the technical problem in related technologies where the coordinated control of multiple transformers in a distribution network results in poor coordinated control performance, leading to unstable operation of the distribution network.

Description

Technical Field

[0001] This invention relates to the field of power systems, and more specifically, to a method, apparatus, and electronic device for the coordinated control of multiple transformers. Background Technology

[0002] In related technologies, power distribution networks utilize multiple transformers for voltage regulation, load distribution, and power transmission to adapt to the electricity demands of different areas and improve power supply coverage and capacity. However, these technologies suffer from poor coordination between multiple transformers in the distribution network, leading to instability in network operation.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, and electronic device for the coordinated control of multiple transformers, in order to at least solve the technical problem in the related art where the coordinated control of multiple transformers in a distribution network is ineffective, leading to unstable operation of the distribution network.

[0005] According to one aspect of the present invention, a method for coordinated control of multiple transformers is provided, comprising: acquiring multiple control cycles of a distribution network, wherein the distribution network includes multiple transformers; controlling the multiple transformers of the distribution network in the corresponding control cycle according to the execution order of the multiple control cycles, in the following manner, until the multiple control cycles are completed: acquiring operating data corresponding to the multiple transformers at the start time of the corresponding control cycle; determining grid state parameters corresponding to the distribution network in the corresponding control cycle based on the operating data corresponding to the multiple transformers; determining operation probability distributions corresponding to the multiple transformers in the corresponding control cycle based on the grid state parameters; determining target control operations corresponding to the multiple transformers in the corresponding control cycle based on the operation probability distributions corresponding to the multiple transformers; and controlling the multiple transformers of the distribution network in the corresponding control cycle based on the target control operations corresponding to the multiple transformers.

[0006] Optionally, determining the target control operation corresponding to each of the multiple transformers within the corresponding control period based on the operation probability distributions corresponding to each of the multiple transformers includes: determining the power parameters corresponding to each of the multiple transformers and the operating constraint parameters corresponding to the distribution network at the start time of the corresponding control period; determining the power deviation characteristics corresponding to each of the multiple transformers based on the power parameters corresponding to each of the multiple transformers; determining the operating deviation characteristics corresponding to the distribution network based on the operating constraint parameters; and determining the target control operation corresponding to each of the multiple transformers within the corresponding control period based on the power deviation characteristics, the operating deviation characteristics, and the operation probability distributions corresponding to each of the multiple transformers.

[0007] Optionally, determining the operating deviation characteristics corresponding to the distribution network based on the operating constraint parameters includes: when the operating constraint parameters include current constraint parameters, voltage constraint parameters, and load constraint parameters, determining the operating deviation characteristics corresponding to the distribution network based on the current constraint parameters, the voltage constraint parameters, and the load constraints.

[0008] Optionally, determining the target control operation corresponding to each of the multiple transformers within the corresponding control period based on the operation probability distributions corresponding to each of the multiple transformers includes: after determining the target control operation corresponding to the first control period among the multiple control periods, determining the target control operation corresponding to the corresponding target period in the following manner according to the execution order of the multiple target periods: determining the preceding control period corresponding to the corresponding target period, wherein the multiple target periods are control periods after the first control period among the multiple control periods; determining the preceding control operation corresponding to each of the multiple transformers within the preceding control period; determining the operation deviation characteristics corresponding to each of the multiple transformers based on the preceding control operation corresponding to each of the multiple transformers; and determining the target control operation corresponding to the corresponding target period based on the operation probability distributions and operation deviation characteristics corresponding to each of the multiple transformers.

[0009] Optionally, determining the operation probability distribution corresponding to each of the plurality of transformers within the corresponding control period based on the power grid state parameters includes: for any one of the plurality of transformers, determining the operation probability distribution corresponding to that transformer within the corresponding control period using the following method: determining the tap control constraint and multiple voltage control taps corresponding to that transformer, wherein the tap control constraint includes a first tap constraint and a second tap constraint, the first tap constraint constraining the control operation of the first control tap, the second tap constraint constraining the control operation of the second control tap, the first control tap being the largest voltage control tap among the plurality of voltage control taps, and the second control tap being the smallest voltage control tap among the plurality of voltage control taps; determining multiple candidate control operations corresponding to each of the plurality of voltage control taps, wherein the multiple candidate control operations are used to adjust the corresponding voltage control taps at different taps; and determining the operation probability distribution corresponding to that transformer within the corresponding control period based on the power grid state parameters, the tap control constraint, and the multiple candidate control operations corresponding to each of the plurality of voltage control taps.

[0010] Optionally, determining multiple voltage control levels corresponding to any one of the transformers includes: determining equipment parameters corresponding to any one of the transformers, wherein the equipment parameters include rated capacity, winding connection method, and number of circuit terminals; determining voltage compensation parameters corresponding to each of the transformers based on the equipment parameters, wherein the voltage compensation parameters are used to reflect the voltage compensation characteristics of the transformer to the distribution network; and determining multiple voltage control levels corresponding to any one of the transformers based on the voltage compensation parameters.

[0011] Optionally, before obtaining multiple control cycles of the distribution network, the process includes: determining scenario parameters and operating requirement parameters corresponding to the distribution network; and determining multiple control cycles corresponding to the distribution network based on the scenario parameters and operating requirement parameters.

[0012] According to one aspect of the present invention, a coordinated control device for multiple transformers is provided, comprising:

[0013] An acquisition module is used to acquire multiple control cycles of a distribution network, wherein the distribution network includes multiple transformers; a determination module is used to control the multiple transformers of the distribution network in the corresponding control cycle according to the execution order of the multiple control cycles, in the following manner, until the multiple control cycles are completed: at the start time of the corresponding control cycle, acquire the operating data corresponding to each of the multiple transformers; based on the operating data corresponding to each of the multiple transformers, determine the grid state parameters corresponding to the distribution network in the corresponding control cycle; based on the grid state parameters, determine the operation probability distribution corresponding to each of the multiple transformers in the corresponding control cycle; based on the operation probability distribution corresponding to each of the multiple transformers, determine the target control operation corresponding to each of the multiple transformers in the corresponding control cycle; based on the target control operation corresponding to each of the multiple transformers, control the multiple transformers of the distribution network in the corresponding control cycle.

[0014] According to one aspect of the present invention, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions to implement the cooperative control method for multiple transformers as described in any of the preceding claims.

[0015] According to one aspect of the present invention, a computer-readable storage medium is provided, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the cooperative control method for multiple transformers as described above.

