Benefit evaluation model and construction method of coal-fired unit coupled with molten salt energy storage system

Abstract

The present application relates to the technical field of energy system evaluation, and more particularly to a coal-fired unit and molten salt energy storage coupling system benefit evaluation model and construction method, comprising: constructing a multi-source data processing module, calibrating and time-space registering multi-source operation data of core components of the coupling system, forming a space-condition-operation parameter three-dimensional data cube; constructing a multi-dimensional feature extraction module, extracting energy efficiency, component collaborative coupling and time sequence dynamic response features of each core evaluation node; constructing a feature fusion and mapping module, generating a multi-dimensional coupling feature map associated with space-condition; constructing a distribution map generation module, outputting a space-condition distribution map of key benefit indicators; constructing an evaluation optimization module, identifying core components and condition intervals with benefit attenuation, quantifying the influence of local short boards on overall benefit, and generating evaluation results and optimization suggestions. Through the present application, the spatial heterogeneity and condition dynamic fluctuation of system benefit can be captured, and the short board can be located and quantified.

Description

Technical Field

[0001] This invention relates to the technical field of energy system assessment, and in particular to a benefit assessment model and construction method for a coal-fired power unit coupled with molten salt energy storage system. Background Technology

[0002] Coupling coal-fired power units with molten salt energy storage systems, as a novel energy-saving and consumption-reducing technology, can mitigate load fluctuations and enhance the absorption capacity of renewable energy through molten salt energy storage, demonstrating significant application value in energy transition scenarios. Currently, the typical method for evaluating the benefits of this system is to calculate comprehensive indicators such as energy efficiency improvement, energy consumption reduction, and economic benefits throughout the entire lifecycle based on the overall operating data of the unit and energy storage device, thereby forming an overall benefit evaluation conclusion.

[0003] However, in the spatial dimension, the coupling and coordination efficiency of core components such as boilers, turbines, molten salt storage tanks, and heat exchange units in coal-fired power units varies. Problems such as heat exchange losses and energy storage leakage in local components can lead to a significantly higher rate of regional benefit decay than in other parts. In the operating condition dimension, the charging and discharging response speed of molten salt energy storage and its coordination matching degree with the unit dynamically fluctuate under different operating conditions such as rated load, peak load, and start-up / shutdown transition, resulting in an uneven distribution of benefits across different operating condition ranges. Existing evaluation models use overall data homogenization for calculation, which simplifies the calculation process but inevitably smooths out the differences in the coordination benefits of various components in the spatial dimension and the dynamic benefit fluctuations in the operating condition dimension. It can only give the average benefit level of the system as a whole. It cannot accurately locate the key components with spatial benefit decay, the optimal / worst operating range under the operating conditions, or quantify the impact of local shortcomings on the overall benefit. This results in insufficient guidance value for fault warning, optimization and transformation, and operating condition adaptation of coupled systems. Summary of the Invention

[0004] This invention provides a benefit evaluation model and construction method for a coal-fired power unit coupled with molten salt energy storage system, which can effectively solve the problems in the background art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for constructing a benefit evaluation model for a coal-fired power unit coupled with molten salt energy storage system, comprising: A multi-source data processing module is constructed to perform data calibration and spatiotemporal registration on the multi-source operating data of the core components of the coal-fired power unit and molten salt energy storage coupling system under different operating conditions, and to construct a three-dimensional data cube containing spatial dimension, operating condition dimension and operating parameter dimension. A multi-dimensional feature extraction module is constructed to extract energy efficiency features, component collaborative coupling features, and time-series dynamic response features under different operating conditions from each core evaluation node in the three-dimensional data cube. A feature fusion and mapping module is constructed to bidirectionally map the same type of node-level features to their spatial location and corresponding working condition information, generating a multi-dimensional coupled feature map that integrates spatial-working condition relationships. A distribution map generation module is constructed to output a spatial-condition distribution map of at least one key benefit indicator per node and per condition, based on a multi-dimensional coupled feature map and a benchmark feature library of each core component and working condition, through multi-dimensional feature correlation analysis. An evaluation and optimization module is constructed to identify core components and operating condition ranges with abnormal benefit attenuation in the coupled system based on the spatial-operating condition distribution map and preset evaluation benchmarks, quantify the impact of local shortcomings on overall benefits, and generate evaluation results that include overall benefit assessment and local optimization suggestions.

[0006] Furthermore, the core components include boilers, steam turbines, molten salt storage tanks, heat exchange units, and auxiliary equipment.

[0007] Furthermore, spatiotemporal registration achieves temporal dimension alignment based on data timestamps and spatial dimension association based on the spatial installation coordinates of core components.

[0008] Furthermore, the spatial dimension of the three-dimensional data cube corresponds to the physical installation location of the core components and key monitoring points, the operating condition dimension includes rated load, peak load, start-stop transition and other preset operating conditions, and the operating parameter dimension includes multi-source operating data after calibration and registration.

[0009] Furthermore, the core evaluation nodes are divided according to the physical structure of the core components and the distribution of monitoring points. A single core evaluation node corresponds to a specific monitoring area of ​​a single component or a collaborative interaction area of ​​multiple components.

[0010] Furthermore, the energy efficiency characteristics are calculated based on node energy consumption data and output energy data, the component collaborative coupling characteristics are calculated based on the difference in operating parameter response between associated nodes, and the time-series dynamic response characteristics are extracted based on the sequence of operating parameter changes of the same node during the switching of operating conditions.

[0011] Furthermore, the bidirectional mapping includes spatial dimension mapping and working condition dimension mapping; spatial dimension mapping associates the same type of features of core evaluation nodes in different spatial locations under the same working condition according to spatial topological relationships; working condition dimension mapping associates the same type of features of core evaluation nodes in the same spatial location according to the working condition switching order.

[0012] Furthermore, multi-dimensional feature correlation analysis includes: By comparing the actual characteristics of the core evaluation nodes with the corresponding characteristics in the benchmark feature library, and combining the working condition weight coefficient and the spatial weight coefficient, the actual values ​​of the key benefit indicators under the corresponding working conditions of each node are calculated.

