Circulating water energy-saving control method and device, electronic equipment and storage medium

Abstract

The invention provides a circulating water energy-saving control method and device, electronic equipment and a storage medium, and belongs to the technical field of intelligent water treatment.The method comprises the steps that equipment distribution data and operation data of a circulating water system are obtained; the operation data comprises water flow, pipeline pressure, water pump load and circulating water temperature, and the equipment distribution data comprises water pump distribution data, pipeline distribution data, cooling equipment distribution data, heat exchange equipment distribution data and water storage and water treatment equipment distribution data; constructing a spatial partition data association graph based on the equipment distribution data and the operation data; determining the working condition type of each partition based on the spatial partition data association graph; based on the working condition type and the operation data of each partition, determining an equipment regulation and control strategy through a target energy-saving control large model; and controlling the circulating water system based on the equipment regulation and control strategy. The energy-saving effect of different circulating water systems under different working conditions can be improved.

Description

Technical Field

[0001] This application belongs to the field of smart water treatment technology, and more specifically, relates to methods and devices for energy-saving control of circulating water, electronic equipment, and storage media. Background Technology

[0002] Circulating water systems are core auxiliary systems in industrial production, widely used in chemical, steel, and pharmaceutical industries. Their main function is to cool equipment or exchange heat through water, ensuring stable operation of production processes. With the increasing urgency of industrial energy conservation, energy-saving control of circulating water systems has become a focus of industry attention. Currently, the energy consumption of industrial circulating water systems accounts for a high proportion of a company's total energy consumption; therefore, achieving energy conservation in water pumps through scientific regulation is the core direction for energy conservation in circulating water systems.

[0003] Existing circulating water energy-saving control technologies mainly fall into two categories: one is closed-loop control based on a single operating parameter, which adjusts the pump frequency using a PID algorithm by collecting water flow or pipeline pressure data; the other is global regulation based on a general energy-saving model, which is directly applied to different systems after training the model using industry-standard data. However, the problem with traditional methods is that the regulation strategies lack specificity and are difficult to adapt to complex fluctuation scenarios; the general models do not consider the individual operating characteristics of enterprises, resulting in poor adaptability and limited energy-saving effects. Therefore, there is an urgent need for a circulating water energy-saving control method that can adapt to different operating conditions, solving the problems of insufficient specificity, poor adaptability, and limited energy-saving effects of existing technologies. Summary of the Invention

[0004] The purpose of this application is to provide energy-saving control methods and devices for circulating water, electronic equipment, and storage media to improve the energy-saving effect of different circulating water systems under different operating conditions.

[0005] A first aspect of this application provides a circulating water energy-saving control method, comprising: Acquire equipment distribution data and operational data of the circulating water system; the operational data includes water flow rate, pipeline pressure, pump load and circulating water temperature, and the equipment distribution data includes pump distribution data, pipeline distribution data, cooling equipment distribution data, heat exchange equipment distribution data and water storage and water treatment equipment distribution data; A spatial partition data association diagram is constructed based on the device distribution data and the operation data; the spatial partition data association diagram includes various partitions divided according to the device distribution data, and each partition includes multiple devices and their respective operation data; The operating condition type of each partition is determined based on the spatial partition data association diagram. Based on the operating conditions and data of each partition, the equipment control strategy is determined through the target energy-saving control model. The circulating water system is controlled based on the aforementioned equipment regulation strategy.

[0006] A second aspect of this application provides a circulating water energy-saving control device, comprising: The data acquisition module is used to acquire equipment distribution data and operation data of the circulating water system; the operation data includes water flow rate, pipeline pressure, water pump load and circulating water temperature, and the equipment distribution data includes water pump distribution data, pipeline distribution data, cooling equipment distribution data, heat exchange equipment distribution data and water storage and water treatment equipment distribution data; The partitioning module is used to construct a spatial partitioning data association diagram based on the device distribution data and the operation data; the spatial partitioning data association diagram includes various partitions divided according to the device distribution data, and each partition includes multiple devices and their respective operation data; The operating condition analysis module is used to determine the operating condition type of each partition based on the spatial partition data association diagram. The strategy generation module is used to determine the equipment control strategy based on the operating condition type and operation data of each partition through the target energy-saving control big model; The control module is used to control the circulating water system based on the equipment regulation strategy.

[0007] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described circulating water energy-saving control method.

[0008] In a fourth aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described circulating water energy-saving control method.

[0009] The beneficial effects of the circulating water energy-saving control method and device, electronic equipment, and storage medium provided in this application embodiment are as follows: Existing technologies often employ global unified control, ignoring the operational differences of equipment in different areas. This application's embodiment, by acquiring equipment distribution data, divides the space into dedicated areas based on equipment spatial layout, and constructs a correlation diagram corresponding to the area, equipment, and operational data. This allows for precise analysis of the actual operational status of each area, avoiding over-control of low-load areas and under-control of high-load areas.

[0010] Traditional general models cannot be adapted to the specific equipment conditions of an enterprise, making it difficult to match actual operational needs. The energy-saving control in this application not only relies on the enterprise's own equipment distribution data but also integrates real-time operational data, making it more suitable for the enterprise's equipment characteristics and operating habits, and significantly improving the adaptability of the control.

[0011] This application's embodiments accurately determine the operating condition type of each area, ensuring precise matching between the control strategy and the operating condition status. This minimizes ineffective energy consumption by water pumps, significantly improves energy efficiency, and helps enterprises reduce production costs. Because the control strategy is based on actual regional data and operating condition types, frequent adjustments to equipment operating parameters are avoided. This effectively reduces the start-up and shutdown impacts and load fluctuations of core equipment such as water pumps, ensuring the continuous and stable operation of the circulating water system, preventing production processes from being affected by improper control, reducing equipment wear, extending equipment lifespan, and reducing maintenance and replacement costs for enterprises. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic flowchart of a circulating water energy-saving control method provided in an embodiment of this application; Figure 2 A structural block diagram of a circulating water energy-saving control device provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0015] It is understood that in the embodiments of this application, data such as user information are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with relevant laws, regulations and standards.

[0016] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0017] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a circulating water energy-saving control method provided in an embodiment of this application. The method can be executed by an electronic device, and specifically, the method may include S101 to S105.

[0018] S101: Obtain equipment distribution data and operation data of the circulating water system; operation data includes water flow rate, pipeline pressure, water pump load and circulating water temperature, and equipment distribution data includes water pump distribution data, pipeline distribution data, cooling equipment distribution data, heat exchange equipment distribution data and water storage and water treatment equipment distribution data.