[0016] In this embodiment of the invention, multiple control cycles of a distribution network are obtained, wherein the distribution network includes multiple transformers. According to the execution order of the multiple control cycles, the multiple transformers of the distribution network are controlled in the corresponding control cycles in the following manner until the multiple control cycles are completed: at the start time of the corresponding control cycle, operating data corresponding to each of the multiple transformers is obtained; based on the operating data corresponding to each of the multiple transformers, the grid state parameters corresponding to the distribution network within the corresponding control cycle are determined; based on the grid state parameters, the operation probability distribution corresponding to each of the multiple transformers within the corresponding control cycle is determined; based on the operation probability distribution corresponding to each of the multiple transformers, the target control operation corresponding to each of the multiple transformers within the corresponding control cycle is determined; based on the target control operation corresponding to each of the multiple transformers, the multiple transformers of the distribution network are controlled within the corresponding control cycle. By acquiring multiple control cycles of a distribution network containing multiple transformers, and collecting the operating data of each transformer at the start time of each control cycle in the execution sequence, the real-time operating status of each transformer can be comprehensively captured, providing accurate data support for subsequent decision-making. Based on this operating data, the grid state parameters of the distribution network can be determined, and the operational correlation information of multiple transformers can be integrated to form a unified reference reflecting the overall network operation status. Determining the operation probability distribution for each transformer based on the grid state parameters ensures that the control decisions of each transformer are aligned with the overall network operation needs, avoiding the blindness of decisions made by a single device. Combining the operation probability distribution to clarify the target control operations of each transformer ensures that the control actions of each device are reasonable and meet the requirements of coordination. Synchronously controlling multiple transformers based on the target control operations within each control cycle and continuing until all cycles are completed enables dynamic coordinated adjustment of multiple transformers, avoiding mutual interference caused by independent control, ensuring the stability of the distribution network operation, and thus solving the technical problem in related technologies where the coordinated control of multiple transformers in a distribution network results in poor coordinated control effects and unstable distribution network operation. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0018] Figure 1 This is a flowchart of a method for coordinated control of multiple transformers according to an embodiment of the present invention;

[0019] Figure 2 This is a schematic flowchart of a collaborative control method for multiple transformers in an optional embodiment of the present invention.

[0020] Figure 3This is a structural block diagram of a collaborative control device for multiple transformers according to an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Example 1

[0024] According to an embodiment of the present invention, an embodiment of a collaborative control method for multiple transformers is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0025] Figure 1 This is a flowchart of a coordinated control method for multiple transformers according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0026] S102, obtain multiple control cycles of the distribution network, wherein the distribution network includes multiple transformers;

[0027] This involves the distribution network, which is a power grid system containing multiple transformers and other electrical equipment. It is responsible for transmitting electrical energy from the generation end to the consumption end, directly connecting with the end-user's electricity demand.

[0028] This involves multiple control cycles, which are used to achieve dynamic and coordinated control of multiple transformers in the distribution network. The continuous operation process of the distribution network is divided into a series of discrete and ordered time intervals, each of which is an independent control cycle. Multiple control cycles are executed in a preset order until the preset control task is completed.

[0029] This involves multiple transformers, which are devices in the power distribution network that perform voltage transformation, power transmission and distribution functions. For example, these multiple transformers can be multiple thyristor-controlled hybrid transformers (TCHTs), that is, these multiple transformers can be multiple TCHTs.

[0030] By dividing the continuous operation of the power distribution network into multiple discrete and ordered control cycles, it is possible to provide fine-grained time-dimensional settings for the coordinated control of multiple transformers, ensuring that the operating status of the transformers can be accurately adjusted within each control cycle, and avoiding insufficient control accuracy due to complex dynamic changes during continuous operation.

[0031] S104, following the execution sequence of multiple control cycles, control multiple transformers in the distribution network within their respective control cycles in the following manner until the multiple control cycles are completed: at the start time of the corresponding control cycle, acquire the operating data corresponding to each of the multiple transformers; based on the operating data corresponding to each of the multiple transformers, determine the grid state parameters corresponding to the distribution network within the corresponding control cycle; based on the grid state parameters, determine the operation probability distribution corresponding to each of the multiple transformers within the corresponding control cycle; based on the operation probability distribution corresponding to each of the multiple transformers, determine the target control operation corresponding to each of the multiple transformers within the corresponding control cycle; based on the target control operation corresponding to each of the multiple transformers, control the multiple transformers in the distribution network within the corresponding control cycle.

[0032] This involves the execution order of multiple control cycles, which follows a pre-defined logical sequence (e.g., in chronological order, starting with the first control cycle and executing subsequent control cycles sequentially until all control cycles are completed). This ensures that the control processes within each control cycle are connected in an orderly manner, avoiding temporal confusion or overlap.

[0033] This involves the start time of the corresponding control cycle. The start time of the corresponding control cycle is the specific time point at the beginning of each control cycle, which is the starting point for a series of control operations such as data acquisition, status analysis, and decision generation within that control cycle.

[0034] This includes operational data, which is collected at the start of the control cycle and reflects various data on the operating status of multiple transformers, including the current tap position of each transformer, actual active and reactive power transmission, node voltage, current, and line load rate.

[0035] This involves power grid state parameters, which are parameters used to comprehensively reflect the overall operating state of the distribution network within the corresponding control cycle and can be represented by the current system state vector. The power grid state parameters are used as the current system state vector. As an example, this current system state vector It can be constructed based on the current tap position of multiple transformers, actual active and reactive power transmission, node voltage, current and line load rate and other operating data, combined with the power command issued by the dispatcher.

[0036] This involves an operation probability distribution, which reflects the probability of various control operations that each transformer may perform within a corresponding control cycle. This operation probability distribution can be an action probability distribution. Taking the action probability distribution as an example, for an agent corresponding to multiple transformers, the current system state vector... By inputting the policy network corresponding to each transformer (such as TCHT) in the multi-agent reinforcement learning control module, the action probability distribution (i.e., operation probability distribution) of each agent is obtained.

[0037] This involves target control operations, which are the specific control actions that each transformer should perform (such as shifting up, shifting down, or maintaining the current shift). Specifically, the target control operation for each transformer can be determined by using the operation probability distributions corresponding to multiple transformers, combined with an action mask, to mask out the restricted actions in the operation probability distributions, and then using a maximum probability selection or a sampling method with a small amount of randomness.