[0013] Furthermore, the determination of abnormal benefit decline is based on the degree of deviation between the actual value of the key benefit indicator and the corresponding preset evaluation benchmark. When the degree of deviation exceeds the preset threshold, it is determined to be abnormal benefit decline.

[0014] On the other hand, the present invention also provides a benefit evaluation model for a coal-fired power unit coupled with a molten salt energy storage system, which is constructed using the above-mentioned method for constructing a benefit evaluation model for a coal-fired power unit coupled with a molten salt energy storage system.

[0015] The technical solution of this invention can achieve the following technical effects: By constructing a three-dimensional data cube, the system integrates spatial, operational, and parameter data across all dimensions. Combined with multi-dimensional feature extraction and bidirectional mapping fusion, the generated spatial-operational distribution map clearly presents the differences and dynamic fluctuations in the efficiency of each core component under different operating conditions. This not only accurately locates key components with diminishing efficiency in space and the optimal / worst operating ranges under different operating conditions, but also quantifies the impact of local shortcomings on overall efficiency. This provides data support and scientific basis for fault warning, precise optimization and transformation, and operational condition adaptation adjustments of coupled systems, thereby enhancing the guiding value of efficiency assessment results.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, 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 recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the method for constructing a benefit evaluation model for a coal-fired power unit coupled with molten salt energy storage in an embodiment of the present invention. Figure 2 This is a structural block diagram of the benefit evaluation model of the coal-fired unit and molten salt energy storage coupling system in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0021] like Figure 1 As shown, the present invention provides a method for constructing a benefit evaluation model for a coal-fired power unit coupled with molten salt energy storage, comprising the following steps: Step S1: Construct a multi-source data processing module to perform data calibration and spatiotemporal registration on the multi-source operating data of each core component of the input coal-fired power unit and molten salt energy storage coupling system under different operating conditions, and construct a three-dimensional data cube containing spatial dimension, operating condition dimension and operating parameter dimension. Step S2: Construct a multi-dimensional feature extraction module to extract energy efficiency features, component collaborative coupling features, and time-series dynamic response features under different operating conditions for each core evaluation node in the three-dimensional data cube. Step S3: Construct a feature fusion and mapping module, which is used to bidirectionally map the same type of node-level features to their spatial location and corresponding working condition information, and generate a multi-dimensional coupled feature map that integrates spatial-working condition relationships. Step S4: Construct a distribution map generation module, which is used to output a spatial-condition distribution map of at least one key benefit indicator per node and per condition based on multi-dimensional coupled feature maps and the benchmark feature library of each core component and working condition through multi-dimensional feature correlation analysis. Step S5: Construct an evaluation and optimization module to identify core components and operating condition ranges with abnormal benefit attenuation in the coupled system based on the spatial-operating condition distribution map and preset evaluation benchmarks, quantify the impact of local shortcomings on overall benefits, and generate evaluation results that include overall benefit evaluation and local optimization suggestions.

[0022] In this embodiment, by deeply coupling spatial collaborative features and dynamic operating condition features, it is possible not only to accurately identify the shortcomings in the efficiency of a single component or a single operating condition, but also to reveal the collaborative attenuation law of different components under different operating conditions. By organically combining static component efficiency features with dynamic operating condition adaptation features, the generated spatial-operating condition distribution map has spatial targeted positioning and dynamic operating condition adaptation capabilities. It can provide precise directions for the hardware optimization and transformation of core components, provide optimization basis for the charging and discharging scheduling of molten salt energy storage under different operating conditions of the unit, realize integrated evaluation and optimization, and enhance the practical application value of energy-saving and consumption-reducing technologies of coupled systems.