[0019] In this embodiment, the circulating water system refers to an auxiliary system used for equipment cooling or heat exchange in industrial production, and is an integrated whole of various equipment working together. Equipment distribution data refers to static data such as the spatial location, specifications, and connection relationships of various equipment in the circulating water system. Operational data refers to dynamic parameter data generated in real time during system operation. Water flow rate refers to the amount of circulating water transported per unit time, used to characterize water transport capacity. Pipeline pressure refers to the pressure value of the circulating water in the pipeline, used to reflect the operating status of the pipeline network. Pump load refers to the ratio of the actual operating load of the pump to the rated load, characterizing the pump's workload. Circulating water temperature refers to the real-time temperature of the circulating water in the system, reflecting the heat exchange or cooling effect. Pump distribution data refers to the spatial location, quantity, and connection relationships of the pumps. Pipeline distribution data refers to the layout, diameter, length, and connection relationships of the pipelines. Cooling equipment distribution data refers to the location, specifications, and connection relationships of the cooling equipment. Heat exchange equipment distribution data refers to the location, specifications, and connection relationships of the heat exchange equipment. Water storage and water treatment equipment distribution data refers to the location, specifications, and connection relationships of the water storage and water treatment equipment.

[0020] Energy-saving control of circulating water systems must be based on a comprehensive understanding of the system. Equipment distribution data reflects the core information of the system's hardware layout, determining static attributes such as water flow paths and equipment coordination relationships; operational data reflects the system's real-time operating status, directly reflecting dynamic characteristics such as load demand and operating efficiency. Precise control cannot be achieved by relying on only one type of data. Therefore, this embodiment acquires both types of data simultaneously, forming complete data support for both static layout and dynamic operation. This ensures accurate subsequent zoning, condition determination, and the development of appropriate strategies, avoiding blind control due to missing data dimensions, and laying the foundation for the feasibility and effectiveness of the entire energy-saving control method.

[0021] For example, this embodiment can obtain relevant data through the following steps. First, acquire equipment distribution data, such as construction design drawings of the circulating water system, equipment procurement ledgers, and installation and acceptance documents, and extract the specifications, design locations, and theoretical connection relationships of various types of equipment from them. Then, through on-site surveys conducted by professional technicians, verify the actual installation locations, connection methods, and layout details of the equipment, supplementing any on-site information not covered by the drawings. This embodiment can employ digital modeling techniques to classify and organize the aforementioned equipment distribution data, establishing a standardized database containing various types of distribution data, and clarifying the relationships between different equipment distribution data.

[0022] This embodiment can acquire operational data through electromagnetic flow meters and pressure sensors installed at key pipeline transmission nodes, load monitoring devices configured at water pump motors, and temperature sensors installed at key equipment locations for circulating water inlet and outlet. Finally, the collected operational data undergoes preliminary screening to remove obviously invalid or abnormal data, and then the operational data is associated and bound with the distribution data of the corresponding equipment using unique equipment identifiers.

[0023] S102: Construct a spatial partition data association diagram based on device distribution data and operation data; the spatial partition data association diagram includes various partitions divided according to device distribution data, and each partition includes multiple devices and their respective operation data.

[0024] In this embodiment, a spatial partition data association graph is constructed based on device distribution data and operational data, including: Determine the equipment type and spatial coordinates of all equipment based on equipment distribution data; Based on the equipment type and spatial coordinates of all devices, the circulating water system is divided into multiple zones using the K-means clustering algorithm; these zones include a heat exchange zone, a cooling zone, a water storage zone, and a water pipeline zone. A device code identifier is generated for each device in each partition. The device code identifier includes the device ID, the partition to which the device belongs, and the device's spatial coordinates. Align the running data with each device to obtain the correspondence between all device codes and all running data; A spatial partition data association diagram is constructed based on the correspondence between all device codes and all operational data.

[0025] In this embodiment, the spatial partition data association diagram refers to the association carrier that integrates the partitions, equipment, and corresponding operational data of the circulating water system, used to intuitively present the correspondence between the spatial distribution of equipment and dynamic operational data. The K-means clustering algorithm is an unsupervised learning algorithm that divides samples into multiple categories based on data features; here, it is used to partition according to equipment attributes. Equipment code identifiers refer to unique information identifiers assigned to equipment, used to distinguish different equipment and associate their affiliation with their location. The heat exchange zone refers to a partition where heat exchange equipment is centrally located and primarily performs heat exchange functions. The cooling zone refers to a partition where cooling equipment is centrally located and primarily performs circulating water cooling functions. The water storage zone refers to a partition where water storage equipment is centrally located and primarily performs circulating water storage functions. The water pipeline zone refers to a partition where water pipelines are centrally located and primarily performs circulating water transportation functions. Equipment type refers to the functional category of the equipment, such as water pumps, pipelines, and cooling equipment. Equipment spatial coordinates refer to the actual spatial location parameters of the equipment within the circulating water system.

[0026] Circulating water systems involve diverse equipment types and complex spatial distribution, requiring precise management through rational zoning and data association. This embodiment first determines the equipment types and spatial coordinates, providing core characteristic basis for subsequent zoning. This embodiment employs the K-means clustering algorithm, enabling objective and reproducible zoning based on equipment attributes. This embodiment divides the system into functional zones such as heat exchange zones and cooling zones, conforming to the functional layout logic of industrial circulating water systems and facilitating the development of control strategies based on zone characteristics. Equipment coding ensures equipment uniqueness, avoiding data association confusion. This embodiment aligns operational data with equipment, achieving spatial binding between "equipment and operational data." The final constructed association diagram integrates static layout and dynamic data, providing a clear data carrier for subsequent operating condition determination and control strategy generation, avoiding blind control caused by a disconnect between equipment and operational data.

[0027] For example, this embodiment can construct a spatial partition data association graph through the following steps.

[0028] (1) Determine the equipment type and spatial coordinates of all equipment based on equipment distribution data. In this embodiment, the type information of various equipment such as water pumps, pipelines, and cooling equipment can be extracted from the equipment distribution data, and the spatial coordinates of the equipment marked in the design documents can also be extracted. Organize technical personnel to bring positioning equipment to the site to verify the actual spatial location of each piece of equipment, correct the deviation between the design coordinates and the actual location, and ensure the accuracy of equipment type and spatial coordinates.

[0029] (2) Based on the equipment type and spatial coordinates of all equipment, partitions are formed using the K-means clustering algorithm. In this embodiment, the verified equipment type (converted into feature vectors) and spatial coordinate data can be imported into the data processing platform. According to the functional requirements of the circulating water system, the number of clusters is set to 4 (corresponding to the heat exchange zone, cooling zone, water storage zone, and water pipeline zone). In this embodiment, the K-means algorithm can be started, and the clustering is based on the similarity of equipment types and spatial distance. The calculation is iteratively performed until the center of each category is stable, and the clustering results are output. The areas where the equipment of the same category is located are respectively determined as the heat exchange zone, cooling zone, water storage zone, and water pipeline zone, and the boundary range of each partition is marked.