[0038] This involves controlling multiple transformers in a distribution network. This control involves translating the determined target control operations for each transformer into specific tap adjustment commands within a corresponding control cycle and sending these commands to field devices. The devices then execute tap switching and other operations to achieve synchronous adjustment of multiple transformers, thereby reconstructing the power flow of the distribution network. This allows the distribution network to reach a new stable operating state after a brief transient process, ensuring the overall stability and coordination of the distribution network operation. Specifically, a joint action vector can be determined by analyzing the target control operations corresponding to each of the multiple transformers. , joint action vector The specific gear adjustment command is converted into a specific gear position adjustment command and sent to each transformer. The corresponding system device of the distribution network then performs the gear position switching to enable the power flow of the distribution network to reach a new steady state after a brief transient process.

[0039] By executing multiple control cycles in a pre-set sequence, transformer operating data is collected at the beginning of each cycle, and the power grid status parameters are determined accordingly. This allows for real-time monitoring of the overall operation of the distribution network. Furthermore, the operation probability distribution of each transformer is calculated based on the power grid status parameters to ensure that control decisions meet the current power grid requirements. Specific target control operations are then determined based on the operation probability distribution and converted into gear adjustment commands for execution. This enables synchronous and precise adjustment of multiple transformers, avoiding mutual interference caused by independent control and ensuring that the distribution network reaches a new stable operating state after each control cycle.

[0040] Through steps S102-S104 above, multiple control cycles of the distribution network are obtained, wherein the distribution network includes multiple transformers. Following the execution order of the multiple control cycles, the multiple transformers of the distribution network are controlled within their respective control cycles in the following manner until the multiple control cycles are completed: at the start time of the corresponding control cycle, the operating data corresponding to each of the multiple transformers is obtained; based on the operating data corresponding to each of the multiple transformers, the grid state parameters corresponding to the distribution network within the corresponding control cycle are determined; based on the grid state parameters, the operation probability distribution corresponding to each of the multiple transformers within the corresponding control cycle is determined; based on the operation probability distribution corresponding to each of the multiple transformers, the target control operation corresponding to each of the multiple transformers within the corresponding control cycle is determined; based on the target control operation corresponding to each of the multiple transformers, the multiple transformers of the distribution network are controlled within the corresponding control cycle. By acquiring multiple control cycles of a distribution network containing multiple transformers, and collecting the operating data of each transformer at the start time of each control cycle in the execution sequence, the real-time operating status of each transformer can be comprehensively captured, providing accurate data support for subsequent decision-making. Based on this operating data, the grid state parameters of the distribution network can be determined, and the operational correlation information of multiple transformers can be integrated to form a unified reference reflecting the overall network operation status. Determining the operation probability distribution for each transformer based on the grid state parameters ensures that the control decisions of each transformer are aligned with the overall network operation needs, avoiding the blindness of decisions made by a single device. Combining the operation probability distribution to clarify the target control operations of each transformer ensures that the control actions of each device are reasonable and meet the requirements of coordination. Synchronously controlling multiple transformers based on the target control operations within each control cycle and continuing until all cycles are completed enables dynamic coordinated adjustment of multiple transformers, avoiding mutual interference caused by independent control, ensuring the stability of the distribution network operation, and thus solving the technical problem in related technologies where the coordinated control of multiple transformers in a distribution network results in poor coordinated control effects and unstable distribution network operation.

[0041] As an optional embodiment, based on the operation probability distributions corresponding to the multiple transformers respectively, the target control operation corresponding to the multiple transformers within the corresponding control period is determined, including: at the start time of the corresponding control period, determining the power parameters corresponding to the multiple transformers and the operating constraint parameters corresponding to the distribution network respectively; determining the power deviation characteristics corresponding to the multiple transformers based on the power parameters corresponding to the multiple transformers respectively; determining the operating deviation characteristics corresponding to the distribution network based on the operating constraint parameters; and determining the target control operation corresponding to the multiple transformers within the corresponding control period based on the power deviation characteristics, the operating deviation characteristics, and the operation probability distributions corresponding to the multiple transformers respectively.

[0042] This involves power parameters, which reflect the power characteristics of multiple transformers at the start of a corresponding control cycle. These parameters include the actual active power, actual reactive power, target active power, and target reactive power for each transformer. For example, taking multiple transformers as multiple TCHTs, the power parameters include the actual active and reactive power transmission power of each TCHT and the target active and reactive power commands issued by the upper-level control system for each TCHT.

[0043] This involves operational constraint parameters, which are electrical parameters (i.e., quantities related to safety constraints) that are associated with the physical and safety limitations during the operation of the distribution network. These parameters include the device current, phase voltage, node voltage, and line load rate of the distribution network.

[0044] This involves power deviation characteristics, which are features used to reflect the difference between the actual power transmission status of each transformer and the target power requirement. These characteristics can be represented by power deviation sub-items. The formula is as follows:

[0045]

[0046] in, The number of multiple transformers; Let be the active power transmitted by the i-th transformer at time t; Let be the target active power transmitted by the i-th transformer; The reactive power transmission power of the i-th transformer at time t; Let be the target reactive power transmission power of the i-th transformer.

[0047] This involves operational deviation characteristics, which are features used to reflect the degree of fit between the current operating state of the distribution network and the safety operating constraints. These characteristics can be represented by operational safety constraint penalty terms. The formula is:

[0048]

[0049] in, To implement safety constraint penalties; This is an indicator of whether the k-th constraint is triggered at time t, or a quantized value of the corresponding degree of violation. is the weight coefficient of the k-th constraint.

[0050] By synchronously acquiring transformer power parameters and distribution network operation constraint parameters at the start of the control cycle, the real-time operating status of equipment and system safety boundary conditions can be comprehensively obtained. Then, based on the power parameters, deviation characteristics are calculated, and the degree of operational deviation is evaluated according to the constraint parameters. The gap between the current state and the target requirements can be quantitatively analyzed. Finally, when determining the target control operation by combining the operation probability distribution, both power tracking accuracy and system safety constraints can be taken into account. This ensures that the multi-transformer collaborative control meets the scheduling requirements while avoiding the risk of exceeding limits, thus achieving the dual goals of safe and stable operation and optimized regulation of the distribution network.

[0051] As an optional embodiment, determining the operating deviation characteristics corresponding to the distribution network based on operating constraint parameters includes: when the operating constraint parameters include current constraint parameters, voltage constraint parameters, and load constraint parameters, determining the operating deviation characteristics corresponding to the distribution network based on the current constraint parameters, voltage constraint parameters, and load constraints.