[0023] In a specific implementation, as one example, given that existing coupled system benefit evaluation models use overall data homogenization for accounting, the input data cannot retain the spatial dimension of component differences and the dynamic fluctuation information of the operating condition dimension. This phenomenon stems from the lack of a mechanism for sorting out the spatiotemporal correlation of multi-source heterogeneous data and for accurate preprocessing. Based on this, this embodiment adopts a heterogeneous processing logic: First, the operating data of each core component of the coupled system comes from heterogeneous sources, has varying accuracy, and suffers from temporal asynchrony and spatial non-correlation. Therefore, data noise and spatiotemporal deviations need to be eliminated through calibration and registration. Second, a structured data carrier capable of simultaneously carrying three-dimensional information of space, operating conditions, and operating parameters needs to be constructed to achieve the associated storage and rapid retrieval of information from different dimensions. This differs from the existing approach that only focuses on overall data statistics. Through refined decomposition and association of data dimensions, the local and dynamic information required for evaluation is retained. Specifically, this includes the following steps: Step S11: Obtain multi-source operating data of each core component of the coal-fired power unit and molten salt energy storage coupling system under different operating conditions. This multi-source operating data originates from the monitoring systems and existing data acquisition links of each core component of the coal-fired power unit and molten salt energy storage coupling system. Core components include the boiler, turbine, molten salt tank, heat exchange unit, and auxiliary equipment. When selecting multi-source operating data for each component and different operating conditions, it is necessary to accurately reflect the component's operating status, energy efficiency level, and collaborative interaction characteristics, avoiding redundant parameters and invalid data. The operating parameters for each component and their corresponding selections are as follows: a. Boiler: Select data on furnace outlet flue gas temperature, steam drum pressure, feedwater flow rate, coal consumption, and exhaust oxygen content. Among these, furnace outlet flue gas temperature and exhaust oxygen content are directly related to boiler combustion efficiency, steam drum pressure and feedwater flow rate reflect boiler steam production capacity, and coal consumption is used to calculate energy consumption. These parameters together constitute the basic data for boiler energy efficiency assessment and are also used to analyze the synergistic characteristics with the heat exchange unit. The data can be obtained through the boiler's built-in thermocouple temperature sensors, pressure transmitters, electromagnetic flow meters, belt scales, and zirconia analyzers. b. Steam Turbine: Select data on inlet steam temperature, inlet steam pressure, exhaust steam temperature, turbine speed, and power generation. Among these, inlet / exhaust steam temperature and pressure directly affect the turbine's working efficiency, while speed and power generation characterize the turbine's output capacity and reflect the energy conversion effect when the turbine and molten salt energy storage system operate in tandem. Data can be obtained through temperature / pressure sensors on the turbine's inlet and exhaust pipes, shaft speed sensors, and generator outlet power transmitters. c. Molten Salt Storage Tank: Select data on molten salt temperature, molten salt level, molten salt inlet and outlet flow rate, and tank outer wall temperature at different layers within the tank. Among these, the multi-layer molten salt temperature reflects the uniformity of molten salt temperature distribution within the tank and is directly related to energy storage efficiency and heat dissipation loss. The liquid level and inlet / outlet flow rate are used to calculate the total energy charge and release, and the outer wall temperature is used to monitor the risk of energy storage leakage. The data can be obtained through platinum resistance temperature sensors arranged in layers within the tank, radar level gauges, electromagnetic flow meters on the molten salt pipelines, and thermocouple arrays deployed on the outer wall of the tank. d. Heat exchange unit: Select inlet and outlet temperatures on the molten salt side, inlet and outlet temperatures on the working fluid side, heat exchange area utilization rate, and heat transfer coefficient data; among them, the temperature difference between the inlet and outlet on both sides directly reflects the heat exchange efficiency, and the heat exchange area utilization rate and heat transfer coefficient can quantify the coordinated matching ability of the heat exchange unit and be used to evaluate heat exchange loss; the data can be obtained by temperature sensors on the molten salt and working fluid pipelines of the heat exchange unit and flow sensors based on the differential pressure method. e. Auxiliary Equipment: Select data on circulating water pump speed, oil pump pressure, cooling fan airflow, and valve opening. The operating parameters of auxiliary equipment directly affect the operational stability and energy consumption level of core components. For example, the circulating water pump speed is related to the circulation efficiency of molten salt and working fluid, and the valve opening affects the accuracy of medium flow regulation. These parameters are used to provide auxiliary data support for the extraction of collaborative coupling features. The data is obtained through water pump speed sensors, oil pump pressure transmitters, fan airflow sensors, and electric valve opening encoders. The above data acquisition frequency can be set uniformly to ensure that instantaneous parameter changes during the switching of operating conditions can be captured; Step S12: Perform differentiated calibration on multi-source data according to the error sources of different types of data to eliminate the influence of systematic errors, random errors and environmental interference; Among them, systematic errors are inherent errors of the sensors, such as zero-point drift of temperature sensors and range deviation of pressure sensors. The benchmark calibration method can be used to retrieve the calibration benchmark value of each sensor at a preset period, compare the collected raw data with the corresponding benchmark value, and eliminate systematic deviations through linear correction formulas. For example, the calibration process of temperature sensor is as follows: obtain the current measurement value of the sensor, retrieve the corresponding benchmark temperature, which is pre-calibrated by national metrological certification standard equipment, calculate the correction value, and uniformly superimpose the correction value on the subsequently collected temperature data to achieve calibration, ensuring that the temperature measurement error is controlled within the allowable range. Random error refers to random fluctuations caused by electromagnetic interference during data transmission, such as instantaneous jumps in power data. A sliding window filtering method can be used, which involves setting a sliding window of a preset size, calculating the mean of the original data within the window, and using the mean as the calibrated data. At the same time, outliers exceeding a preset threshold within the window are removed to avoid the impact of extreme data on the overall accuracy. The selection criteria for the window size are: to filter out high-frequency random noise without masking normal parameter changes during operating condition switching, thus balancing the filtering effect with data timeliness. Environmental interference error is the error generated by the sensor affected by environmental factors in outdoor or high-temperature environments. For example, the thermocouple on the outer wall of the storage tank is affected by the environmental temperature, and the pressure sensor of the open-air pipeline is affected by the atmospheric pressure fluctuation. The environmental compensation method can be used. By additionally arranging environmental monitoring sensors, environmental parameters are collected in real time, and the original data is corrected based on a preset compensation model. For example, the temperature calibration of the outer wall of the storage tank requires superimposing the difference compensation between the environmental temperature and the temperature of the outer wall of the storage tank to eliminate the interference of the environmental temperature on leakage monitoring and ensure that the outer wall temperature data can truly reflect the heat dissipation state of the storage tank; Step S13: After calibration, verify the validity of the data, set the reasonable value range of each parameter. This reasonable value range is preset based on the normal operating conditions of the coupling system. Data outside the value range is determined as invalid data. At the same time, call the calibration data of the previous acquisition cycle for supplementation to ensure data continuity; Step S14: Achieve synchronous alignment based on the time stamps carried by each data. When collecting data, record the time stamp of the collection moment synchronously, sort all data according to the time stamp, and use the linear interpolation method to process data with non-coincident time stamps. For example, the time stamp of the flue gas temperature data at the outlet of the boiler furnace is t1, t3, and the time stamp of the steam inlet temperature data of the steam turbine is t2, t1 < t2 < t3. Then, the flue gas temperature data at the outlet of the furnace at time t2 is obtained through interpolation calculation to ensure that the data of each component and each monitoring point at the same moment correspond one by one, meeting the requirement for extracting the timing characteristics during the working condition switching; Step S15: Based on the spatial installation coordinates of the core components, establish the spatial topological relationship of each monitoring point to bind the data with the physical position. The installation coordinates of each core component are collected in advance. The three-dimensional rectangular coordinate system can be used, with the center point of the unit as the origin, the X-axis along the length direction of the unit, the Y-axis along the width direction, and the Z-axis along the height direction. At the same time, record the relative coordinates of each monitoring point on the corresponding component, and associate and store the coordinate information with the calibration data of the corresponding monitoring point. For example, the coordinate of the temperature sensor in the middle layer of the molten salt storage tank is (X1, Y1, Z1), and its calibrated temperature data will carry this coordinate information to achieve the association of data, monitoring points, and component positions. At the same time, construct a spatial association matrix to clarify the spatial distance and interaction relationship between the monitoring points of different components, such as the spatial association between the monitoring points of the heat exchange unit and the inlet and outlet pipelines of the molten salt storage tank; Step S16: Based on the calibrated and spatiotemporally registered data, construct a three-dimensional data cube containing spatial, operational, and parameter dimensions. The spatial dimension corresponds to the physical installation locations and key monitoring points of core components, and can be divided into three levels: the first level is the core components; the second level is the monitoring areas of each component, such as the furnace area, steam drum area, and flue gas area for boilers, and the internal and external wall areas for molten salt storage tanks; the third level is the specific monitoring points within each monitoring area, such as the furnace outlet flue gas temperature monitoring point in the furnace area and the steam drum pressure monitoring point in the steam drum area. Each monitoring point corresponds to a unique spatial code to ensure the uniqueness of its spatial location. The operational dimension includes rated load, peak load, start-stop transition, and other preset operating conditions. Each operating condition is divided and identified based on the unit's operating load, molten salt charging and discharging status, and operating commands. The rated load condition can be defined as the unit load maintaining 90%-100% of its rated capacity. 00% molten salt energy storage system is in steady-state operation; peak load operation can be defined as the unit load fluctuating between 30% and 70% of rated capacity, with the molten salt energy storage system alternately charging and discharging energy; start-stop transition operation can be defined as the process from unit start-up to reaching stable load or from stable load to shutdown, with the molten salt energy storage system in preheating or unloading state; each operation corresponds to a unique operation code, and the operation switching time and duration are recorded to achieve temporal correlation of operation conditions; the operating parameter dimension includes all multi-source operating data after calibration and registration, and is stored in categories of energy efficiency parameters, collaborative parameters, and status parameters; energy efficiency parameters include coal consumption, power generation, and total molten salt charging and discharging energy; collaborative parameters include heat transfer coefficient and parameter response difference between components; status parameters include temperature, pressure, liquid level, and speed, each parameter corresponds to a unique parameter code, and is associated with its data source, calibration method, and accuracy index to ensure data traceability; The three-dimensional data cube is constructed using a distributed storage architecture. Data for each dimension is stored in a dimension table and a fact table, respectively. The dimension tables include spatial dimension tables, operating condition dimension tables, and operating parameter dimension tables, which store the corresponding dimension codes, attributes, and relationships. The fact table stores the specific data values ​​at the intersection of each dimension, such as the flue gas temperature data value of the boiler-furnace area-furnace outlet flue gas temperature monitoring point under rated load conditions, and establishes a relationship with each dimension table through dimension codes. At the same time, to improve the efficiency of data query and retrieval, multi-dimensional indexes are built in the cube, including spatial indexes, operating condition indexes, and parameter indexes. The update frequency of the data cube is consistent with the data acquisition frequency, realizing real-time data updates and dynamic maintenance.

