General evaluation system for model operation results

Abstract

The invention relates to the technical field of electronics, in particular to a universal evaluation system for model operation results, and the system comprises a result obtaining module which is used for operating a plurality of to-be-evaluated models to process to-be-processed information, and obtaining the operation result of each to-be-evaluated model by adopting each to-be-evaluated model; the information acquisition module is used for evaluating the operation results obtained by the result acquisition module by utilizing a TOPSIS mechanism to obtain evaluation information and an initial evaluation result of each to-be-evaluated model; wherein the evaluation information at least comprises accuracy evaluation, stability evaluation, adaptability evaluation and efficiency evaluation; the initial evaluation result is the adaptation degree of the to-be-evaluated model and the application scene; and the target evaluation module is used for processing the evaluation information obtained by the information acquisition module and the initial evaluation result by using a Pareto mechanism to obtain a target evaluation result of each to-be-evaluated model. Evaluation errors caused by subjective reasons of workers are avoided.

Description

Technical Field

[0001] The present invention relates to the field of electronic technology, and in particular to a universal evaluation system for model operation results. Background Art

[0002] Modeling technology has become an important tool for solving scientific problems and optimizing management. For example, cascade reservoir operation models can simulate the dynamic processes of reservoirs within a river basin, providing decision support for flood control, power generation, and ecological protection. In energy management, optimization models help balance supply and demand and achieve efficient resource allocation. In meteorology, forecasting models can provide early warning of extreme weather events and mitigate the impact of disasters. The widespread application of these models relies on the scientific evaluation of their operational results.

[0003] Currently, model evaluation is typically based on a series of sub-items. However, existing technologies require manual data collation, indicator setting, and result calculation, which is time-consuming and labor-intensive, and is susceptible to human subjectivity, resulting in low accuracy in model evaluation. Summary of the Invention

[0004] (1) Technical issues to be resolved

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a universal evaluation system for model operation results, which solves the technical problems that the existing model evaluation is easily affected by human subjectivity, resulting in inaccurate evaluation, low efficiency and high error rate.

[0006] (2) Technical solution

[0007] In order to achieve the above objectives, the main technical solutions adopted by the present invention include:

[0008] In the first aspect, a general evaluation system for model operation results proposed in an embodiment of the present invention is applied to the field of energy management, and the system includes: a result acquisition module, which is used to run multiple models to be evaluated, and use each model to be evaluated to process the information to be processed, and obtain the operation results of each model to be evaluated; an information acquisition module, which is used to use the TOPSIS mechanism to evaluate the operation results obtained by the result acquisition module respectively, and obtain evaluation information and initial evaluation results of each model to be evaluated; wherein, the evaluation information includes at least: accuracy evaluation, stability evaluation, adaptability evaluation and efficiency evaluation; the initial evaluation result is the degree of adaptation of the model to be evaluated to the application scenario; a target evaluation module, which is used to use the Pareto mechanism to process the evaluation information and the initial evaluation result obtained by the information acquisition module, and obtain the target evaluation result of each model to be evaluated.

[0009] Optionally, the information acquisition module further includes: an initial evaluation module, configured to obtain initial evaluation results of each to-be-evaluated model based on the evaluation information and its evaluation weight.

[0010] Optionally, the system also includes a sensor network, which is used to obtain environmental information; wherein the environmental information includes at least one of the following: water conservancy information, energy information, and meteorological information; the information acquisition module is also used to input the environmental information obtained by the sensor network into a trained weight evaluation model, and output an evaluation weight corresponding to the environmental information; wherein the weight evaluation model includes at least one of the following: a hierarchical analysis model, a principal hierarchical analysis model, a fuzzy comprehensive evaluation model, and a gray correlation evaluation model.

[0011] Optionally, the target evaluation module is also used to evaluate the evaluation information of each model to be evaluated using the Pareto mechanism to determine at least one optimal model; the target evaluation module is specifically used to calculate the adjustment coefficient of each optimal model based on the distance between each optimal model and the positive ideal solution; wherein the adjustment coefficient and the distance are negatively correlated; the adjustment coefficient is used to regulate the evaluation results of each optimal model, and the adjustment coefficient is greater than 1; and the target evaluation result of each optimal model is obtained by using the initial evaluation result of the optimal model and the adjustment coefficient.