[0052] This involves current constraint parameters, which are electrical parameters associated with current safety limits during the operation of the distribution network, such as the device current corresponding to the distribution network.

[0053] This involves voltage constraint parameters, which are electrical parameters associated with voltage safety limits during the operation of the distribution network, such as the phase voltage and node voltage of the distribution network.

[0054] This involves load constraint parameters, which are power parameters associated with load safety limitations during the operation of the distribution network, such as the line load rate of the distribution network.

[0055] By integrating current constraint parameters, voltage constraint parameters, and load constraint parameters, the deviation of each key link of the distribution network (such as equipment current, node voltage, and line load) from the safety limit can be fully quantified. This allows for an accurate assessment of whether the system's operating status meets safety requirements, avoiding local optimization traps caused by single parameter evaluation. It ensures that the operating deviation characteristics truly reflect the overall safety risk of the system, providing a reliable basis for subsequent formulation of target control operations that balance efficiency and safety.

[0056] As an optional embodiment, based on the operation probability distributions corresponding to the multiple transformers, the target control operation corresponding to each of the multiple transformers within the corresponding control period is determined. This includes: after determining the target control operation corresponding to the first control period among the multiple control periods, the target control operation corresponding to the corresponding target period is determined in the following manner according to the execution order of the multiple target periods: determining the preceding control period corresponding to the corresponding target period, wherein the multiple target periods are control periods following the first control period among the multiple control periods; determining the preceding control operation corresponding to each of the multiple transformers within the preceding control period; determining the operation deviation characteristics corresponding to the multiple transformers based on the preceding control operations corresponding to each of the multiple transformers; and determining the target control operation corresponding to the corresponding target period based on the operation probability distributions and operation deviation characteristics corresponding to each of the multiple transformers.

[0057] This involves the first control cycle, which is the first control cycle to be started in the preset execution order among multiple control cycles, and is the initial execution stage of the entire collaborative control process.

[0058] This involves multiple target cycles, which are all control cycles following the first control cycle. The target control operations corresponding to these multiple target cycles can be dynamically adjusted based on the control situation of the previous control cycle to ensure the continuity and adaptability of the control strategy.

[0059] This involves the preceding control cycle, which is the control cycle that precedes the corresponding target cycle in the execution sequence.

[0060] This involves the previous control operation, which is the specific control action (such as shifting up, shifting down, or maintaining the current shift) determined and executed for each transformer in the previous control cycle. This is key information reflecting the adjustment status of each transformer in the previous control cycle.

[0061] This involves operational deviation characteristics, which are used to reflect the difference in transformer control operation between the corresponding target cycle and the previous control cycle, in order to quantify the degree of deviation between the previous control operation and the optimal control direction of the corresponding target cycle. These characteristics can be represented by a penalty term related to the number of actions or gear changes.

[0062] Specifically, based on the operation probability distribution and operation deviation characteristics corresponding to multiple transformers, the target control operation corresponding to the target period is determined as follows: based on the power deviation characteristics, operation deviation characteristics and operation deviation characteristics, the fusion deviation characteristics are determined; based on the fusion deviation characteristics and the operation probability distribution corresponding to multiple transformers, the target control operation corresponding to multiple transformers within the corresponding control period is determined.

[0063] Among them, fusion bias characteristics It can be expressed in the form of a reward function, with the formula:

[0064]

[0065] in, Penalties are related to the number of actions or gear changes.

[0066] After determining the target control operation for the first control cycle, the previous control cycle corresponding to each target cycle is identified according to the execution order of multiple target cycles. Then, the previous control operation, such as adjusting the tap position of each transformer in the previous cycle, is obtained. The deviation quantification result between the previous control operation and the current optimal direction can be used for dynamic correction to avoid frequent reverse adjustment or over-adjustment. This ensures that the control actions of each target cycle are connected and adapted to the previous cycle, and meet the requirements of the real-time power grid status. Ultimately, the dynamic optimization and continuous coordination of transformer control strategies in multiple control cycles are achieved, ensuring the long-term stable operation of the distribution network.

[0067] As an optional embodiment, based on the grid state parameters, the operation probability distribution corresponding to multiple transformers within the corresponding control period is determined, including: for any one of the multiple transformers, the operation probability distribution corresponding to that transformer within the corresponding control period is determined in the following manner: determining the tap control constraint and multiple voltage control taps corresponding to that transformer, wherein the tap control constraint includes a first tap constraint and a second tap constraint, the first tap constraint is used to constrain the control operation of the first control tap, the second tap constraint is used to constrain the control operation of the second control tap, the first control tap is the largest voltage control tap among the multiple voltage control taps, and the second control tap is the smallest voltage control tap among the multiple voltage control taps; determining multiple candidate control operations corresponding to each of the multiple voltage control taps, wherein the corresponding multiple candidate control operations are used to perform different tap adjustments on the corresponding voltage control taps; and determining the operation probability distribution corresponding to any one transformer within the corresponding control period based on the grid state parameters, the tap control constraint, and the multiple candidate control operations corresponding to each of the multiple voltage control taps.

[0068] This involves any one transformer, which is any one of multiple transformers.

[0069] This involves gear position control constraints, which are the limitations on performing gear position control operations for any transformer. These constraints regulate the feasible range of gear position adjustments, preventing equipment damage and abnormal grid operation caused by operations exceeding physical limits or safety boundaries. The constraints include first and second gear position constraints. These constraints can be implemented through an action masking mechanism, which automatically masks actions that would cause the gear position to exceed the limits when the transformer is in certain extreme gear positions (such as the first and second control gear positions), retaining only gear position changes that maintain the current gear position or the feasible direction.

[0070] This involves multiple voltage control levels, which are the set of all switchable discrete voltage regulation levels for any transformer.

[0071] This involves a first-level constraint, which is used to limit the control operation of the first control level, so as to prohibit the execution of adjustment actions from this level that would cause the voltage to exceed the maximum compensation range, and ensure that the operation of the first control level meets the safe operation conditions of the equipment.

[0072] This involves a second-level constraint, which is used to limit the control operation of the second control level to prohibit the execution of adjustment actions that would cause the voltage to fall below the minimum compensation range from this level, thus ensuring that the operation of the second control level meets the safe operating conditions of the equipment.

