A cascade power station intelligent operation method and system

Abstract

A method and system for intelligent operation of cascade power stations, comprising the following steps: step S1, continuously updating a preset operation knowledge graph; step S2, monitoring the operation status information of the cascade power station; step S3, deciding an operation control mode based on the operation status information and the operation knowledge graph; step S4, executing the operation control mode on the cascade power station. The method and system for intelligent operation of cascade power stations of the present invention decides an operation control mode based on the operation status information and the operation knowledge graph of the cascade power station, executes the operation control mode on the cascade power station, and adaptively and automatically manages the operation mode of the cascade power station, which greatly reduces the labor cost and improves the efficiency of the operation mode management, and is also more intelligent.

Description

Technical Field

[0001] The present invention belongs to the technical field of operation control of cascade power stations, and in particular relates to an intelligent operation method and system for cascade power stations. Background Art

[0002] At present, operation mode management is an important part of the cascade power production process. The main work is to coordinate the relevant needs of the power generation side and the grid side, coordinate the grid dispatching and power stations, reasonably arrange the operation mode of plant and station equipment, and formulate power generation, maintenance and commissioning plans. Operation mode management has the following characteristics: First, it involves a wide range of areas, complicated matters, and many boundary conditions; second, it is related to power safety production, with low fault tolerance and high accuracy requirements; third, the work content covers hydrology and power technology, power generation side and grid side, and power market and other types of information; fourth, there are many docking units and a large amount of coordination, which is an important source of decision-making support and an important production node for information integration and interactive release.

[0003] However, the current operation mode management is performed by a large number of dispatching and operating personnel in the cascade power stations. As a result, the labor cost is high and the efficiency of operation mode management is low. In addition, the degree of intelligence is insufficient. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an intelligent operation method and system for cascade power stations, which decides the operation and control mode based on the operation status information and operation knowledge graph of the cascade power station, executes the operation and control mode on the cascade power station, and adaptively and automatically manages the operation mode of the cascade power station, which greatly reduces the labor cost and improves the efficiency of operation mode management, and at the same time, is more intelligent.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for intelligent operation of a cascade power station, comprising the following steps:

[0007] Step S1, continuously updating the preset operation knowledge graph;

[0008] Step S2, monitoring the operating status information of the cascade power station;

[0009] Step S3: Based on the operation status information and the operation knowledge graph, decide on the operation control mode;

[0010] Step S4: Execute the operation control mode for the cascade power station.

[0011] Preferably, the sub-steps of step S1 are:

[0012] Step S1.1, obtaining attribute information of the cascade power station;

[0013] Step S1.2, based on the attribute information, determine the knowledge applicable conditions;

[0014] Step S1.3, based on the knowledge applicability condition, incrementally search the first knowledge source;

[0015] Step S1.4, parsing the source type of the first knowledge source;

[0016] Step S1.5: When a first data extraction template corresponding to the source type exists in the preset first template library, first operation knowledge is extracted from the first knowledge source based on the first data extraction template; and the first operation knowledge is added to the operation knowledge graph.

[0017] Preferably, in step S1.5, when the first data extraction template does not exist in the first template library, the following steps are adopted:

[0018] Step S1.6, obtaining a second data extraction template from the first knowledge source;

[0019] Step S1.7: based on the second data extraction template, try to extract the second operation knowledge from the first knowledge source;

[0020] Step S1.8: When the attempt is successful, the second operation knowledge is added to the operation knowledge graph.

[0021] Preferably, in step S1.6, obtaining the second data extraction template of the first knowledge source includes:

[0022] Obtain the second knowledge source and the third operating knowledge provided by the source of the first knowledge source N times before; the second knowledge source and the third operating knowledge correspond one to one; N is a positive integer;

[0023] determining whether to use a third data extraction template when obtaining the third operating knowledge based on the second knowledge source;

[0024] When in use, the third data extraction template is used as the second data extraction template; otherwise, the content position of the third operation knowledge in the second knowledge source is obtained;

[0025] Extracting features of target content within a preset position range before and after the content position in the second knowledge source to obtain content features;

[0026] Determine data extraction rules based on content features;

[0027] Based on the data extraction rule, the second data extraction template is generated; and / or the content type set of the first knowledge source is parsed; a fourth data extraction template is determined from a preset second template library; the matching degree between the preset standard content type set and the content type set corresponding to the fourth data extraction template is greater than or equal to a preset matching degree threshold;

[0028] The fourth data extraction template is used as the second data extraction template.