[0024] In this embodiment, errors are eliminated through multi-source data calibration, and the spatiotemporal correlation of data is established through spatiotemporal registration, which can preserve local and dynamic information. The three-dimensional data cube fully carries three-dimensional information of space, working conditions, and operating parameters, which can realize the correlation of components, monitoring points, working conditions, and parameters, avoiding information loss caused by homogenization processing in existing technologies. It can directly support benefit analysis on a node-by-node and working condition-by-working condition basis, providing a basis for locating key components with benefit decay and the optimal / worst working condition range. The constructed multi-source data processing module is a hierarchical architecture of data input, calibration processing, spatiotemporal registration, and cube construction.

[0025] In some embodiments of the present invention, existing feature extraction for the benefit assessment of coupled systems is mostly based on extracting single energy efficiency features from the overall system operation data. This approach fails to consider the spatial distribution of components and dynamic changes in operating conditions. Limited by the drawbacks of data homogenization, it cannot reflect the local feature differences between different components and collaborative areas, and it is difficult to capture the dynamic fluctuations of features during operating condition switching. Consequently, the assessment can only rely on the overall average features, making it impossible to pinpoint the root causes of local benefit attenuation and shortcomings in operating condition adaptation. Therefore, feature extraction of the three-dimensional data cube adopts a node-based approach: first, core assessment nodes are divided according to the physical structure of components and the distribution of monitoring points, decomposing the overall system into independently assessable local units and collaborative units; then, for each node, energy efficiency features, component collaborative coupling features, and time-series dynamic response features are extracted. These three types of features correspond to the node's own energy conversion level, the degree of collaborative matching between nodes, and the node's adaptability to changes in operating conditions, respectively. Through multi-dimensional feature complementarity, the information required for benefit assessment is covered. Specifically, the following steps are included: Step S21: Divide the core evaluation nodes according to the physical structure of the core components and the distribution of monitoring points. The division principle is to ensure that a single node can accurately correspond to the operating status of a specific physical area, while covering two types of scenarios: independent operation of a single component and collaborative operation of multiple components, and adapting to the extraction requirements of local features and collaborative features, as follows: a. For core components such as boilers, steam turbines, molten salt storage tanks, heat exchange units, and auxiliary equipment, nodes are divided according to their physical structure and the coverage of monitoring points. Each node corresponds to a specific monitoring area of ​​a single component, ensuring that the node boundary is consistent with the physical zoning of the component, and that each node covers at least one set of operating parameter data from monitoring points. For example, boilers are divided into furnace nodes, steam drum nodes, and flue gas nodes according to the physical structure of the furnace, steam drum, and flue gas pipes. Molten salt storage tanks are divided into upper-layer nodes, middle-layer nodes, lower-layer nodes, and outer-wall nodes according to the upper, middle, and lower monitoring areas inside the tank and the outer wall monitoring area. The division is based on the fact that the operating states of different areas of a single component are different. After division, the local characteristics of each area can be accurately captured, avoiding the overall characteristics of a single component from masking local loss problems. b. For component combinations with energy interaction or operational linkage, nodes are divided according to collaborative interaction areas. A single node corresponds to the collaborative interaction areas of multiple components, covering parameter transmission and operational linkage between components. The definition of collaborative areas is based on the physical connection relationship and energy flow path between components. For example, the steam transmission pipeline area between the boiler and the heat exchange unit, the molten salt transmission channel area between the molten salt storage tank and the heat exchange unit, and the speed regulation linkage area between the steam turbine and auxiliary equipment are all separately divided into collaborative nodes. During the division, the corresponding monitoring point data of each collaborative component are synchronously associated to ensure that the node can capture the operational parameter response relationship of multiple components within the interaction area. Step S22: Assign a unique identifier to each core evaluation node to identify the associated node type, corresponding physical area, covered monitoring points and associated component information. At the same time, establish a mapping relationship between the node and the three-dimensional data cube to achieve accurate association between the node and spatial location, working condition and operating parameters, and ensure that all operating parameter data under the corresponding working condition can be retrieved by node. Step S23, Energy Efficiency Feature Extraction: Calculate the energy efficiency feature parameters of each node according to different working conditions. Based on the effective data during the continuous working period, extract two core features: unit energy consumption output energy and energy loss ratio. For collaborative nodes, additional energy transfer efficiency features are extracted. All are converted by the corresponding energy data ratio to reflect the energy conversion efficiency, loss degree and collaborative transfer effect respectively. Step S24: Extraction of component cooperative coupling features: Under steady-state conditions, parameters are collected at preset intervals and the response difference sequence is calculated. Two features are extracted: the mean difference and the difference fluctuation amplitude, which reflect the level of cooperative deviation and the operational stability, respectively. When switching operating conditions, the difference response delay feature is extracted to reflect the timeliness of response through the time difference of parameter changes. Step S25, Extraction of time-series dynamic response features: Collect the parameter change sequence of the entire switching process, extract three features: response speed, adjustment stability, and steady-state adaptability, and calculate them respectively through parameter change duration, fluctuation, and deviation between steady-state value and benchmark value, and retain the original time-series information; Step S26: Perform preliminary verification on the three types of features, and remove invalid and abnormal features. The verification is based on the validity of the original parameters in the 3D data cube, the rationality of the node division, and the accuracy of the working condition definition. If the original parameters corresponding to a feature have a high proportion of invalid data, incorrect node identification, or deviation in the working condition boundary definition, it is determined to be an invalid feature and marked. At the same time, it is completed based on the feature mean of the same type of node. For abnormal features that exceed the reasonable value range, the corresponding node's operating parameter change sequence is reviewed to confirm whether it is caused by abnormal node operation. Abnormal features are marked separately and associated with the original parameter log. Step S27: Label each feature with the corresponding node identifier, working condition type, and data statistics period to ensure that the features are accurately bound to the nodes and working conditions, and at the same time associate them with the original parameter data in the 3D data cube.