[0073] This involves the first control level, which is the level with the strongest voltage compensation capability and the largest corresponding compensation voltage among multiple voltage control levels. It is the upper limit level for transformer voltage regulation.

[0074] This involves the second control level, which is the level with the weakest voltage compensation capability and the smallest corresponding compensation voltage among multiple voltage control levels. It is the lower limit level for transformer voltage regulation.

[0075] This involves multiple selectable control operations, which are control operations used to adjust multiple voltage control levels in different directions or magnitudes, including adjusting up to an adjacent level, adjusting down to an adjacent level, and keeping the current level unchanged.

[0076] By defining the control constraints and multiple voltage control levels, the safety boundaries and feasible range of transformer level adjustment can be delineated, preventing adjustment actions from exceeding the physical limits of the equipment. Furthermore, by identifying multiple candidate control operations such as upward adjustment, downward adjustment, and holding corresponding to each voltage control level, specific action options can be provided for subsequent probability calculations. Finally, by combining real-time grid state parameters to perform probability assessments of each operation, the generated transformer operation probability distribution can be ensured to conform to both the physical safety limits of the equipment and the current grid operation requirements, avoiding equipment damage or grid anomalies caused by exceeding limits. Simultaneously, it provides a realistic and optimized decision-making basis for subsequent coordinated control.

[0077] As an optional embodiment, determining multiple voltage control levels corresponding to any one transformer includes: determining equipment parameters corresponding to any one transformer, wherein the equipment parameters include rated capacity, winding connection method and number of circuit terminals; determining voltage compensation parameters corresponding to each transformer based on the equipment parameters, wherein the voltage compensation parameters are used to reflect the voltage compensation characteristics of any transformer to the distribution network; and determining multiple voltage control levels corresponding to any one transformer based on the voltage compensation parameters.

[0078] This involves equipment parameters, which are parameters used to reflect the attribute characteristics of the corresponding transformer, including rated capacity, winding connection method and number of circuit joints. The number of circuit joints can be expressed by the number of taps.

[0079] This includes the rated capacity, which is the maximum apparent power that a transformer is allowed to output safely and continuously under specified rated voltage and frequency conditions.

[0080] This involves the winding connection method, which is the connection form of the primary (input end) and secondary (output end) windings of the transformer. Different connection methods will affect the voltage phase, turns ratio and output voltage stability of the transformer, thereby affecting the implementation method and effect of voltage compensation.

[0081] This involves the number of circuit connectors, which is the total number of taps installed on the transformer windings. These connectors can be switched to change the number of winding turns, thereby adjusting the transformer output voltage. This is the physical basis for realizing voltage level regulation, and its number affects the upper limit of the number of voltage control levels that the transformer can be divided into.

[0082] This involves voltage compensation parameters, which are used to quantify the transformer's ability to regulate and compensate for the voltage of the distribution network.

[0083] Specifically, based on the voltage compensation parameters, multiple voltage control levels corresponding to any one transformer are determined. This also includes determining the upper and lower limits of the levels (such as the upper and lower limits of the levels allowed by the device), the level change step size (such as the maximum allowable level change step size), and the operating frequency (such as the maximum allowable operating frequency) for each of the multiple transformers, in order to limit the feasible range of the operation. Based on the upper and lower limits of the levels, the level change step size, the operating frequency, and other operating constraints corresponding to each of the multiple transformers, and in combination with the voltage compensation parameters, multiple voltage control levels corresponding to any one transformer are determined.

[0084] By determining the transformer equipment parameters, we can clarify its physical characteristics and the hardware basis for voltage regulation. Then, based on these parameters, we can calculate quantitative indicators (voltage compensation parameters) that reflect the voltage compensation capability, ensuring an accurate characterization of the transformer's regulation characteristics. Dividing the voltage compensation parameters into multiple voltage control levels allows the level settings to not only conform to the actual regulation capability of the equipment but also meet the accuracy requirements of the distribution network voltage regulation. This avoids insufficient regulation or exceeding limits due to unreasonable level design, ultimately achieving precise and adaptable transformer voltage regulation.

[0085] As an optional embodiment, before obtaining multiple control cycles of the distribution network, the process includes: determining the scenario parameters and operating requirement parameters corresponding to the distribution network; and determining multiple control cycles corresponding to the distribution network based on the scenario parameters and operating requirement parameters.

[0086] This involves scenario parameters, which are operating condition parameters related to the external environment, network structure, and load characteristics of the distribution network, used to reflect the operating conditions of the distribution network. These scenario parameters include the distribution network topology, time parameters (such as day and night, season), meteorological conditions (such as temperature, humidity, and light intensity), load type distribution, etc., used to characterize the spatiotemporal background and boundary conditions of the distribution network operation.

[0087] This includes operational requirement parameters, which are parameters related to the core objectives and performance indicators that the distribution network needs to meet during operation. These parameters clarify the control direction and requirements of the distribution network, including but not limited to voltage quality standards (such as the allowable range of voltage deviation), power transmission accuracy requirements, power supply reliability indicators, equipment operating efficiency targets, and load balancing requirements.

[0088] By determining the scenario parameters of the distribution network, we can characterize its spatiotemporal background and boundary conditions. Combined with the operational demand parameters, we can clarify the control priorities and constraints under different operating conditions. Based on these parameters, we can divide the control cycle into multiple control cycles, which can ensure that the control strategy of each cycle is accurately matched with the current scenario characteristics and operational requirements. This avoids the situation where the control cycle is set too densely, leading to excessive equipment adjustment and increased losses, or set too sparsely, failing to respond to changes in the grid status in a timely manner. This ensures that the subsequent coordinated control of multiple transformers is more targeted and effective.

[0089] Based on the above embodiments and optional embodiments, an optional implementation method is provided, which is described in detail below.

[0090] In related technologies, power distribution networks utilize multiple transformers for voltage regulation, load distribution, and power transmission to adapt to the electricity demands of different areas and improve power supply coverage and capacity. However, these technologies suffer from poor coordination of multiple transformers in the distribution network, leading to instability in network operation.

[0091] There is currently no effective solution to the above problems.

[0092] In view of this, an optional embodiment of the present invention provides a method for coordinated control of multiple transformers, which can effectively solve the above-mentioned technical problems.

[0093] Figure 2 This is a schematic flowchart of a collaborative control method for multiple transformers in an optional embodiment of the present invention, as shown below. Figure 2 As shown, multiple transformers are represented as multiple TCHTs as an example, and the following is a detailed description.