[0029] Preferably, the sub-steps of step S3 are:

[0030] Step S3.1, determining a graph knowledge query formula based on the operation status information;

[0031] Step S3.2: Based on the graph knowledge query formula, query the first graph knowledge to be selected from the running knowledge graph;

[0032] Step S3.3: when the first candidate graph knowledge is unique, determine the operation control mode based on the first candidate graph knowledge; otherwise, determine the second candidate graph knowledge that meets the standard knowledge condition from the first candidate graph knowledge;

[0033] Step S3.4: when the second graph knowledge to be selected is unique, determine the operation control mode based on the second graph knowledge to be selected; otherwise, obtain the effect value of the second graph knowledge to be selected;

[0034] Step S3.5: Determine the operation control mode based on the second candidate graph knowledge with the maximum effect value.

[0035] Preferably, in step S3.3, the standard knowledge conditions include:

[0036] The number of knowledge of the target graph knowledge in the running knowledge graph is greater than or equal to a preset number threshold; there is at least one association relationship of a preset standard association type between the target graph knowledge and the second selected graph knowledge;

[0037] The sum of the preset type weights corresponding to the query type of the graph knowledge query that the second graph knowledge to be selected conforms to is greater than or equal to the preset type weight and threshold;

[0038] The durability of the second to-be-selected graph knowledge is greater than or equal to a preset durability threshold.

[0039] Preferably, obtaining the effect value of the second candidate graph knowledge includes:

[0040] 1) Obtaining historical knowledge application information of the second candidate graph knowledge;

[0041] 2) Based on a preset knowledge application effect evaluation system, the effect value is determined according to historical knowledge application information.

[0042] A cascade power station intelligent operation system adopts a cascade power station intelligent operation method, including:

[0043] Update module, used to continuously update the preset running knowledge graph;

[0044] A monitoring module, used to monitor the operating status information of the cascade power station;

[0045] A decision module, used to decide on an operation control mode based on the operation status information and the operation knowledge graph;

[0046] An execution module is used to execute the operation control mode on the cascade power station.

[0047] The present invention can achieve the following beneficial effects:

[0048] The present invention decides the operation and control mode based on the operation status information and operation knowledge graph of the cascade power station, executes the operation and control mode on the cascade power station, and adaptively and automatically manages the operation mode of the cascade power station, which greatly reduces the labor cost and improves the efficiency of operation mode management. At the same time, it is also more intelligent. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0050] Figure 1 This is a flow chart of an intelligent operation method of a cascade power station according to the present invention;

[0051] Figure 2 The figure is a schematic diagram of an intelligent operation system of a cascade power station according to the present invention. DETAILED DESCRIPTION

[0052] Embodiment 1:

[0053] The preferred solution is Figure 1 As shown, a method and system for intelligent operation of a cascade power station, the steps are:

[0054] Step S1: continuously updating the preset running knowledge graph;

[0055] A large amount of operation knowledge of cascade power stations is marked on the operation knowledge graph. The operation knowledge indicates the appropriate operation and control methods under different operation conditions of the cascade power stations. The operation knowledge can be set in advance by the dispatching and operation personnel of the cascade power stations; the operation status information reflects the operation status of the cascade power stations, which can be: power station units, transformers, switches, busbars, lines and external transmission systems near-grid AC, cross-regional DC operation status, unit output, total branch plant output, line output, safety control switching status, inflow flow, outflow flow, upstream and downstream representative station water levels, tail water level, head, water consumption rate, adjustable storage capacity, etc.; based on the operation status information and the operation knowledge graph, the operation and control method can be decided; the operation and control method is executed for the cascade power stations.

[0056] The present invention decides the operation and control mode based on the operation status information and operation knowledge graph of the cascade power station, executes the operation and control mode on the cascade power station, and adaptively and automatically manages the operation mode of the cascade power station, which greatly reduces the labor cost and improves the efficiency of operation mode management. At the same time, it is also more intelligent.