[0030] (3) Generate equipment code identifiers for each device in each partition. In this embodiment, coding rules can be defined. The first segment of the code is the equipment ID (a unique serial number assigned according to the equipment installation order), the middle segment is the partition name to which the equipment belongs (e.g., HYQ for heat exchange zone), and the last segment is the spatial coordinates of the equipment (simplified expression according to the plant coordinate system). Through data processing tools, this embodiment can automatically generate a unique equipment code identifier for each device according to the rules and store it in the equipment information database.

[0031] (4) Align the operational data with each device. In this embodiment, real-time operational data can be retrieved from the data acquisition system, and the acquisition timestamp of each operational data point and the basic information of the corresponding acquisition device can be extracted. In this embodiment, the device code identifier can be used as the matching basis, combined with the timestamp synchronization principle, to associate and match the operational data such as water flow and pipeline pressure of the same device in the same time dimension with the device code identifier, forming a one-to-one correspondence between "device code identifier - operational data", and eliminating abnormal data that fail to match.

[0032] (5) Construct a spatial partition data association diagram based on the correspondence. In this embodiment, an industrial visualization modeling tool can be used to import the basic map of the circulating water system plant area. In this embodiment, the heat exchange area, cooling area, water storage area, and water pipeline area can be marked on the map according to the partition boundaries. Within each partition, the equipment location can be marked according to the equipment spatial coordinates and associated with the equipment code identifier. In this embodiment, the correspondence between "equipment code identifier - operating data" can be embedded into the model, so that clicking on the equipment icon can view its real-time operating data. Finally, an interactive spatial partition data association diagram is generated and stored in the system management platform for subsequent operation condition judgment and control.

[0033] S103: Determine the operating condition type of each partition based on the spatial partition data association diagram.

[0034] In this embodiment, the operating condition type of each partition is determined based on the spatial partition data association diagram, including: For each partition, obtain the corresponding equipment and the operating data of each equipment from the spatial partition data association diagram; calculate the flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load and temperature deviation rate based on the operating data; determine the operating condition type of the partition based on the flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load and temperature deviation rate; Operating conditions include high stability under low load, low stability under low load, or fluctuating load.

[0035] In this embodiment, the operational data includes water flow rate data and pipe pressure data for all pipes in the partition, water load data for all operating water pumps, and circulating water temperature data for the partition. Based on operational data, the flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load, and temperature deviation rate are calculated, including: Calculate the flow fluctuation coefficient of the partition based on the water flow data of all pipes in the partition; The deviation rate of the pipeline resistance coefficient is calculated based on the pipeline pressure data of all pipelines in the partition. The average pump load is calculated based on the pump load data of all operating pumps. The temperature deviation rate is calculated based on the circulating water temperature data in the partition.

[0036] In this embodiment, the flow fluctuation coefficient is a parameter reflecting the degree of change in water flow volume, used to characterize the stable state of water demand within a zone. The pipeline resistance coefficient deviation rate refers to the proportion of deviation between the actual pipeline resistance coefficient and the design reference value, used to reflect the degree of abnormality in pipeline resistance. The average pump load refers to the average operating intensity of pumps within a zone, used to characterize the load level of the zone. High-stability low-load type refers to a condition where the zone load is at a low level and parameters such as flow, resistance, and temperature are stable. Low-stability load type refers to a condition where the zone load is at a low level but at least one operating parameter exhibits slight fluctuations. Fluctuating load type refers to a condition where the zone operating parameters fluctuate drastically or the load is at a high level. Operating pumps refer to pumps that are currently in operation and not shut down.

[0037] Considering that the operating conditions of circulating water systems are affected by multiple factors, a comprehensive assessment using multi-dimensional parameters is necessary to ensure accuracy. This embodiment first selects the flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load, and temperature deviation rate, covering four core dimensions: water flow stability, pipeline status, equipment load, and heat exchange effect. This avoids the one-sidedness of judging by a single parameter; relying solely on load cannot reflect parameter fluctuations, and relying solely on flow rate cannot reflect abnormal pipeline resistance. This embodiment classifies operating conditions into high-stability low-load, low-stability load, and fluctuating load categories, reflecting the actual operating scenarios of industrial circulating water systems. Low-load scenarios exhibit differences between stability and slight fluctuations, while high-load or drastic fluctuation scenarios require separate classification, providing a basis for subsequently developing differentiated control strategies for different operating conditions. This embodiment obtains data based on spatial partition data association diagrams, ensuring precise binding of parameters to partitions and equipment, avoiding misjudgments of operating conditions caused by cross-partition data confusion, and laying the foundation for subsequent partitioned control.

[0038] For example, this embodiment can determine the operating condition type of each partition through the following steps: (1) Obtain the partition equipment and corresponding operating data from the spatial partition data association diagram. The operator logs into the circulating water system management platform and selects the target partition (such as the heat exchange zone) in the spatial partition data association diagram interface. In this embodiment, the water flow data, pipe pressure data, and pump load data of all pipes in the partition can be automatically retrieved, and the circulating water temperature data in the partition can be extracted simultaneously. In this embodiment, the retrieved data can be time-period calibrated to ensure that all data come from the same continuous monitoring period and avoid parameter calculation deviations caused by time asynchrony.

[0039] (2) Calculate the four core parameters based on operating data. In this embodiment, when calculating the flow fluctuation coefficient, the maximum, minimum and average values ​​of the water flow of all pipelines in the selected time period are first counted, and the degree of fluctuation is calculated through the correspondence of the three. In this embodiment, when calculating the deviation rate of the pipeline resistance coefficient, the actual pipeline resistance coefficient is calculated using pipeline pressure data, and then compared with the reference value of the resistance coefficient in the pipeline design document to obtain the deviation ratio. In this embodiment, when calculating the average load of the water pump, all operating water pumps in the zone are selected, and the load data of each water pump in the monitoring period are counted and the average value is taken. In this embodiment, when calculating the temperature deviation rate, the average value of the circulating water temperature in the zone in the monitoring period is first calculated, and then compared with the rated temperature of the core equipment (such as heat exchanger) in the zone to obtain the deviation ratio.

[0040] (3) Determine the zone's operating condition type by combining four parameters. In this embodiment, the calculated flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load, and temperature deviation rate can be compared with preset operating condition judgment criteria. If the average pump load is at a low level and the other three parameters are all in the stable range, it is judged as a high-stability low-load type; if the average pump load is at a low level, but at least one parameter exceeds the stable range and does not reach the drastic fluctuation threshold, it is judged as a low-stability load type; if any parameter reaches the drastic fluctuation threshold, or the average pump load is at a high level, it is judged as a fluctuating load type. After the judgment is completed, this embodiment can label the corresponding operating condition type label for the zone in the spatial zone data association diagram.