[0094] Step A: Define the motion space based on the TCHT device parameters;

[0095] The action space is the set of all operable gear adjustment behaviors of each TCHT within its operating constraints. The multi-agent reinforcement learning controller is directly trained online in the actual distribution network operating environment. To ensure that the training and control process meets the physical and safety constraints of the device, the corresponding discrete action space (i.e., the action space) must first be defined based on the structural parameters and operating constraints of each hybrid phase-shifting transformer (TCHT).

[0096] Specifically, step A includes:

[0097] A1, Acquisition device information and operational constraints;

[0098] For each TCHT connected to the distribution network, obtain or register its basic parameters such as rated capacity, winding connection method, number of taps and corresponding compensation voltage vector set; obtain the device's allowed upper and lower limits of tap positions, maximum allowable tap position change step size, maximum allowable operating frequency and other operating constraints to limit the feasible range of operation.

[0099] A2, based on the set of compensation vectors, divides feasible gears;

[0100] Based on the actual achievable set of compensation voltage vectors for each TCHT, it is discretized into a finite number of gears, each gear corresponding to a unique three-phase or single-phase compensation voltage vector code; gears located near physical limits are marked, and operations that would cause out-of-bounds operations are prohibited from being issued from these gears in subsequent action spaces.

[0101] A3 is used to construct a discrete set of actions for reinforcement learning agents.

[0102] Using "gear shifting mode" as the basic control action, including shifting to one of several adjacent compensation vector gears and keeping the current gear unchanged, all the above operable gear shifting modes are encoded into a finite number of discrete actions to form the corresponding TCHT action set.

[0103] For each set of actions, an action masking mechanism (i.e., gear control constraint) is established that matches the upper and lower limits of the TCHT device gear: when the TCHT is in certain extreme gears, actions that would cause the gear to go out of bounds are automatically masked, and only gear changing actions that maintain the gear or the feasible direction are retained.

[0104] A4, constructing a multi-agent joint action space;

[0105] Each TCHT's defined set of discrete actions is considered as an agent's local action space. In each control cycle, all devices (i.e., TCHTs) simultaneously select actions, and their combination constitutes a joint action vector of multiple agents, which is used to drive the actual system to complete a power flow adjustment. During the online training and operation phases, the joint actions are independently generated or sampled by each agent's policy network based on the real-time state, and the action mask ensures that the device gear boundaries and basic operating constraints are not violated.

[0106] Through the above steps, without constructing a complex simulation model, the action space of multiple TCHT agents is defined based solely on the device's (i.e., TCHT) own parameters and operational constraints, providing an analytical basis for subsequent online closed-loop control and strategy optimization.

[0107] Step B: Define the reward function;

[0108] In online training scenarios, reinforcement learning agents (each agent is a corresponding TCHT control module) interact directly with the actual power distribution network. They need to use real-time measurement data to construct reward functions to characterize the comprehensive impact of a gear adjustment on the system's operational objectives and safety constraints.

[0109] Specifically, step B includes:

[0110] B1, collects actual system usage;

[0111] After each control cycle ends, real-time quantities related to the control objective are collected, including but not limited to the actual active and reactive power transmission power of each TCHT, the target active and reactive power commands issued by the upper control system to each TCHT, and quantities related to safety constraints, such as device current, phase voltage, node voltage, and line load rate.

[0112] B2, constructing the power deviation term;

[0113] The deviation between the actual transmitted complex power of each TCHT station and the target power command is used as the basis for the reward, and a power deviation sub-item is defined as follows:

[0114]

[0115] in, For the power deviation sub-item, this item is negative, so that the smaller the power deviation, the larger the reward value, guiding the agent to quickly approach the target power command within a finite number of steps; The number of multiple TCHTs; Let be the active power transmitted at time t for the i-th TCHT. Let be the target active power transmitted by the i-th TCHT; The reactive power transmission power of the i-th TCHT at time t; Let be the target reactive power transmission power of the i-th TCHT.

[0116] B3, add runtime safety constraint penalty items;

[0117] When any of the following conditions are detected: TCHT device current exceeds the rated value or approaches the thermal limit; access node voltage exceeds the upper or lower limit or approaches the over-limit threshold; or the load rate of related lines or transformers exceeds the set upper limit, a penalty is added to the reward for that step. A weighting coefficient is set for each type of constraint violation. Construct operational safety constraint penalty items:

[0118]

[0119] in, To implement safety constraint penalties; This is an indicator of whether the k-th constraint is triggered at time t, or a quantized value of the corresponding degree of constraint violation.

[0120] B4, considering adjustment efficiency and motion smoothness;

[0121] To avoid equipment wear caused by frequent gear shifting and excessive movements in the actual system, additional penalty terms related to the number of movements, the magnitude of gear shifting, and the number of reverse adjustments in a short period of time are introduced into the reward. If the total gear shifting of all devices is large within a certain control cycle, a penalty is applied to the reward for that step to encourage the agent to achieve the same power tracking effect with fewer movements; a higher penalty is applied to continuous reverse adjustments in a short period of time to guide the strategy to form a smooth and monotonous adjustment trajectory.

[0122] B5, in the form of a comprehensive reward function.

[0123] The power deviation term, the operational safety constraint penalty term, and the regulation efficiency and smoothness penalty term are weighted and summed to obtain the instantaneous reward for each control cycle. (That is, fusion bias characteristics), the formula is:

[0124]

[0125] in, Penalties are related to the number of actions or gear changes; the weights of each item can be adjusted in actual operation based on system experience and performance requirements to achieve a trade-off between tracking accuracy, adjustment speed and equipment life.

[0126] Through the above reward function design, the multi-agent reinforcement learning strategy can automatically balance multiple objectives such as power command tracking, system safety constraints, and the number of equipment actions during the online interaction process of a real power distribution network.

[0127] Step C involves constructing an online closed loop of "observation-decision-execution-reward-strategy optimization," where control and strategy optimization are completed simultaneously within the same closed loop, enabling continuous online training and collaborative control.

[0128] Steps A and B abstract the actual power system into an "action-reward" structure usable by a reinforcement learning framework. Step C then constructs a continuously operating closed loop in the actual distribution network: observation and decision-making are completed within the control cycle, and the collected data is used to update the strategy between cycles, thereby continuously optimizing the collaborative control strategy of multiple TCHTs. Specifically, after completing the design of the action space and reward function, a single closed loop is constructed in the actual distribution network, completing observation, decision-making, execution, reward calculation, and strategy optimization within the same cycle, achieving integrated control and training.