[0012] Optionally, the target evaluation module is further used to perform normalization processing based on the distance between each optimal model and the positive ideal solution to obtain the processing distance of each optimal model; and to obtain the adjustment coefficient of each optimal model using the processing distance and the distance between each optimal model and its corresponding positive ideal solution.

[0013] Optionally, the information acquisition module is used to calculate the positive ideal solution and the negative ideal solution of each model to be evaluated; the information acquisition module is also used to obtain evaluation information of each model to be evaluated based on the positive ideal solution and the negative ideal solution.

[0014] Optionally, the information acquisition module is further configured to acquire the running time and resource usage information of each model to be evaluated; and to perform evaluation using the running time and resource usage information to obtain an efficiency evaluation of each model to be evaluated.

[0015] Optionally, the system further includes a report generation module, the report generation module being configured to generate an evaluation report for each model to be evaluated based on the target evaluation result obtained by the target evaluation module; wherein the evaluation report includes at least one of the following: a radar chart, a bar chart;

[0016] The histogram includes evaluation information of each model to be evaluated.

[0017] Optionally, the model to be evaluated includes at least one of the following: a cascade reservoir model, a distributed hydrological model, an energy system planning model, and a numerical weather forecast model.

[0018] Optionally, the result acquisition module is used to obtain the operation results of each model to be evaluated based on the processing time and / or resource consumption of the information to be processed of each model to be evaluated.

[0019] Beneficial effects

[0020] The beneficial effects of the present invention are as follows: the present invention provides a universal evaluation system for the model operation results, and the present invention utilizes the result acquisition module in the system to operate multiple models to be evaluated to process the information to be processed, so as to obtain the operation results of each model to be evaluated. The TOPSIS mechanism is used to evaluate the operation results of multiple models to be evaluated respectively, so as to obtain the evaluation information of each model to be evaluated. In other words, the present invention eliminates the influence of each evaluation indicator and obtains the initial evaluation results that can be compared on the same scale. Next, the present invention utilizes the Pareto mechanism to find the model with the best trade-off among the multiple evaluation dimensions of each model to be evaluated, so as to obtain the target evaluation result. Specifically, the present invention utilizes the TOPSIS mechanism and the Pareto mechanism to jointly evaluate the model to be evaluated, which avoids the participation of staff in the evaluation process to a certain extent, and avoids evaluation errors due to subjective reasons of staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A schematic diagram of the structure of a general evaluation system for model operation results provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] Currently, modeling technology has become an important tool for solving scientific problems and optimizing management in various fields. For example, cascade reservoir operation models can simulate the dynamic processes of reservoirs within a river basin, providing decision support for flood control, power generation, and ecological protection. In energy management, optimization models help balance supply and demand and achieve efficient resource allocation. In meteorology, forecasting models can provide early warning of extreme weather events and reduce the impact of disasters.

[0023] In order to improve the accuracy and efficiency of models in solving scientific problems and optimizing management in various fields, it is necessary to evaluate different models based on the application field and the evaluation of operation results, so as to make the best choice.

[0024] Currently, model performance evaluation is typically based on a series of metrics. For example, error analysis measures accuracy, runtime evaluates efficiency, exception handling capabilities reflect the robustness of the model, and cross-scenario application effectiveness assesses its applicability. Evaluation methods include both quantitative analysis based on mathematical statistics and qualitative analysis based on expert opinion.

[0025] However, existing technologies require staff to manually organize data, set indicators and calculate results, which is time-consuming and labor-intensive and easily affected by human subjectivity.

[0026] In order to solve the above problems, the present invention provides a universal evaluation system for model operation results.

[0027] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation methods in conjunction with the accompanying drawings.