[0041] S104: Based on the operating conditions and data of each zone, determine the equipment control strategy through the target energy-saving control big model.

[0042] In this embodiment, the target energy-saving control model is obtained by training an initial energy-saving control model based on the historical operating conditions and historical operation data of each partition of multiple circulating water systems. The initial energy-saving control model includes an initialized input layer, an attention feature fusion layer, a Transformer encoding layer, a decoding layer, and a physical constraint layer.

[0043] In this embodiment, based on the operating condition type, operational data, and equipment distribution data of each partition, the equipment control strategy is determined through a target energy-saving control model, including: The operation data and device distribution data of each partition are preprocessed. The preprocessed device distribution data and operation data of each partition are then processed by the input layer to extract features, thereby obtaining the device distribution feature vector and operation feature vector of each partition. The attention feature fusion layer performs attention fusion on the working condition type, equipment distribution feature vector and operation feature vector of each partition to obtain the partition fusion feature vector; Based on the partitioned fusion feature vector, a deep semantic feature vector is obtained through a Transformer encoding layer; Based on deep semantic feature vectors, a preliminary control strategy vector is generated through a decoding layer; The equipment control strategy is obtained by modifying and formatting the initial control strategy vector through the physical constraint layer.

[0044] In this embodiment, the target energy-saving control model is used to achieve precise energy-saving regulation by zone. Historical operating condition type refers to the high-stability low-load, low-stability load, or fluctuating load status of each zone during past operation of multiple circulating water systems. Historical operating data refers to dynamic data such as water flow and pipeline pressure generated by equipment in each zone during past operation of multiple circulating water systems. Preprocessing refers to the operations of cleaning and standardizing the zone operating data and equipment distribution data. Equipment distribution feature vector refers to the vector extracted by the input layer from the preprocessed equipment distribution data, representing attributes such as equipment location and specifications. Operating feature vector refers to the vector extracted by the input layer from the preprocessed operating data, representing the dynamic operating state of the equipment. Zone fusion feature vector refers to the vector output by the attention feature fusion layer, fusing operating condition type, equipment distribution features, and operating features. Deep semantic feature vector refers to the vector output by the Transformer encoding layer, containing deep correlations of multi-dimensional information about the zones. Equipment regulation strategy refers to the final strategy output by the physical constraint layer, which can be directly used to control the equipment in the circulating water system.

[0045] Considering that the regulation of circulating water systems needs to balance model adaptability and strategy feasibility, this embodiment trains the model using historical data from multiple circulating water systems. This avoids insufficient generalization ability caused by data from a single system and ensures that the model is adaptable to the operating scenarios of different enterprises and different zones. The initial model is set with a multi-level structure to construct a complete logical chain of data processing, feature fusion, deep encoding, strategy generation, and compliance correction. The input layer extracts basic features to solve the data format adaptation problem; the attention fusion layer highlights key information and avoids interference from irrelevant features; the Transformer encoding layer mines deep correlations to improve the policy's targeting; the decoding layer generates a preliminary policy and provides a prototype for regulation; and the physical constraint layer ensures compliance and prevents the policy from exceeding equipment or physical limits. At the same time, this embodiment combines zone operating condition types and equipment distribution data to allow the policy to fit the characteristics of the zone, solving the problems of poor adaptability and blind policy of existing general models, and ensuring that regulation is both energy-saving and meets the actual operating needs of the system.

[0046] For example, this embodiment can determine the equipment control strategy through the following steps.

[0047] (1) Preprocess the operation data and equipment distribution data of each partition. In this embodiment, the operation data (water flow rate, pipeline pressure, water pump load, circulating water temperature) and equipment distribution data (water pump location and specifications, pipeline diameter and length, etc.) of each partition can be retrieved from the spatial partition data association diagram. Industrial data processing software is used to clean the data: remove the load data when the water pump is stopped and the invalid data with zero pipeline pressure, and delete duplicate records in the equipment distribution data. The cleaned data is standardized: the operation data is standardized according to the industry common range (such as pipeline pressure 0-1MPa), and the discrete attributes (such as equipment type) in the equipment distribution data are converted into numerical codes to ensure that the data format meets the model input requirements. The processed data is stored in the model input database.

[0048] (2) Extracting feature vectors through the input layer. In this embodiment, the preprocessed equipment distribution data and operation data can be input into the input layer of the target energy-saving control model. The input layer adopts a fully connected network structure: for the equipment distribution data, the attributes such as equipment location and specifications are transformed into low-dimensional vectors through linear transformation, and then the key features are enhanced by the ReLU activation function to generate equipment distribution feature vectors; for the operation data, the same fully connected network structure is used to transform dynamic parameters such as water flow and pipeline pressure into operation feature vectors with the same dimension as the equipment distribution feature vectors. The two types of vectors are transmitted to the attention feature fusion layer simultaneously.

[0049] (3) Obtain the partition fusion feature vector through the attention feature fusion layer. In this embodiment, the operating condition type of each partition (high stability low load, low stability load, and fluctuating load are converted into corresponding numerical codes), equipment distribution feature vector, and operation feature vector are input into the attention feature fusion layer: This layer calculates the correlation weight of the three. For example, under the fluctuating load condition, the operation feature vector is assigned a higher weight, and the equipment distribution feature vector is assigned a lower weight. Based on the weight, the three are linearly fused. During the fusion process, layer normalization is used to ensure the stability of the vector values. Finally, a partition fusion feature vector containing the partition operating condition, equipment characteristics, and real-time status is generated and transmitted to the Transformer encoding layer.

[0050] (4) Obtain the deep semantic feature vector through the Transformer encoding layer. In this embodiment, the partitioned fused feature vector can be input into the Transformer encoding layer. This layer adopts a multi-head attention mechanism, with 8 attention heads capturing the correlations in different dimensions (e.g., head 1 captures the correlation between pump specifications and load, and head 2 captures the correlation between pipe diameter and pressure). The original fused features are superimposed with the attention output through residual connections to avoid loss of feature information. Then, the features are processed in depth by a feedforward network (composed of linear transformation and ReLU activation function) to explore the potential correlations between working condition type and control requirements, equipment characteristics and operating parameters. Finally, a deep semantic feature vector containing complete semantic information is output and passed to the decoding layer.

[0051] (5) Generate a preliminary control strategy vector through the decoding layer. In this embodiment, the deep semantic feature vector can be input into the decoding layer, which adopts a branch structure design according to the working condition type. For high-stability low-load partitions, a 2-layer fully connected branch is used to generate simple control parameters such as pump frequency adjustment value and single pump operation selection; for low-stability load partitions, a 3-layer fully connected branch is used to generate correction parameters such as frequency fine-tuning amplitude and pressure compensation coefficient; for fluctuating load partitions, a 4-layer fully connected branch is used to generate complex parameters such as the number of pump combinations, frequency allocation of each pump, and valve opening adjustment value; the control parameters output by all branches are integrated into a preliminary control strategy vector, which contains various parameters that the equipment needs to adjust and is temporarily stored in the model intermediate result library.