[0129] C1, control cycle drive and state observation;

[0130] The operation of the distribution network is divided into discrete control cycles, with coordinated regulation of multiple TCHTs performed once per cycle. At the beginning of the control cycle, operating data such as the current tap position of each TCHT, actual active and reactive power transmission, node voltage, current, and line load rate are collected. Combined with the power commands issued by the dispatch center, the current system state vector is constructed. .

[0131] C2, multi-agent collaborative decision-making and action execution;

[0132] State Input the policy network corresponding to each TCHT in the multi-agent reinforcement learning control module to obtain the action probability distribution of each agent;

[0133] Combining the action mask from step A, restricted actions are masked, and then the local actions of each TCHT are determined using either maximum probability selection or sampling with a small amount of randomness, forming a joint action vector. ;

[0134] Joint action The commands are converted into specific gear adjustment instructions and sent to each TCHT. The field devices then execute the gear switching, and the system power flow reaches a new steady state after a brief transient process.

[0135] C3, Observation of operational results and calculation of rewards;

[0136] After the power flow reconfiguration stabilizes, the actual active and reactive power transmission power, voltage, current, and line load rate of each TCHT are collected again to form the state at the next moment. Calculate the immediate reward based on the reward function given in step B. That is, the smaller the power deviation, the higher the reward; negative rewards are given for violating safety constraints or for excessive or strenuous actions. It is stored as an experience sample in the online experience buffer.

[0137] C4, Online optimization of policy and value functions;

[0138] During system operation, a strategy optimization process is initiated at a certain frequency, forming a training loop that operates at the same or lower frequency as the control. Specifically:

[0139] Extract sample sequences from the most recent control cycles from the online experience buffer to construct the state-action-reward-next state trajectory; and apply the set discount factor. and optional generalized advantage estimation parameters The system calculates the discounted reward, temporal difference error, and advantage value at each time step to evaluate the merits of the current policy. Using multi-agent reinforcement learning algorithms such as proximal policy optimization, the system updates the policy network parameters of each agent based on the aforementioned advantage value, while ensuring that the update magnitude is limited. The centralized value network is updated with the goal of minimizing the error between the predicted value function and the actual discounted reward, thereby improving the accuracy of long-term return estimation. The training pace is dynamically adjusted according to the system's operation: the update frequency is increased when the policy is not yet stable, and decreased when the policy performance is stable and meets the operating indicators, retaining only necessary fine-tuning.

[0140] C5, a single closed-loop continuous operation mechanism.

[0141] Steps C1 to C4 continuously cycle during the operation of the distribution network: in the control loop by generate And obtained , In the training loop, based on accumulation The sample updates the parameters of the policy and value network. At any given time, the current version of the policy network serves as both the basis for control decisions and the starting point for subsequent updates, enabling continuous optimization through learning on the fly. When significant changes occur in the power grid structure or operating conditions, the policy can automatically adapt to the new operating environment because the training loop persists, eliminating the need for manual controller readjustment.

[0142] The above optional implementation methods can achieve at least the following beneficial effects:

[0143] (1) Compared with related technologies, this invention no longer relies on complete and accurate distribution network parameters and simulation models, but instead directly conducts online training of multi-agent reinforcement learning based on real-time measurement data in actual distribution network operation. By defining the action space using only device-side parameters such as rated capacity, number of taps, compensation vector set, and tap upper and lower limits, and constructing reward functions with actual measured power, voltage, current and other operating quantities, it avoids complex power flow models and parameter identification processes, and significantly reduces the threshold for engineering implementation.

[0144] (2) Compared with related technologies, this invention abstracts each TCHT as a reinforcement learning agent, and adjusts the gear of each device simultaneously in the same control cycle by the joint action of multiple agents. The power flow coupling relationship between multiple devices is implied in the state transition and reward feedback. The cooperative strategy obtained through training can automatically learn the rules of "how multiple devices cooperate to adjust", effectively avoiding the problems of power grabbing, oscillation and gear switching that occur when each device adjusts independently in the traditional way, making the overall system operation more stable and coordinated.

[0145] (3) Compared with related technologies, this invention adopts the method of "online sampling + periodic strategy update" to continuously collect state-action-reward data during the actual operation of the system, and continuously update the strategy network and value network in small steps. When the load level, distributed power output or network structure changes, the strategy can automatically adapt to the new operating environment and maintain good control performance without frequent manual adjustment of control parameters.

[0146] (4) Compared with related technologies, the present invention can obtain the probability or score of each action by having the intelligent agent perform forward neural network calculations in the control loop, and then select the final action by combining the action mask. The computation is mainly a small number of matrix multiplication and addition operations, which is suitable for deployment in substation automation devices, field controllers or dispatch master station servers. Compared with the method of directly solving mixed integer nonlinear programming problems, it greatly reduces the online computation complexity and meets the real-time requirements of distribution network regulation at the second level or even the sub-second level.

[0147] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0148] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0149] Example 2

[0150] According to embodiments of the present invention, an apparatus for implementing the above-described coordinated control method for multiple transformers is also provided. Figure 3 This is a structural block diagram of a collaborative control device for multiple transformers according to an embodiment of the present invention, such as... Figure 3 As shown, the device includes an acquisition module 302 and a determination module 304. The device will be described in detail below.

[0151] The acquisition module 302 is used to acquire multiple control cycles of the distribution network, wherein the distribution network includes multiple transformers;

[0152] The determining module 304, connected to the acquiring module 302, is used to control multiple transformers in the distribution network according to the execution order of multiple control cycles, in the following manner within the corresponding control cycles, until the multiple control cycles are completed: at the start time of the corresponding control cycle, acquiring the operating data corresponding to each of the multiple transformers; based on the operating data corresponding to each of the multiple transformers, determining the grid state parameters corresponding to the distribution network within the corresponding control cycle; based on the grid state parameters, determining the operation probability distribution corresponding to each of the multiple transformers within the corresponding control cycle; based on the operation probability distribution corresponding to each of the multiple transformers, determining the target control operation corresponding to each of the multiple transformers within the corresponding control cycle; and based on the target control operation corresponding to each of the multiple transformers, controlling the multiple transformers in the distribution network within the corresponding control cycle.