[0026] In this embodiment, the construction of the multi-dimensional feature extraction module is based on a three-dimensional data cube. After dividing the core evaluation nodes according to rules, energy efficiency features, component collaborative coupling features, and time-series dynamic response features are extracted simultaneously to form a multi-dimensional feature system covering energy conversion, spatial collaboration, and operating condition adaptation. The extraction process is node-based, follows the dynamic change law of component physical structure and operating conditions, avoids the problem of homogeneous extraction in existing technologies, and ensures the accuracy and traceability of features through preliminary verification and feature calibration.

[0027] In practical implementation, as an example, existing methods for feature fusion often adopt a single-dimensional overlay approach, or only aggregate features in the spatial dimension, or only sort out the temporal changes of features under different operating conditions, without establishing a two-way correlation logic between spatial location and operating condition information. Although node-level features cover three dimensions: energy efficiency, collaboration, and temporal sequence, the features are still scattered and correspond to a single node and a single operating condition, lacking cross-space and cross-operating condition correlation and integration. This fails to reflect the feature collaboration relationship of different nodes under the same operating condition, and it is also difficult to reflect the feature evolution law of the same node in different operating condition switching. Based on this, this embodiment maps and associates the same type of features of different nodes under the same operating condition through spatial dimension mapping to capture spatial collaboration differences, and maps and connects the same type of features of the same node under different operating conditions through operating condition dimension mapping to capture the dynamic response law of operating conditions. Then, the two types of mapping results are fused to generate a coupled feature map that combines spatial topological correlation and operating condition temporal correlation, realizing the structured integration of fragmented features. Specifically, it includes the following steps: Step S31: Classify and organize the three types of node-level features, dividing them into energy efficiency feature group, component collaborative coupling feature group, and time-series dynamic response feature group according to feature type. Establish a preliminary index within each group according to node identifier and operating condition type; at the same time, retrieve the node spatial topology relationship data and operating condition switching sequence data from the three-dimensional data cube. Step S32, Spatial Dimension Mapping: For feature groups of the same type under the same working condition, feature association is completed according to the spatial topology of nodes. First, a single working condition is locked, and the same type of features of all core evaluation nodes under this working condition are extracted. Then, based on the spatial installation coordinates and topological relationships of nodes in the three-dimensional data cube, the same type of features of nodes in different spatial locations are associated according to the physical connection path and energy flow direction. The association follows the principle of proximity priority and collaborative guidance. Directly associated nodes are given priority to establish feature association links, and indirectly associated nodes achieve feature transmission association through intermediate nodes. During the mapping process, the spatial association strength is marked for each link. The association strength is determined based on the spatial distance between nodes and the energy transfer efficiency. Step S33, Working Condition Dimension Mapping: For the same type of feature group of the same node, complete the feature association according to the working condition switching order; first, lock a single core evaluation node, extract the same type of features of the node under all working conditions, and then, based on the working condition dimension information in the three-dimensional data cube, string the features together according to the working condition switching time series; during the association process, retain the feature mutation information at the time of working condition switching, mark the feature change amplitude and trend between adjacent working conditions, and form the working condition time series link of the same type of features of a single node; at the same time, associate the working condition duration and switching interval to provide a time dimension basis for the analysis of feature change patterns, and realize the working condition time series integration of the same type of features of the same node; Step S34: Using a method of superimposing bidirectional mapping results and integrating related information, the feature association network formed by spatial dimension mapping and the temporal link formed by working condition dimension mapping are fused together; taking node identifier as the core hub, the temporal links of the same node under different working conditions are embedded into the corresponding spatial association chain to realize the linkage of node working condition temporal features with other nodes of the same type; taking working condition type as the auxiliary hub, the spatial association network of the same working condition and the feature network of adjacent working conditions are connected through node feature changes to form a cross-working condition spatial feature linkage system; during the fusion, the annotation information such as spatial association strength and working condition feature change amplitude are retained, and the three types of feature group results are integrated according to the three-dimensional structure of node-working condition-feature, duplicate association information is removed, and the feature storage structure is optimized. Step S35: Based on the fused feature data, a graph structure storage mode is used to construct the graph. Graph nodes correspond to core evaluation nodes, and node identifiers, spatial locations, and associated component information are labeled. Graph edges are divided into spatial edges and operating condition edges, corresponding to spatial dimension mapping links and operating condition dimension mapping links, respectively, and the spatial association strength and the magnitude of feature changes during operating condition switching are simultaneously labeled. Each core node is associated with three types of feature data under different operating conditions, and each feature data corresponds to associated spatial edge and operating condition edge information, forming a three-dimensional coupled feature graph with interconnected nodes, interconnected operating conditions, and linked features.