[0028] The present invention utilizes the result acquisition module in the system to run multiple models to be evaluated to process the information to be processed, so as to obtain the operating results of each model to be evaluated. The TOPSIS mechanism and the operating results of the multiple models to be evaluated are evaluated separately to obtain the evaluation information of each model to be evaluated. In other words, the present invention eliminates the influence of each evaluation indicator and obtains the initial evaluation results that can be compared on the same scale. Next, the present invention utilizes the Pareto mechanism to find the model with the best trade-off among the multiple evaluation dimensions of each model to be evaluated, thereby obtaining the target evaluation result. Specifically, the present invention utilizes the TOPSIS mechanism and the Pareto mechanism to jointly evaluate the model to be evaluated, which to a certain extent avoids the participation of staff in the evaluation process and avoids evaluation errors due to subjective reasons of staff.

[0029] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0030] Please refer to Figure 1 ,like Figure 1 As shown, the general evaluation system for model operation results involved in the present invention specifically includes:

[0031] The result acquisition module is used to run multiple models to be evaluated to process the information to be processed, so as to obtain the operation results of each model to be evaluated.

[0032] An information acquisition module is used to evaluate the operating results obtained by the result acquisition module using the TOPSIS mechanism to obtain evaluation information and initial evaluation results of each model to be evaluated; wherein the evaluation information includes at least: accuracy evaluation, stability evaluation, adaptability evaluation and efficiency evaluation; the initial evaluation result is the degree of adaptation of the model to be evaluated to the application scenario.

[0033] The target evaluation module is used to process the evaluation information and the initial evaluation results obtained by the information acquisition module using the Pareto mechanism to obtain the target evaluation results of each model to be evaluated.

[0034] The universal evaluation system for model operation results provided by the present invention can be applied to the field of energy management, where the energy management field includes but is not limited to at least one of the following: water energy management, meteorology, solar energy management, and meteorology. In other words, the universal evaluation system for model operation results provided by the present invention can be applied to different models to be evaluated in different fields.

[0035] Specifically, the general evaluation system for the model operation results includes at least but is not limited to at least one of the following: a result acquisition module, an information acquisition module, and a target evaluation module.

[0036] When multiple models to be evaluated need to be evaluated, each model to be evaluated can be first input into the result acquisition module, and then evaluated based on the running results of the information to be processed output by it to obtain its performance, thereby evaluating and / or selecting each model to be evaluated. For example, the general evaluation system for model running results can score each model to be evaluated according to actual conditions, and use the model to be evaluated with the highest score as the optimal model, so as to be applied in subsequent data processing. For another example, the score obtained by the general evaluation system for model running results provided by the present invention can be used to evaluate the model to be evaluated, and the score can be used to determine whether it needs to be adjusted, thereby improving the accuracy and efficiency of subsequent data processing.

[0037] It should be noted that the present invention does not specifically limit the information included in the run results, which includes but is not limited to at least one of the following: prediction results, prediction accuracy, run time (e.g., word simulation time, average simulation time), resource consumption, exception handling results, and robustness. Furthermore, the present invention does not limit the specific type of model to be evaluated, which includes but is not limited to at least one of the following: cascade reservoir models, distributed hydrological models, energy system planning models, and numerical weather forecast models.

[0038] Furthermore, the present invention does not specifically limit the size and content of the information to be processed; users can determine this based on the desired application area. For example, if a user wishes to apply the information to water resource management, the information to be processed could include runoff data for a particular region over a year, relevant policies, and possibly some abnormal data to test the processing capabilities of the model to be evaluated. In addition to the data that needs to be processed by each model to be evaluated, the information to be processed could also include its actual results. For example, when using a model to be evaluated to predict data for a specific time period, the actual results could be the actual runoff data for that time period.

[0039] After obtaining the operation results, the information acquisition module can use the TOPSIS mechanism to evaluate the operation results to obtain evaluation information. Specifically, error analysis and / or prediction accuracy measurement can be performed based on the operation results and actual results. Taking the runoff prediction in the field of water resources management as an example, the operation results at this time at least include the predicted value of runoff, and the runoff prediction error is the difference between the runoff value predicted by the model and the actual observed runoff value. For example, a certain model predicts that the runoff of a certain river in a certain period of time is 1000m 3 / s, while the actual measured runoff is 1100m 3 / s, then the prediction error here is 100m 3 / s. Thus, the accuracy evaluation of each model to be evaluated can be obtained.