[0052] (6) Obtain the equipment control strategy through the physical constraint layer. In this embodiment, the preliminary control strategy vector can be input into the physical constraint layer. This layer calls the preset physical rule library (such as the pipeline pressure not exceeding the equipment rated pressure, and the water pump frequency adjustment range not exceeding 5Hz) and the equipment parameter library (such as the water pump rated power and the valve maximum opening) to verify each parameter in the preliminary strategy. If a parameter exceeds the constraint range (such as the water pump frequency exceeding the rated frequency), it is corrected according to the principle of proximity constraint (the frequency is adjusted to the upper limit of the rated frequency). After correction, the parameter is formatted into an instruction form that can be recognized by industrial equipment (such as water pump 01: frequency adjusted to 42Hz; pipeline valve 03: opening adjusted to 80%), and finally the equipment control strategy is generated and transmitted to the circulating water system control terminal for subsequent equipment control.

[0053] S105: Controlling the circulating water system based on equipment regulation strategies.

[0054] In this embodiment, the equipment control strategy refers to the set of specific instructions output by the target energy-saving control model that can be directly used to adjust the operating status of the circulating water system equipment, including parameters such as pump frequency and valve opening. The circulating water system refers to an auxiliary system used in industry for equipment cooling or heat exchange, consisting of pumps, pipes, cooling equipment, etc. Control refers to the operation of adjusting the operating status of the equipment within the system according to the equipment control strategy, so that the system operates under expected conditions.

[0055] The consideration behind this embodiment is to translate the strategies generated by the target energy-saving control model into actual operations, ensuring that the control measures are implemented effectively. Simply generating strategies cannot achieve energy savings; control mechanisms are needed to make the equipment act according to the strategies. At the same time, the control must accurately correspond to the equipment in each zone to avoid a disconnect between strategy and execution, ultimately achieving the goal of system energy saving and stable operation.

[0056] For example, this embodiment can control the circulating water system through the following steps.

[0057] (1) The control strategy is parsed into equipment instructions. In this embodiment, the control strategy can be split according to the equipment type (pump, valve, and cooling equipment, etc.). For pumps, the frequency value to be adjusted and the number of pumps in operation are extracted and converted into digital instructions that can be recognized by the pump controller; for valves, the opening value is extracted and converted into analog signal instructions adapted to the valve actuator; after parsing, this embodiment can mark each instruction with a unique identifier corresponding to the equipment (such as pump 01, valve 03) to avoid incorrect instruction delivery.

[0058] (2) Sending instructions and controlling equipment execution. In this embodiment, the parsed instructions can be sent to the controllers of the corresponding equipment via industrial communication links (such as Modbus-RTU, Profinet). For example, after receiving the frequency instruction, the water pump controller adjusts the output frequency of the frequency converter and drives the water pump to run at the new frequency; after receiving the opening instruction, the valve actuator drives the valve stem to rotate through the motor and adjusts the valve to the specified opening; after receiving the instruction, the cooling equipment controller adjusts parameters such as the fan speed to ensure that the equipment operates according to the strategy.

[0059] (3) Feedback on operating status and handling of anomalies. After the equipment executes the command, it can collect operating status data (such as the actual frequency of the water pump and the actual opening degree of the valve) in real time through sensors and send it back. In this embodiment, the feedback data can be compared with the target parameters in the control strategy. If they are consistent, the control is deemed effective. If the deviation exceeds the allowable range (such as the water pump frequency not reaching the target value), this embodiment can resend the command. If it fails multiple times, an alarm will be triggered to notify the operation and maintenance personnel to troubleshoot the equipment and ensure the control effect.

[0060] As can be seen from the above, existing technologies mostly adopt global unified control, ignoring the operational differences of equipment in different areas. This embodiment obtains equipment distribution data, divides dedicated areas according to the spatial layout of equipment, and constructs a correlation diagram corresponding to the area, equipment, and operational data. It accurately analyzes the actual operating status of each area, avoiding the problems of over-controlling low-load areas and under-controlling high-load areas.

[0061] Traditional general-purpose models cannot be adapted to the specific equipment conditions of enterprises, making it difficult to match actual operational needs. The energy-saving control in this embodiment not only relies on the enterprise's own equipment distribution data but also integrates real-time operational data, making it more suitable for the enterprise's equipment characteristics and operating habits, and significantly improving the adaptability of the control.

[0062] This embodiment accurately determines the operating condition type of each area, ensuring precise matching between the control strategy and the operating status. This minimizes ineffective energy consumption by water pumps, significantly improves energy efficiency, and helps enterprises reduce production costs. Because the control strategy is based on actual regional data and operating condition types, frequent adjustments to equipment operating parameters are avoided. This effectively reduces the start-up and shutdown impacts and load fluctuations of core equipment such as water pumps, ensuring the continuous and stable operation of the circulating water system, preventing production processes from being affected by improper control, reducing equipment wear, extending equipment lifespan, and reducing maintenance and replacement costs for enterprises.

[0063] In one embodiment of this application, the operating condition type of the zone is determined based on the flow fluctuation coefficient, the pipeline resistance coefficient deviation rate, the average pump load, and the temperature deviation rate, including: If the flow fluctuation coefficient, the deviation rate of the pipeline resistance coefficient, the average pump load, and the temperature deviation rate meet the first condition, then the operating condition type of this zone is determined to be high stability and low load. If the flow fluctuation coefficient, the deviation rate of the pipeline resistance coefficient, the average pump load, and the temperature deviation rate meet the second condition, then the operating condition type of this zone is determined to be low stable load. If the flow fluctuation coefficient, the deviation rate of the pipeline resistance coefficient, the average pump load, and the temperature deviation rate meet the third condition, then the operating condition type of this zone is determined to be the fluctuating load type. The first condition is that the average load of the water pump is less than the target load threshold, the flow fluctuation coefficient is less than the first fluctuation coefficient threshold, the absolute value of the deviation rate of the pipeline resistance coefficient is less than the first resistance deviation rate threshold, and the absolute value of the temperature deviation rate is less than the first temperature deviation rate threshold. The second condition is that the average load of the water pump is less than the target load threshold, and at least one of the following conditions is met: the flow fluctuation coefficient is not less than the first fluctuation coefficient threshold and less than the second fluctuation coefficient threshold; the absolute value of the pipeline resistance coefficient deviation rate is not less than the first resistance deviation rate threshold and less than the second resistance deviation rate threshold; and the absolute value of the temperature deviation rate is not less than the first temperature deviation rate threshold and less than the second temperature deviation rate threshold. The third condition is that the flow fluctuation coefficient is not less than the second fluctuation coefficient threshold, the absolute value of the pipeline resistance coefficient deviation rate is not less than the second resistance deviation rate threshold, the absolute value of the temperature deviation rate is not less than the second temperature deviation rate threshold, and the average pump load is not less than the target load threshold.