[0057] In this embodiment, the preset operation knowledge graph is continuously updated, including:

[0058] 1) Obtain the attribute information of cascade power stations;

[0059] 2) Determine the applicable conditions of knowledge based on attribute information;

[0060] 3) Incrementally search the first knowledge source based on the knowledge applicability conditions;

[0061] 4) Analyze the source type of the first knowledge source;

[0062] When a first data extraction template corresponding to the source type exists in the preset first template library, first operation knowledge is extracted from the first knowledge source based on the first data extraction template; and the first operation knowledge is added to the operation knowledge graph.

[0063] The attribute information includes: the models of various equipment in the cascade power station, the type of geographical location, etc.; the knowledge applicability condition is the condition that the first operating knowledge extracted from the first knowledge source is applicable to the cascade power station, for example: the cascade power station equipment model involved in the first knowledge source exists in the attribute information; incremental search refers to searching for the new first knowledge source generated after the last update of the operating knowledge graph; the source type of the first knowledge source is divided into cascade power station operation control experience, etc.; the first template library has a first data extraction template corresponding to the source type, and the first data extraction template is a system execution template that extracts the first operating knowledge from the first knowledge source of the source type, for example: if the source type is cascade power station operation control experience, then the first data extraction template is to extract the cascade power station operation status and the control scheme faced by the cascade power station operation control from the first knowledge source as the first operating knowledge; when the first data extraction template exists, the first operating knowledge is extracted from the first knowledge source directly based on the first data extraction template, and the first operating knowledge is added to the operation knowledge graph.

[0064] This embodiment continuously updates the operation knowledge graph, thereby improving the ability of the operation knowledge graph to be used for decision-making on operation control methods. In addition, the first data extraction template is introduced to improve the efficiency of knowledge graph updating.

[0065] In this embodiment, continuously updating the preset operation knowledge graph also includes:

[0066] 1) When the first data extraction template does not exist in the first template library, a second data extraction template from the first knowledge source is obtained.

[0067] 2) Based on the second data extraction template, try to extract the second operational knowledge from the first knowledge source.

[0068] 3) When the attempt is successful, the second operation knowledge is added to the operation knowledge graph.

[0069] Wherein, obtaining the second data extraction template of the first knowledge source includes:

[0070] Obtain the second knowledge source and the third operating knowledge provided by the source of the first knowledge source N times before; the second knowledge source and the third operating knowledge correspond one to one; N is a positive integer.

[0071] It is determined whether to use a third data extraction template when obtaining third operational knowledge based on the second knowledge source.

[0072] When used, the third data extraction template is used as the second data extraction template; otherwise, the content position of the third operating knowledge in the second knowledge source is obtained.

[0073] Feature extraction is performed on target content within a preset position range before and after the content position in the second knowledge source to obtain content features.

[0074] Based on content features, determine data extraction rules.

[0075] Based on the data extraction rules, a second data extraction template is generated; and / or a content type set of the first knowledge source is parsed; a fourth data extraction template is determined from a preset second template library; the degree of match between the preset standard content type set and the content type set corresponding to the fourth data extraction template is greater than or equal to a preset matching degree threshold; and the fourth data extraction template is used as the second data extraction template.

[0076] When the first data extraction template does not exist, it means that the second data extraction template needs to be obtained immediately, and based on the second data extraction template, the second operation knowledge is tried to be extracted from the first knowledge source; the source of the first knowledge source can be a certain trapezoidal power station, a certain trapezoidal power station dispatching and operation personnel, etc.; N can be set in advance by the technical personnel; for example, it is: 10. In the past, the source party provided the second knowledge source, and the third operation knowledge was obtained based on the second knowledge source. If the third data extraction template was used when the third operation knowledge was obtained based on the second knowledge source, it means that the source party has recently and shortly provided a second knowledge source from which the third operation knowledge can be extracted by the third data extraction template. The third data extraction template can be used as the second data extraction template to try to extract the second operation knowledge; if the third data extraction template is not used when the third operation knowledge is obtained based on the second knowledge source, the second data extraction template needs to be generated; the third operation knowledge is derived from the second knowledge source, so the content in the third operation knowledge comes from the second knowledge source, and has a corresponding content position in the second knowledge source; the preset position range can be, for example: 3 content positions. The target content is other content within the preset position range before and after the content position in the second knowledge source; content features include: content type, whether the publisher of the content is the same as the publisher of the third operation knowledge, etc.; generally, when the first data extraction template cannot be used directly, the source type of the first knowledge source can be a forum page for sharing the operation and control experience of cascade power plants, for example: the operation and control personnel of different cascade power plants will share their experience in the internal forum. At this time, the content position of the third operation knowledge in the second knowledge source is a position in the forum page, that is, the publishing position of the publisher of the third operation knowledge. The content type of other forum page content within a certain position range before and after the content position, whether the publisher of the content is the same as the publisher of the third operation knowledge, etc. determine what kind of forum page content can be used as the third operation knowledge. For example: the third operation knowledge is a scheme for controlling cascade power plants, the control scheme question of a cascade power plant operation and control personnel before the content position, and the approval and recognition words of a cascade power plant operation and control personnel after the content position.