[0028] In this embodiment, a feature fusion and mapping module is constructed based on a three-dimensional data cube and node-level features. Feature integration is completed through spatial-operating condition bidirectional mapping to generate a multi-dimensional coupled feature map. Throughout the process, fragmented node-level features are transformed into structured map data that combines spatial correlation and operating condition time series through bidirectional mapping and fusion. This can compensate for the shortcomings of single-dimensional fusion and fully present the spatial distribution pattern and dynamic change pattern of the coupled system features. The information such as the spatial correlation strength and the magnitude of operating condition feature changes carried by the map can directly support multi-dimensional feature correlation analysis, providing structured feature basis for accurately calculating the key benefit indicators of each node under the corresponding operating condition, and facilitating the location of abnormal benefit areas at the intersection of spatial and operating conditions.

[0029] In a specific implementation, as one example, to ensure that the generated distribution map reflects the fluctuations of indicators at each node and working condition, and reflects the benefit distribution pattern of the spatial-working condition cross-dimensional dimension, this embodiment constructs a logic of benchmark feature library-weight adaptation-feature correlation analysis-indicator calculation-map generation based on the spatial-working condition-feature linkage capability of multi-dimensional coupled feature maps: First, a benchmark feature library that fits the characteristics of components and working condition requirements is established to provide standardized comparison references; then, working condition and spatial weight coefficients are introduced to adapt to the degree of benefit impact of different working conditions and nodes; through multi-dimensional feature correlation analysis, accurate comparison between actual features and benchmark features is achieved, and node-by-node and working condition indicators are calculated in combination with weights; finally, a distribution map that integrates spatial location, working condition type, and indicator values ​​is generated, making up for the shortcomings of inaccurate calculation of existing technical indicators and fragmented map information; the specific implementation steps are as follows: Step S41: Construction and Calibration of the Benchmark Feature Library; The benchmark feature library is constructed according to the type of core component and the type of operating condition, covering all core evaluation nodes and preset operating conditions, providing a standardized reference for comparing actual feature differences; During construction, historical data under normal operating conditions of the coupled system are used as the basis, and rated operating data and industry standard values ​​of the same type of unit and molten salt energy storage coupled system are retrieved. Combined with the design parameters of this system, benchmark features are sorted out according to the three-level structure of component, node and operating condition; the benchmark features of a single component node are set according to the functional characteristics of the corresponding component, the benchmark features of the collaborative node are set according to the optimal operating state of multi-component collaboration, and the benchmark features under different operating conditions are adjusted in combination with the design operating parameters of the operating condition; After the benchmark feature library is constructed, it is calibrated regularly, and the benchmark values ​​are updated based on the effective data of the long-term operation of the system, eliminating the impact of abnormal historical data on the accuracy of the benchmark, and marking the applicable operating condition range and node correspondence of the benchmark features to ensure the matching degree between the benchmark features and the actual features; Step S42: Determine the operating condition weight coefficient and the spatial weight coefficient. These two types of weight coefficients are adapted to the impact of operating conditions and spatial nodes on overall benefits, ensuring that the calculated indicators align with the actual operation of the system. The operating condition weight coefficient is determined based on the duration, energy consumption ratio, and contribution to system benefits of different operating conditions. Operating conditions with longer durations and higher energy consumption ratios are assigned higher weights, and the sum of the weight coefficients meets the normalization requirements. The spatial weight coefficient is determined based on the energy transfer importance of core evaluation nodes and the strength of their collaborative relationship with other nodes. Key energy transfer nodes and nodes with strong collaborative relationships are assigned higher weights, and the weight values ​​are derived from the spatial topology of the nodes and the proportion of energy flow. After the two types of weight coefficients are determined, they are associated with the corresponding operating conditions and nodes, stored in the coefficient library, and periodically reviewed and adjusted according to the system's operating status to avoid weight fixation leading to deviations in indicator calculations. Step S43: Using a multi-dimensional coupled feature map as a carrier, the analysis is completed according to the logic of feature matching, difference quantification, and weight fusion. The actual values ​​of key benefit indicators under the corresponding working conditions of each node are calculated. First, for each core evaluation node, the benchmark features of the corresponding components and working conditions are matched in the benchmark feature library to ensure the dimensional consistency between the actual features and the benchmark features. Then, the difference between the actual features and the benchmark features is quantified, and the difference value is calculated according to the feature type. The difference of energy efficiency features is derived based on the deviation of energy conversion efficiency, the difference of collaborative coupling features is derived based on the deviation of parameter response difference, and the difference of time-series dynamic response features is derived based on the deviation of working condition switching response law. Finally, the difference values ​​of various features are fused with the corresponding working condition weight coefficient and spatial weight coefficient, and integrated into the actual value of key benefit indicators according to the preset rules. Each node and each working condition corresponds to a unique indicator value to ensure that the indicator can accurately reflect the local benefit level. Step S44: Select at least one core key benefit indicator as the object of the graph presentation. The selection principle is that it is strongly related to the core needs of system benefit assessment and can accurately reflect the energy utilization efficiency and collaborative benefits under the node operating conditions. For example, select the node energy efficiency utilization rate and the collaborative benefit contribution value as core indicators. After selection, verify the actual value of the indicator, compare the rationality of the indicator distribution of the same type of node under the same operating conditions, remove extreme indicator values ​​caused by feature matching errors or abnormal weight coefficients, and fill in the missing values ​​based on the mean of the effective indicators in the same batch. At the same time, associate and couple the original feature information in the feature graph to realize the traceability of indicator values ​​and feature data, and ensure that the accuracy of the indicators can be verified. Step S45: Based on the actual values ​​of the key benefit indicators after screening and verification, a distribution map is generated using a visual graph structure. The map simultaneously carries spatial, working condition, and indicator dimension information. The map is based on a three-dimensional coordinate system. The spatial dimension corresponds to the physical installation location of the core evaluation nodes, and the node identifiers and associated component information are marked. The working condition dimension is presented in layers according to preset working condition types, and the working condition switching sequence and continuous characteristics are marked. The indicator dimension is associated with the corresponding nodes and working conditions through color gradients and numerical annotations, presenting the distribution of indicator values ​​of different nodes under different working conditions. The map supports multi-dimensional interactive queries, and can retrieve cross-working condition indicator changes by node and cross-spatial indicator distribution by working condition, while retaining the correlation information such as difference values ​​and weight coefficients in the indicator calculation process.