[0064] In this embodiment, the target load threshold refers to the benchmark value used to distinguish between high and low loads in a zone, and is the core load boundary for determining high-stability low-load, low-stability load, and fluctuating load categories. The first fluctuation coefficient threshold refers to the benchmark value used to distinguish between stable flow and slight fluctuations; a value below this indicates stable flow. The second fluctuation coefficient threshold refers to the benchmark value used to distinguish between slight and severe flow fluctuations; a value not lower than this indicates severe flow fluctuations. The first resistance deviation rate threshold refers to the benchmark value used to distinguish between normal and slightly abnormal network resistance; an absolute value below this indicates normal resistance. The second resistance deviation rate threshold refers to the benchmark value used to distinguish between slightly abnormal and severely abnormal network resistance; an absolute value not lower than this indicates severely abnormal resistance. The first temperature deviation rate threshold refers to the benchmark value used to distinguish between normal and slightly deviated circulating water temperature; an absolute value below this indicates normal temperature. The second temperature deviation rate threshold refers to the benchmark value used to distinguish between slightly and severely deviated circulating water temperature; an absolute value not lower than this indicates severely deviated temperature. The first condition refers to the parameter combination standard for determining a zone as a high-stability low-load category. The second condition refers to the parameter combination standard for determining a zone as a low-stability load category. The third condition refers to the parameter combination criteria for determining whether a zone is a fluctuating load type.

[0065] This embodiment achieves precise classification of operating conditions through hierarchical thresholds and multi-parameter combinations, avoiding misjudgments caused by single parameters or thresholds. This embodiment first locks in low-load scenarios by setting a target load threshold, and then further subdivides high stability (all parameters normal) and low stability (minor local anomalies) using first fluctuation, resistance, and temperature thresholds. This not only aligns with the two typical states of low-load industrial operation but also avoids misjudging minor fluctuations as severe fluctuations. This embodiment sets a second threshold to define severe fluctuations / serious anomalies, and directly classifies high loads as fluctuating loads because high loads result in high system energy consumption and high parameter sensitivity; even if short-term stability is achieved, sudden fluctuations are likely, requiring close monitoring and dynamic control. The multi-parameter and / or combined condition design of this embodiment ensures the stringency of high-stability operating conditions (all parameters meet standards) while also accommodating the flexibility of low-stability and fluctuating operating conditions (local parameter triggering), providing precise operating condition basis for subsequent differentiated energy-saving control and solving the problems of ambiguous operating condition classification and insufficient targeted control in existing technologies.

[0066] For example, this embodiment can determine the operating condition type of the partition through the following steps.

[0067] (1) Obtain the core parameters of the partition to be judged. In this embodiment, the flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load and temperature deviation rate of the partition can be retrieved from the circulating water system data platform. Ensure that the four parameters are all from the same continuous monitoring period (such as the last 5 minutes), and that the data has been preprocessed to remove outliers. The parameter format is uniformly in percentage form (retaining two decimal places).

[0068] (2) Set each judgment threshold. In this embodiment, the thresholds can be determined by referring to the economic operation specifications of industrial circulating water system and the historical operation data of the enterprise. For example, the target load threshold is set in combination with the average load during the low load operation period of the enterprise; the first fluctuation coefficient threshold, the first resistance deviation rate threshold, and the first temperature deviation rate threshold are set according to the normal parameter fluctuation range allowed by the equipment; the second fluctuation coefficient threshold, the second resistance deviation rate threshold, and the second temperature deviation rate threshold are set according to the critical fluctuation range of safe operation of the equipment. After all thresholds are entered, they are stored in the working condition judgment rule library to support subsequent adjustments as needed.

[0069] (3) Compare the judgment conditions sequentially. In this embodiment, the first condition can be compared to determine whether the average load of the water pump is less than the target load threshold and whether the flow fluctuation coefficient is less than the first fluctuation coefficient threshold. At the same time, the absolute value of the deviation rate of the pipeline resistance coefficient is less than the first resistance deviation rate threshold and the absolute value of the temperature deviation rate is less than the first temperature deviation rate threshold. If all four conditions are met, it is judged as a high-stability low-load type. The operating condition type is marked on the system interface and the judgment time is recorded. If the first condition is not met, the second condition is compared. First, it is confirmed that the average load of the water pump is still less than the target load threshold. Then, it is judged whether any one of the following conditions is met: the flow fluctuation coefficient is between the first and second fluctuation coefficient thresholds, the absolute value of the deviation rate of the pipeline resistance coefficient is between the first and second resistance deviation rate thresholds, and the absolute value of the temperature deviation rate is between the first and second temperature deviation rate thresholds. If any one of these conditions is met, it is judged as a low-stability load type and the condition is marked and recorded simultaneously. If the first two conditions are not met, then compare with the third condition to determine if any of the following conditions are met: the flow fluctuation coefficient is not less than the second fluctuation coefficient threshold, the absolute value of the pipeline resistance coefficient deviation rate is not less than the second resistance deviation rate threshold, the absolute value of the temperature deviation rate is not less than the second temperature deviation rate threshold, or the average pump load is not less than the target load threshold. If any of these conditions are met, the load is classified as fluctuating load, and the labeling and recording are completed.

[0070] This embodiment uses multi-threshold stratification and multi-parameter combination to determine operating condition types, accurately distinguishing between high-stability low-load, low-stability load, and fluctuating load types, avoiding control deviations caused by the ambiguity of operating condition classification in existing technologies. The stringent determination criteria for high-stability operating conditions ensure that lightweight energy-saving strategies (such as historical energy efficiency matching) can be adopted subsequently, reducing redundant control energy consumption. Local anomaly identification for low-stability operating conditions can trigger targeted correction strategies to prevent anomaly escalation. The multi-trigger condition design for fluctuating operating conditions can promptly capture high load or severe fluctuation risks, adapt to dynamic optimization strategies, and ensure system safety. Ultimately, this achieves precise operating condition determination and differentiated control strategies, improving the energy efficiency and operational stability of the circulating water system and reducing equipment operation and maintenance costs.

[0071] Corresponding to the circulating water energy-saving control method in the above embodiments, Figure 2 This is a structural block diagram of a circulating water energy-saving control device provided according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The circulating water energy-saving control device 20 includes: a data acquisition module 21, a partitioning module 22, an operating condition analysis module 23, a strategy generation module 24, and a control module 25.