[0077] Therefore, data extraction rules can be determined based on content features; for example: if the content feature is a question about the control solution and whether the publisher of the content is different from the publisher of the third operating knowledge, then the data extraction rule is generated to extract the content after the question about the control solution published by the publisher of the non-extraction object in the forum page; based on the data extraction rule, a second data extraction template is generated, and when generating the template, the data extraction rules are integrated; the content type set includes different content types in the first knowledge source; the fourth data extraction template corresponds to a preset standard content type set, and when the matching degree between the content type set and the standard content type set is greater than or equal to the preset matching degree threshold, it indicates that the corresponding fourth data extraction template is available; the fourth data extraction template is used as the second data extraction template; the matching degree threshold can be, for example: 80; the preset standard content type set corresponding to the fourth data extraction template can be set in advance by technical personnel according to actual needs.

[0078] When the first data extraction template cannot be used directly to extract the first operation knowledge, the embodiment of the present invention introduces a second data extraction template to extract the second operation knowledge. When obtaining the second data extraction template, two acquisition schemes are introduced to further improve the comprehensiveness and efficiency of updating the operation knowledge graph.

[0079] Step S2: monitoring the operating status information of the cascade power station;

[0080] Step S3: Based on the operation status information and the operation knowledge graph, decide on the operation control method;

[0081] Based on the operation status information, determine the graph knowledge query formula;

[0082] Based on the graph knowledge query formula, the first candidate graph knowledge is queried from the running knowledge graph;

[0083] When the first candidate graph knowledge is unique, the operation control mode is determined based on the first candidate graph knowledge; otherwise, the second candidate graph knowledge that meets the standard knowledge condition is determined from the first candidate graph knowledge;

[0084] When the second candidate graph knowledge is unique, the operation control mode is determined based on the second candidate graph knowledge; otherwise, the effect value of the second candidate graph knowledge is obtained;

[0085] Determine the operation control method based on the knowledge of the second candidate graph with the maximum effect value;

[0086] Among them, standard knowledge conditions include:

[0087] The number of knowledge of the target graph knowledge in the running knowledge graph is greater than or equal to a preset number threshold; there is at least one association relationship of a preset standard association type between the target graph knowledge and the second selected graph knowledge;

[0088] The sum of the preset type weights corresponding to the query type of the graph knowledge query that the second selected graph knowledge matches is greater than or equal to the preset type weight and threshold;

[0089] The durability of the second candidate graph knowledge is greater than or equal to a preset durability threshold.