[0030] In this embodiment, a distribution map generation module is constructed based on a multi-dimensional coupled feature map and a benchmark feature library. Through weight adaptation and multi-dimensional feature correlation analysis, key benefit indicators are calculated and distribution maps are generated for each node and each working condition. The entire process follows the system operation rules and the requirements for refined evaluation, avoiding the problems of homogeneous calculation and fragmented map information in existing technologies. This ensures the accuracy of indicators and the practicality of the map, providing a visualized and traceable basis for benefit distribution for evaluation and optimization. It makes up for the shortcomings of existing technologies in local benefit positioning and dynamic distribution presentation, and supports the refined evaluation of the benefits of the coupled system.

[0031] In a specific implementation, as one example, given that existing technologies for evaluating and optimizing coupled systems largely rely on overall benefit indicators, they can only determine whether there is overall benefit attenuation in the system, but cannot pinpoint the specific components and operating conditions corresponding to the attenuation. Furthermore, it is difficult to quantify the impact of local bottlenecks on overall benefits. Moreover, optimization suggestions are mostly general guidelines, not designed to incorporate the characteristics of abnormal parts and the requirements for adapting to operating conditions, thus limiting their guiding value. Therefore, it is necessary to establish a complete logical chain for anomaly location, impact quantification, and precise suggestions by combining refined benefit distribution across spatial and operating conditions. This embodiment uses refined benefit data and weighting coefficients from the spatial-operating condition distribution map to construct a hierarchical evaluation and optimization logic to overcome the above deficiencies, achieving accurate anomaly identification, quantification of local impacts, and the formulation of targeted optimization suggestions. The specific steps are as follows: Step S51: Setting and calibrating preset evaluation benchmarks; preset evaluation benchmarks are set according to core component type, core evaluation node, and operating condition type, providing a standardized reference for judging abnormal benefit attenuation, complementing the benchmark feature library. The former is used for judging the qualification of indicators, and the latter is used for the reference of indicator calculation; the benchmark setting is based on the rated benefit level of the coupled system design, and retrieves the benefit data of long-term stable operation of similar units, industry energy efficiency standards, and component performance parameter thresholds, and sorts the benchmark values ​​according to the three-level structure of component-node-operating condition; the benchmark of a single component node corresponds to the benefit index range under the rated operating state of the component, the benchmark of a collaborative node corresponds to the optimal benefit range of multi-component collaboration, and the benchmarks under different operating conditions are adjusted in combination with the design benefit target of the operating condition; after the benchmark is set, it is calibrated regularly, based on the effective operating data of the system and the update of industry standards, marking the applicable scope of the benchmark and the correspondence between nodes and operating conditions, and setting the benchmark floating range to adapt to normal operating fluctuations, avoiding abnormal misjudgment due to the solidification of the benchmark. Step S52: Based on the actual values ​​of key benefit indicators in the spatial-operating condition distribution map and the preset evaluation benchmark, identify anomalies according to the node-operating condition-cross-dimensional integration logic; first, retrieve the actual values ​​of indicators of each core evaluation node in the map under different operating conditions, compare them with the corresponding preset evaluation benchmark, and calculate the degree of deviation; then set a deviation threshold, and if the deviation exceeds the threshold, it is determined that the node has an abnormal benefit decay under the corresponding operating condition; this deviation threshold is derived based on the normal operating fluctuation range of the system and is dynamically adjusted according to the type of operating condition and the importance of the node; during the identification process, the spatial location and operating condition information in the map are synchronously associated to locate the physical components, monitoring areas, and operating condition intervals corresponding to the core evaluation nodes of the anomaly; cross-node and cross-operating condition related anomalies are marked as systemic anomalies, and the correlation relationships are sorted out separately; Step S53: Based on the operating condition weight coefficient, spatial weight coefficient, and abnormal data, quantify the impact of local bottlenecks. First, extract the spatial weight coefficient corresponding to the abnormal node and the operating condition weight coefficient corresponding to the abnormal operating condition. Combine the difference between the actual value of the node's index under the abnormal operating condition and the preset benchmark to calculate the impact of a single abnormal node-operating condition combination on the overall benefit. Then, summarize the impact of all abnormal combinations within the same component and the same operating condition interval to obtain the total local impact of a single component and a single operating condition interval. Finally, compare the total local impact with the total overall system benefit to obtain the proportion of the impact of local bottlenecks on the overall benefit, and identify the core bottleneck as the abnormal component with the highest impact proportion and the key bottleneck operating condition as the abnormal operating condition interval with the most significant impact. During the quantification process, retain the calculation basis, correlate the index data and weight coefficients in the graph, and ensure that the quantification results are traceable. Step S54: Develop overall benefit assessment and local optimization recommendations. The overall benefit assessment is based on the statistical analysis of all indicators from the spatial-operating condition distribution map, combined with the results of local impacts, to calculate the overall system benefit achievement rate and overall benefit loss, and to clarify the overall system benefit level and core sources of loss. Local optimization recommendations are developed for different types and locations of anomalies, taking into account component operating characteristics, causes of anomalies, and operating condition adaptation requirements: For anomalies caused by heat exchange losses, it is recommended to optimize the heat exchange unit structure or replace sealing components; for anomalies caused by insufficient coordination and matching, it is recommended to adjust operating condition switching parameters or optimize component linkage logic; for anomalies caused by operating condition adaptation deviations, it is recommended to limit the optimal operating condition range or adjust the operating parameters under that condition. Recommendations should include the applicable scenarios, implementation paths, and corresponding expected benefit improvements to ensure relevance and feasibility. Step S55: Integrate the overall benefit assessment conclusions, anomaly identification results, local impact quantification data, and optimization suggestions to form standardized assessment results; the results are organized according to the structure of overall overview - local anomaly details - impact analysis - optimization scheme, and the source and reliability level of each data point are marked; a visualization report is generated simultaneously, linked to the S4 space - operating condition distribution map, and the locations of abnormal areas and core shortcomings are marked; the output format is adapted to industrial application scenarios and can be exported as structured documents and data interfaces to facilitate fault diagnosis, transformation implementation, and operating condition adjustment.