[0072] The data acquisition module 21 is used to acquire equipment distribution data and operation data of the circulating water system. The operation data includes water flow, pipeline pressure, water pump load and circulating water temperature. The equipment distribution data includes water pump distribution data, pipeline distribution data, cooling equipment distribution data, heat exchange equipment distribution data and water storage and water treatment equipment distribution data. The partitioning module 22 is used to construct a spatial partitioning data association diagram based on device distribution data and operational data. The spatial partitioning data association diagram includes various partitions divided according to device distribution data, and each partition includes multiple devices and their respective operational data. The operating condition analysis module 23 is used to determine the operating condition type of each partition based on the spatial partition data association diagram; The strategy generation module 24 is used to determine the equipment control strategy based on the operating condition type and operation data of each partition through the target energy-saving control big model. Control module 25 is used to control the circulating water system based on equipment regulation strategies.

[0073] In one embodiment of this application, the partitioning module 22 is specifically used for: Determine the equipment type and spatial coordinates of all equipment based on equipment distribution data; Based on the equipment type and spatial coordinates of all devices, the circulating water system is divided into multiple zones using the K-means clustering algorithm; these zones include a heat exchange zone, a cooling zone, a water storage zone, and a water pipeline zone. A device code identifier is generated for each device in each partition. The device code identifier includes the device ID, the partition to which the device belongs, and the device's spatial coordinates. Align the running data with each device to obtain the correspondence between all device codes and all running data; A spatial partition data association diagram is constructed based on the correspondence between all device codes and all operational data.

[0074] In one embodiment of this application, the working condition analysis module 23 is specifically used for: For each partition, obtain the corresponding equipment and the operating data of each equipment from the spatial partition data association diagram; calculate the flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load and temperature deviation rate based on the operating data; determine the operating condition type of the partition based on the flow fluctuation coefficient, pipeline resistance coefficient deviation rate, average pump load and temperature deviation rate; Operating conditions include high stability under low load, low stability under low load, or fluctuating load.

[0075] In one embodiment of this application, the operating data includes water flow rate data and pipe pressure data for all pipes in the partition, pump load data for all operating pumps, and circulating water temperature data for the partition; the operating condition analysis module 23 is further used for: Calculate the flow fluctuation coefficient of the partition based on the water flow data of all pipes in the partition; The deviation rate of the pipeline resistance coefficient is calculated based on the pipeline pressure data of all pipelines in the partition. The average pump load is calculated based on the pump load data of all operating pumps. The temperature deviation rate is calculated based on the circulating water temperature data in the partition.

[0076] In one embodiment of this application, the working condition analysis module 23 is further used for: If the flow fluctuation coefficient, the deviation rate of the pipeline resistance coefficient, the average pump load, and the temperature deviation rate meet the first condition, then the operating condition type of this zone is determined to be high stability and low load. If the flow fluctuation coefficient, the deviation rate of the pipeline resistance coefficient, the average pump load, and the temperature deviation rate meet the second condition, then the operating condition type of this zone is determined to be low stable load. If the flow fluctuation coefficient, the deviation rate of the pipeline resistance coefficient, the average pump load, and the temperature deviation rate meet the third condition, then the operating condition type of this zone is determined to be the fluctuating load type. The first condition is that the average load of the water pump is less than the target load threshold, the flow fluctuation coefficient is less than the first fluctuation coefficient threshold, the absolute value of the deviation rate of the pipeline resistance coefficient is less than the first resistance deviation rate threshold, and the absolute value of the temperature deviation rate is less than the first temperature deviation rate threshold. The second condition is that the average load of the water pump is less than the target load threshold, and at least one of the following conditions is met: the flow fluctuation coefficient is not less than the first fluctuation coefficient threshold and less than the second fluctuation coefficient threshold; the absolute value of the pipeline resistance coefficient deviation rate is not less than the first resistance deviation rate threshold and less than the second resistance deviation rate threshold; and the absolute value of the temperature deviation rate is not less than the first temperature deviation rate threshold and less than the second temperature deviation rate threshold. The third condition is that the flow fluctuation coefficient is not less than the second fluctuation coefficient threshold, the absolute value of the pipeline resistance coefficient deviation rate is not less than the second resistance deviation rate threshold, the absolute value of the temperature deviation rate is not less than the second temperature deviation rate threshold, and the average pump load is not less than the target load threshold.

[0077] In one embodiment of this application, the target energy-saving control model is trained based on the historical operating conditions and historical operation data of each zone of multiple circulating water systems, and is obtained through the following method: The initial energy-saving control model is trained based on the historical operating conditions and historical operation data of each zone of multiple circulating water systems to obtain the target energy-saving control model. The initial energy-saving control model includes an initialized input layer, an attention feature fusion layer, a Transformer encoding layer, a decoding layer, and a physical constraint layer.

[0078] In one embodiment of this application, the strategy generation module 24 is specifically used for: The operation data and device distribution data of each partition are preprocessed. The preprocessed device distribution data and operation data of each partition are then processed by the input layer to extract features, thereby obtaining the device distribution feature vector and operation feature vector of each partition. The attention feature fusion layer performs attention fusion on the working condition type, equipment distribution feature vector and operation feature vector of each partition to obtain the partition fusion feature vector; Based on the partitioned fusion feature vector, a deep semantic feature vector is obtained through a Transformer encoding layer; Based on deep semantic feature vectors, a preliminary control strategy vector is generated through a decoding layer; The equipment control strategy is obtained by modifying and formatting the initial control strategy vector through the physical constraint layer.

[0079] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the data acquisition module 21, partitioning module 22, operating condition analysis module 23, strategy generation module 24, and control module 25 are shown.

[0080] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0081] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0082] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0083] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation methods described in the embodiments of the circulating water energy-saving control method provided in the embodiments of this application, or they can execute the implementation methods of the electronic device 300 described in the embodiments of this application, which will not be repeated here.

[0084] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0085] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD) card, flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0086] Those skilled in the art will recognize that the modules / units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0087] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0088] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules, units, or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or modules / units, or it may be an electrical, mechanical, or other form of connection.

[0089] The modules / units described as separate components may or may not be physically separate. Similarly, the components shown as modules / units may or may not be physical modules / units; they may be located in one place or distributed across multiple network modules / units. Some or all of the modules / units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0090] Furthermore, the functional modules / units in the various embodiments of this application can be integrated into one processing module / unit, or each module / unit can exist physically separately, or two or more modules / units can be integrated into one module / unit. The integrated modules / units described above can be implemented in hardware or in the form of software functional modules / units.