[0090] The graph knowledge query is a query for the first candidate graph knowledge (operation knowledge) that queries the operation status information represented by the cascade power station operation status from the operation knowledge graph for the decision-making operation and control method applicable to the operation status. For example, if the operation status information is inflow A and outflow B, then the graph knowledge query is (inflow A) and (outflow B), and and is an and logical relationship. If the first candidate graph knowledge queried is unique, the operation and control method is determined directly based on the first candidate graph knowledge. Otherwise, it means that there are multiple first candidate graph knowledge, and the second candidate graph knowledge is selected from them based on the standard knowledge condition. If the second candidate graph knowledge is unique, the operation and control method is determined directly based on the second candidate graph knowledge. Otherwise, it means that there are multiple second candidate graph knowledge, and the effect value is introduced. Based on the second candidate graph knowledge with the largest effect value, the operation and control method is determined. The operation knowledge graph is used to determine the operation and control method in a layered and progressive manner. Taking into account different query results, the applicability of the system is improved; the number threshold can be, for example: 3; the standard association type can be: after using the second candidate graph knowledge, the target graph knowledge may be used in the next step (the target graph knowledge is applicable to the operating state that may appear in the future after the operating state represented by the operating state information of the volume power station appears), etc.; the query type is divided into one-type query type, two-type query type, three-type query type, etc., for example: (inbound flow A) and (outbound flow B) is a two-type query type; the type weight represents the degree to which the second candidate graph knowledge conforms to the graph knowledge query type. The more formulas in the graph knowledge query type that conform, the greater the degree of conformity; the durability is the retention time of the second candidate graph knowledge on the operating knowledge graph. The longer the retention time, the more real and reliable the second candidate graph knowledge is; the durability threshold can be, for example: 8; set standard knowledge conditions to preliminarily and finely screen out the second candidate graph knowledge.

[0091] In this embodiment, obtaining the effect value of the second candidate graph knowledge includes:

[0092] Acquire historical knowledge application information of the second candidate graph knowledge;

[0093] Based on the preset knowledge application effect evaluation system, the effect value is determined according to the historical knowledge application information.

[0094] The historical knowledge application information is the record of the historical use of the second candidate graph knowledge to determine the operation and control methods of the cascade power station, such as: the stable operation time of the volume power station after use; the knowledge application effect evaluation system is a system that determines the effect value based on the historical knowledge application information, such as: there are effect values ​​corresponding to the stable operation time of the volume power station after use, etc. The longer the stable operation time of the volume power station after use, the greater the corresponding effect value.

[0095] Step S4: Execute the operation control mode for the cascade power station.

[0096] Embodiment 2:

[0097] The embodiment of the present invention provides an intelligent operation system for a cascade power station, such as Figure 2 As shown, including:

[0098] Update module 1, used to continuously update the preset running knowledge graph;

[0099] Monitoring module 2, used to monitor the operating status information of the cascade power station;

[0100] Decision module 3, used to decide the operation control mode based on the operation status information and the operation knowledge graph;

[0101] The execution module 4 is used to execute the operation control mode for the cascade power station.

[0102] The updating module 1 continuously updates the preset operation knowledge graph, including: obtaining the attribute information of the cascade power station; determining the knowledge applicability conditions based on the attribute information; incrementally searching the first knowledge source based on the knowledge applicability conditions; parsing the source type of the first knowledge source; when there is a first data extraction template corresponding to the source type in the preset first template library, extracting the first operation knowledge from the first knowledge source based on the first data extraction template; and adding the first operation knowledge to the operation knowledge graph.

[0103] The decision module 3 decides the operation control mode based on the operation status information and the operation knowledge graph, including: determining the graph knowledge query formula based on the operation status information; querying the first graph knowledge to be selected from the operation knowledge graph based on the graph knowledge query formula; when the first graph knowledge to be selected is unique, determining the operation control mode based on the first graph knowledge to be selected; otherwise, determining the second graph knowledge to be selected that meets the standard knowledge conditions from the first graph knowledge to be selected; when the second graph knowledge to be selected is unique, determining the operation control mode based on the second graph knowledge to be selected; otherwise, obtaining the effect value of the second graph knowledge to be selected; determining the operation control mode based on the second graph knowledge to be selected with the maximum effect value;

[0104] Among them, the standard knowledge conditions include: the number of knowledge of the target graph knowledge in the running knowledge graph is greater than or equal to the preset number threshold; there is at least one association relationship of a preset standard association type between the target graph knowledge and the second selected graph knowledge; the sum of the preset type weights corresponding to the query type of the graph knowledge query formula that the second selected graph knowledge conforms to is greater than or equal to the preset type weight and threshold; the durability of the second selected graph knowledge is greater than or equal to the preset durability threshold.

[0105] The decision module 3 obtains the effect value of the second candidate graph knowledge, including: obtaining the historical knowledge application information of the second candidate graph knowledge; and determining the effect value according to the historical knowledge application information based on a preset knowledge application effect evaluation system.