[0032] In this embodiment, the evaluation and optimization module is constructed based on the spatial-operating condition distribution map and the preset evaluation benchmark. The complete evaluation result is generated by identifying benefit attenuation anomalies, quantifying the impact of local shortcomings, and formulating targeted optimization suggestions. The whole process follows the system operation rules and the requirements of refined evaluation, avoiding the problems of rough anomaly identification, inability to quantify local impact, and general optimization suggestions in the prior art, and ensuring that the evaluation result meets the actual application requirements.

[0033] like Figure 2 As shown, the present invention also provides a benefit evaluation model for a coal-fired power unit coupled with molten salt energy storage system, which is constructed using the above-described method for constructing a benefit evaluation model for a coal-fired power unit coupled with molten salt energy storage system.

[0034] The model in this invention can effectively realize the benefit assessment of the coal-fired power unit and molten salt energy storage coupling system, and the technical effects it can achieve are as described in the above embodiments, which will not be repeated here.

[0035] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined herein, and are to be considered as covering any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for constructing a benefit evaluation model for a coal-fired power unit coupled with molten salt energy storage system, characterized in that, include: A multi-source data processing module is constructed to perform data calibration and spatiotemporal registration on the multi-source operating data of the core components of the coal-fired power unit and molten salt energy storage coupling system under different operating conditions, and to construct a three-dimensional data cube containing spatial dimension, operating condition dimension and operating parameter dimension. A multi-dimensional feature extraction module is constructed to extract energy efficiency features, component collaborative coupling features, and time-series dynamic response features under different operating conditions for each core evaluation node in the three-dimensional data cube. A feature fusion and mapping module is constructed to bidirectionally map the same type of node-level features to their spatial location and corresponding working condition information, generating a multi-dimensional coupled feature map that integrates spatial-working condition relationships. A distribution map generation module is constructed to output a spatial-condition distribution map of at least one key benefit indicator per node and per condition, based on the multi-dimensional coupled feature map and the benchmark feature library of each core component and working condition, through multi-dimensional feature correlation analysis. An evaluation and optimization module is constructed to identify core components and operating condition ranges with abnormal benefit attenuation in the coupled system based on the spatial-operating condition distribution map and preset evaluation benchmarks, quantify the impact of local shortcomings on overall benefits, and generate evaluation results that include overall benefit evaluation and local optimization suggestions.

2. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupled system according to claim 1, characterized in that, The core components include a boiler, a steam turbine, a molten salt storage tank, a heat exchange unit, and auxiliary equipment.

3. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupled system according to claim 1, characterized in that, The spatiotemporal registration is based on data timestamps to achieve time dimension alignment and on the spatial dimension association based on the spatial installation coordinates of core components.

4. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupled system according to claim 1, characterized in that, The spatial dimension of the three-dimensional data cube corresponds to the physical installation location of the core components and key monitoring points. The operating condition dimension includes rated load, peak load, start-stop transition and other preset operating conditions. The operating parameter dimension includes multi-source operating data after calibration and registration.

5. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupled system according to claim 1, characterized in that, The core evaluation nodes are divided according to the physical structure of the core components and the distribution of monitoring points. A single core evaluation node corresponds to a specific monitoring area of ​​a single component or a collaborative interaction area of ​​multiple components.

6. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupled system according to claim 1, characterized in that, The energy efficiency feature is calculated based on node energy consumption data and output energy data; the component collaborative coupling feature is calculated based on the difference in operating parameter response between associated nodes; and the time-series dynamic response feature is extracted based on the sequence of operating parameter changes of the same node during the switching of operating conditions.

7. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupled system according to claim 1, characterized in that, The bidirectional mapping includes spatial dimension mapping and working condition dimension mapping; the spatial dimension mapping associates the same type of features of core evaluation nodes in different spatial locations under the same working condition according to spatial topological relationships; the working condition dimension mapping associates the same type of features of core evaluation nodes in the same spatial location according to the working condition switching order.

8. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupled system according to claim 1, characterized in that, The multi-dimensional feature association analysis includes: By comparing the actual characteristics of the core evaluation nodes with the corresponding characteristics in the benchmark feature library, and combining the working condition weight coefficient and the spatial weight coefficient, the actual values ​​of the key benefit indicators under the corresponding working conditions of each node are calculated.

9. The method for constructing the benefit evaluation model of the coal-fired power unit and molten salt energy storage coupling system according to claim 1, characterized in that, The determination of abnormal benefit decay is based on the degree of deviation between the actual value of the key benefit indicator and the corresponding preset evaluation benchmark. When the degree of deviation exceeds the preset threshold, it is determined to be abnormal benefit decay.

10. A benefit evaluation model for a coal-fired power unit coupled with molten salt energy storage system, characterized in that, The model is constructed using the method described in any one of claims 1 to 9 for evaluating the benefits of a coal-fired power unit coupled with a molten salt energy storage system. The model is configured to evaluate the benefits of the coal-fired power unit coupled with the molten salt energy storage system.

Images