[0091] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A circulating water energy saving control method, characterized by, The method comprises the following steps: obtaining equipment distribution data and operation data of a circulating water system; the operation data comprises water flow, pipeline pressure, water pump load and circulating water temperature, and the equipment distribution data comprises water pump distribution data, pipeline distribution data, cooling equipment distribution data, heat exchange equipment distribution data, and water storage and water treatment equipment distribution data; constructing a spatial partition data association graph based on the equipment distribution data and the operation data; the spatial partition data association graph comprises each partition divided according to the equipment distribution data, and each partition comprises a plurality of equipment and respective operation data of the plurality of equipment; determining the working condition type of each partition based on the spatial partition data association graph; determining an equipment control strategy through a target energy-saving control large model based on the working condition type of each partition, the operation data and the equipment distribution data of each partition; controlling the circulating water system based on the equipment control strategy.

2. The circulating water energy saving control method according to claim 1, wherein The method comprises the following steps: determining the equipment type and spatial coordinates of all equipment based on the equipment distribution data; dividing the circulating water system into a plurality of partitions through a K-means clustering algorithm based on the equipment type and spatial coordinates of all equipment; the plurality of partitions comprise a heat exchange zone, a cooling zone, a water storage zone and a water conveying pipeline zone; generating an equipment coding identifier for the equipment of each partition; the equipment coding identifier comprises an equipment ID, a partition to which the equipment belongs and equipment spatial coordinates; aligning the operation data with each equipment respectively to obtain the correspondence between all equipment coding identifiers and all operation data; constructing a spatial partition data association graph based on the correspondence between all equipment coding identifiers and all operation data.

3. The circulating water energy saving control method according to claim 1, wherein The method comprises the following steps: for each partition, obtaining the equipment corresponding to the partition and the operation data of each equipment from the spatial partition data association graph; calculating a flow fluctuation coefficient, a pipeline resistance coefficient deviation rate, a water pump average load and a temperature deviation rate based on the operation data; determining the working condition type of the partition based on the flow fluctuation coefficient, the pipeline resistance coefficient deviation rate, the water pump average load and the temperature deviation rate; the working condition type comprises stable low load type, fluctuating low load type or fluctuating high load type.

4. The circulating water energy saving control method according to claim 3, wherein The operation data comprises water flow data and pipeline pressure data of all pipelines in the partition, water pump load data of all running water pumps and circulating water temperature data in the partition; the method comprises the following steps: calculating the flow fluctuation coefficient of the partition based on the water flow data of all pipelines in the partition; calculating the pipeline resistance coefficient deviation rate based on the pipeline pressure data of all pipelines in the partition; calculating the water pump average load based on the water pump load data of all running water pumps; calculating the temperature deviation rate based on the circulating water temperature data in the partition.

5. The circulating water energy saving control method according to claim 4, wherein the method comprises the following steps: If the flow fluctuation coefficient, the pipe network resistance coefficient deviation rate, the water pump average load and the temperature deviation rate satisfy a first condition, it is determined that the working condition type of the subzone is stable low load type; If the flow fluctuation coefficient, the pipe network resistance coefficient deviation rate, the water pump average load and the temperature deviation rate satisfy a second condition, it is determined that the working condition type of the subzone is fluctuation low load type; If the flow fluctuation coefficient, the pipe network resistance coefficient deviation rate, the water pump average load and the temperature deviation rate satisfy a third condition, it is determined that the working condition type of the subzone is fluctuation high load type; The first condition is that the water pump average load is less than a target load threshold, the flow fluctuation coefficient is less than a first fluctuation coefficient threshold, the absolute value of the pipe network resistance coefficient deviation rate is less than a first resistance deviation rate threshold, and the absolute value of the temperature deviation rate is less than a first temperature deviation rate threshold; The second condition is that the water pump average load is less than the target load threshold, and at least one of the following conditions is satisfied: the flow fluctuation coefficient is not less than the first fluctuation coefficient threshold and is less than a second fluctuation coefficient threshold, the absolute value of the pipe network resistance coefficient deviation rate is not less than the first resistance deviation rate threshold and is less than a second resistance deviation rate threshold, and the absolute value of the temperature deviation rate is not less than the first temperature deviation rate threshold and is less than a second temperature deviation rate threshold; The third condition is that at least one of the following conditions is satisfied: the flow fluctuation coefficient is not less than the second fluctuation coefficient threshold, the absolute value of the pipe network resistance coefficient deviation rate is not less than the second resistance deviation rate threshold, the absolute value of the temperature deviation rate is not less than the second temperature deviation rate threshold, and the water pump average load is not less than the target load threshold.

6. The circulating water energy saving control method as claimed in claim 1, wherein, The target energy-saving control large model is obtained by training based on historical working condition types and historical running data of each subzone of a plurality of circulating water systems in the following manner: An initial energy-saving control large model is trained based on historical working condition types and historical running data of each subzone of a plurality of circulating water systems to obtain a target energy-saving control large model; the initial energy-saving control large model includes an initialized input layer, an attention feature fusion layer, a Transformer encoding layer, a decoding layer and a physical constraint layer.

7. The circulating water energy saving control method of claim 1, wherein, The target energy-saving control large model includes an input layer, a feature fusion layer, a Transformer encoding layer, a decoding layer and a physical constraint layer; Based on the working condition types, the running data and the device distribution data of each subzone, a device control strategy is determined by the target energy-saving control large model, including: The running data and the device distribution data of each subzone are preprocessed, and the preprocessed device distribution data and running data of each subzone are subjected to feature extraction by the input layer to obtain device distribution feature vectors and running feature vectors of each subzone, respectively; The working condition types, the device distribution feature vectors and the running feature vectors of each subzone are subjected to attention fusion by the attention feature fusion layer to obtain subzone fusion feature vectors; Based on the subzone fusion feature vectors, deep semantic feature vectors are obtained by the Transformer encoding layer; Based on the deep semantic feature vector, a preliminary regulation strategy vector is generated through the decoding layer; The preliminary regulation strategy vector is corrected and formatted through the physical constraint layer to obtain a device regulation strategy.

8. A circulating water energy saving control device, characterized by, Comprise: A data acquisition module is configured to acquire device distribution data and operation data of a circulating water system; the operation data includes water flow, pipeline pressure, water pump load and circulating water temperature, and the device distribution data includes water pump distribution data, pipeline distribution data, cooling device distribution data, heat exchange device distribution data, and water storage and water treatment device distribution data; A partition module is configured to construct a spatial partition data association graph based on the device distribution data and the operation data; the spatial partition data association graph includes each partition divided according to the device distribution data, and each partition includes a plurality of devices and respective operation data of the plurality of devices; A working condition analysis module is configured to determine the working condition type of each partition based on the spatial partition data association graph; A strategy generation module is configured to determine a device regulation strategy through a target energy-saving control large model based on the working condition type and operation data of each partition; A control module is configured to control the circulating water system based on the device regulation strategy.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to realize the steps of the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to realize the steps of the method of any one of claims 1 to 7.

Images