[0106] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A method for intelligent operation of cascade power plants, characterized in that The following steps are involved: Step S1, continuously updating the preset operation knowledge graph; Step S2, monitoring the operating status information of the cascade power station; Step S3: Based on the operation status information and the operation knowledge graph, decide on the operation control mode; Step S4, executing the operation control mode for the cascade power station; The sub-steps of step S3 are: Step S3.1, determining a graph knowledge query formula based on the operation status information; Step S3.2: Based on the graph knowledge query formula, query the first graph knowledge to be selected from the running knowledge graph; Step S3.3: when the first candidate graph knowledge is unique, determine the operation control mode based on the first candidate graph knowledge; Otherwise, determining the second candidate graph knowledge that meets the standard knowledge condition from the first candidate graph knowledge; Step S3.4: when the second graph knowledge to be selected is unique, determine the operation control mode based on the second graph knowledge to be selected; otherwise, obtain the effect value of the second graph knowledge to be selected; Step S3.5, determining an operation control mode based on the second candidate graph knowledge of the maximum effect value; In step S3.3, the standard knowledge conditions include: The number of knowledge of the target graph knowledge in the running knowledge graph is greater than or equal to a preset number threshold; there is at least one association relationship of a preset standard association type between the target graph knowledge and the second selected graph knowledge; The sum of the preset type weights corresponding to the query type of the graph knowledge query that the second graph knowledge to be selected conforms to is greater than or equal to the preset type weight and threshold; The durability of the second to-be-selected graph knowledge is greater than or equal to a preset durability threshold; Obtaining the effect value of the second candidate graph knowledge includes: 1) Obtaining historical knowledge application information of the second candidate graph knowledge; 2) Based on a preset knowledge application effect evaluation system, the effect value is determined according to historical knowledge application information.

2. The intelligent operation method of a cascade power station according to claim 1, characterized in that: Sub-step S of step S1 is: Step S1.1, obtaining attribute information of the cascade power station; Step S1.2, based on the attribute information, determine the knowledge applicable conditions; Step S1.3, based on the knowledge applicability condition, incrementally search the first knowledge source; Step S1.4, parsing the source type of the first knowledge source; Step S1.5: when there is a first data extraction template corresponding to the source type in the preset first template library, extracting first operation knowledge from the first knowledge source based on the first data extraction template; Add the first operation knowledge into the operation knowledge graph.

3. The intelligent operation method of a cascade power station according to claim 2, characterized in that: In step S1.5, when the first data extraction template does not exist in the first template library, the following steps are adopted: Step S1.6, obtaining a second data extraction template from the first knowledge source; Step S1.7: based on the second data extraction template, try to extract the second operation knowledge from the first knowledge source; Step S1.8: When the attempt is successful, the second operation knowledge is added to the operation knowledge graph.

4. The intelligent operation method of a cascade power station according to claim 3, characterized in that: In step S1.6, obtaining the second data extraction template of the first knowledge source includes: Obtain the second knowledge source and the third operating knowledge provided by the source of the first knowledge source N times before; the second knowledge source and the third operating knowledge correspond one to one; N is a positive integer; determining whether to use a third data extraction template when obtaining the third operating knowledge based on the second knowledge source; When in use, the third data extraction template is used as the second data extraction template; otherwise, the content position of the third operation knowledge in the second knowledge source is obtained; Extracting features of target content within a preset position range before and after the content position in the second knowledge source to obtain content features; Determine data extraction rules based on content features; Based on the data extraction rule, the second data extraction template is generated; and / or the content type set of the first knowledge source is parsed; a fourth data extraction template is determined from a preset second template library; the matching degree between the preset standard content type set and the content type set corresponding to the fourth data extraction template is greater than or equal to a preset matching degree threshold; The fourth data extraction template is used as the second data extraction template.

5. An intelligent operation system for cascade power stations, characterized in that: A cascade power station intelligent operation method according to any one of claims 1 to 4 is adopted, comprising: Update module, used to continuously update the preset running knowledge graph; A monitoring module, used to monitor the operating status information of the cascade power station; A decision module, used to decide on an operation control mode based on the operation status information and the operation knowledge graph; An execution module is used to execute the operation control mode on the cascade power station.

Images