[0153] It should be noted that the above-mentioned acquisition module 302 and determination module 304 correspond to steps S102 to S104 in the method for implementing coordinated control of multiple transformers. The multiple modules and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiment 1.

[0154] Example 3

[0155] According to another aspect of the present invention, an electronic device is also provided, comprising: a processor; and a memory for storing processor-executable instructions, wherein the processor is configured to execute instructions to implement the cooperative control method for multiple transformers of any of the above embodiments.

[0156] Example 4

[0157] According to another aspect of the present invention, a computer-readable storage medium is also provided, which, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform the cooperative control method of multiple transformers described above.

[0158] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0159] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0160] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0161] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0162] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0163] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0164] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for coordinated control of multiple transformers, characterized in that, include: Multiple control cycles of a power distribution network are obtained, wherein the power distribution network includes multiple transformers; Following the execution sequence of multiple control cycles, within each corresponding control cycle, multiple transformers in the distribution network are controlled in the following manner until the multiple control cycles are completed: At the start time of each corresponding control cycle, operational data corresponding to each of the multiple transformers is acquired; based on the operational data corresponding to each of the multiple transformers, grid state parameters corresponding to the distribution network within the corresponding control cycle are determined; based on the grid state parameters, the operation probability distribution corresponding to each of the multiple transformers within the corresponding control cycle is determined; based on the operation probability distribution corresponding to each of the multiple transformers, the target control operation corresponding to each of the multiple transformers within the corresponding control cycle is determined; based on the target control operation corresponding to each of the multiple transformers, the multiple transformers in the distribution network are controlled within the corresponding control cycle.

2. The method according to claim 1, characterized in that, The step of determining the target control operation corresponding to each of the multiple transformers within the corresponding control period based on the operation probability distribution of each transformer includes: At the start time of the corresponding control cycle, determine the power parameters corresponding to the plurality of transformers and the operating constraint parameters corresponding to the distribution network. Based on the power parameters corresponding to the multiple transformers, determine the power deviation characteristics corresponding to the multiple transformers; Based on the aforementioned operational constraint parameters, determine the operational deviation characteristics corresponding to the distribution network; Based on the power deviation characteristics, the operating deviation characteristics, and the operation probability distributions corresponding to the multiple transformers, the target control operations corresponding to the multiple transformers within the corresponding control cycle are determined.

3. The method according to claim 2, characterized in that, The step of determining the operational deviation characteristics corresponding to the distribution network based on the operational constraint parameters includes: When the operating constraint parameters include current constraint parameters, voltage constraint parameters, and load constraint parameters, the operating deviation characteristics corresponding to the distribution network are determined based on the current constraint parameters, the voltage constraint parameters, and the load constraints.

4. The method according to claim 1, characterized in that, The step of determining the target control operation corresponding to each of the multiple transformers within the corresponding control period based on the operation probability distribution of each transformer includes: After determining the target control operation corresponding to the first control cycle among the plurality of control cycles, the target control operation corresponding to the corresponding target cycle is determined in the following manner according to the execution order of the plurality of target cycles: Determine the preceding control cycle corresponding to the corresponding target cycle, wherein the plurality of target cycles are control cycles following the first control cycle among the plurality of control cycles; Determine the previous control operation corresponding to each of the plurality of transformers within the previous control cycle; Based on the previous control operation corresponding to each of the multiple transformers, determine the operation deviation characteristics corresponding to the multiple transformers; Based on the operation probability distribution and operation deviation characteristics corresponding to the multiple transformers, the target control operation corresponding to the target period is determined.

5. The method according to claim 1, characterized in that, The step of determining the operation probability distribution corresponding to each of the plurality of transformers within the corresponding control cycle based on the power grid state parameters includes: For any one of the plurality of transformers, the operation probability distribution corresponding to that transformer within the corresponding control cycle is determined using the following method: Determine the tap control constraints and multiple voltage control taps corresponding to any one of the transformers, wherein the tap control constraints include a first tap constraint and a second tap constraint, the first tap constraint is used to constrain the control operation of the first control tap, the second tap constraint is used to constrain the control operation of the second control tap, the first control tap is the largest voltage control tap among the multiple voltage control taps, and the second control tap is the smallest voltage control tap among the multiple voltage control taps; Determine multiple candidate control operations corresponding to the multiple voltage control levels, wherein the multiple candidate control operations are used to adjust the corresponding voltage control levels at different levels; Based on the power grid state parameters, the tap control constraints, and the multiple candidate control operations corresponding to the multiple voltage control taps, the operation probability distribution corresponding to any one transformer within the corresponding control cycle is determined.

6. The method according to claim 5, characterized in that, Determining multiple voltage control levels corresponding to any one of the transformers, including: Determine the equipment parameters corresponding to any one of the transformers, wherein the equipment parameters include rated capacity, winding connection method and number of circuit connectors; Based on the equipment parameters, voltage compensation parameters corresponding to each of the transformers are determined, wherein the voltage compensation parameters are used to reflect the voltage compensation characteristics of each transformer to the distribution network; Based on the voltage compensation parameters, multiple voltage control levels corresponding to any one of the transformers are determined.

7. The method according to any one of claims 1 to 6, characterized in that, Before obtaining multiple control cycles of the distribution network, the following are included: Determine the scenario parameters and operational requirement parameters corresponding to the power distribution network; Based on the scenario parameters and operational requirements parameters, multiple control cycles corresponding to the power distribution network are determined.

8. A coordinated control device for multiple transformers, characterized in that, include: An acquisition module is used to acquire multiple control cycles of a power distribution network, wherein the power distribution network includes multiple transformers; The determination module is configured to control multiple transformers in the distribution network according to the execution sequence of multiple control cycles, and within the corresponding control cycles, control them in the following manner until the multiple control cycles are completed: at the start time of the corresponding control cycle, acquire the operating data corresponding to each of the multiple transformers; based on the operating data corresponding to each of the multiple transformers, determine the grid state parameters corresponding to the distribution network within the corresponding control cycle; based on the grid state parameters, determine the operation probability distribution corresponding to each of the multiple transformers within the corresponding control cycle; based on the operation probability distribution corresponding to each of the multiple transformers, determine the target control operation corresponding to each of the multiple transformers within the corresponding control cycle; and based on the target control operation corresponding to each of the multiple transformers, control the multiple transformers in the distribution network within the corresponding control cycle.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the collaborative control method for multiple transformers as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the coordinated control method of the plurality of transformers as described in any one of claims 1 to 7.

Images