[0040] If the information to be processed does not include actual results, the information acquisition module can select one or more existing benchmark models or classic algorithms before processing and compare the results of the model to be evaluated with them. In this way, the accuracy of each model to be evaluated can be evaluated.

[0041] The processing efficiency of each model under evaluation can be evaluated based on the model's runtime and / or resource consumption (e.g., computing resource utilization). For example, a model with a shorter single-run processing time can be considered more efficient, and its efficiency rating will also increase accordingly. Furthermore, resource consumption can be combined, for example, by assigning different weights to processing time and resource consumption to obtain an efficiency rating.

[0042] The stability of each model can be determined based on its ability and robustness to handle anomalies. If the model can correctly identify abnormal data and take appropriate action (such as properly filling missing values, correcting or ignoring outliers), and the output is highly accurate and not significantly affected, then the model has strong anomaly handling capabilities and good stability. For example, if a runoff prediction model is fed with clearly abnormal runoff data (such as negative runoff data), and the model can identify and properly handle it, rather than providing unreasonable runoff prediction results, then the model is considered to be highly stable.

[0043] Adaptability primarily refers to a model's ability to adapt and adjust to changes in the external environment, particularly policy changes. In many fields, policy changes can impact related systems and behaviors. For example, in the field of environmental protection, new government pollution emission standards or water resource management policies can affect businesses' production methods and resource utilization, and thus, the variables and relationships involved in the relevant models. Models must be able to respond promptly to these policy changes and accurately reflect the actual situation after policy adjustments.

[0044] The initial evaluation result is the degree of adaptation of the model to be evaluated to the application scenario. In other words, by comprehensively considering the performance of each model to be evaluated on different evaluation dimensions (for example, the weight of each dimension in the application scenario), the model's suitability for a specific application scenario is measured. For example, the initial evaluation result can be understood as the initial score of the model to be evaluated.

[0045] Exemplarily, the information acquisition module may also include an initial evaluation module, which is used to obtain initial evaluation results for each model to be evaluated based on the evaluation information and its evaluation weights. For example, in flood prevention scenarios, data and prediction accuracy are extremely important. Accurate predictions of information such as flood occurrence time, water level, and peak flow directly impact human safety and property damage. Similarly, flood prevention models must maintain stable performance under various complex circumstances. Environmental conditions during floods are complex and changeable, potentially leading to data anomalies and external interference. Therefore, the accuracy weight can be set to 40%, the stability weight to 30%, the efficiency weight to 20%, and the applicability weight to 10%. Calculating with an accuracy rating of 70, a stability rating of 80, an adaptability rating of 90, and an efficiency rating of 90, the initial evaluation result is 79 points. In power generation scenarios, on the other hand, the goal is to fully utilize energy resources and reduce energy waste. Therefore, the efficiency can be increased to 30%, while the accuracy weight can be reduced to 30%, resulting in an initial evaluation result of 81 points. In this way, the weights of various evaluation indicators can be flexibly adjusted according to changes in business needs and application scenarios, enhancing the system's adaptability to multi-scenario requirements.

[0046] It should be noted that the present invention does not specifically limit the specific size and acquisition method of each evaluation weight in different fields. Users can set it themselves based on experience, or they can obtain the weight of each evaluation through environmental information. For example, based on the environmental information in the flood control scenario, a higher water level and a larger flood peak flow can be obtained. At this time, it can be considered that the current environment is unstable. In such an unstable environment, the stability of the model to be evaluated is particularly important. Therefore, the weight of the stability evaluation of the model to be tested can be increased, so that flood-related situations can be analyzed, predicted and evaluated more reliably. Exemplarily, the information acquisition module can input the environmental information obtained by the sensor network into the trained weight evaluation model, and output the evaluation weight corresponding to the environmental information. Wherein, the weight evaluation model includes at least one of the following: hierarchical analysis model, main hierarchical analysis model, fuzzy comprehensive evaluation model, and gray correlation evaluation model.

[0047] The environmental information here can be obtained from a sensor network, which includes but is not limited to at least one of the following: a water level meter, a flow meter, a weather station, and an energy metering device.

[0048] At this time, the target evaluation module can use the Pareto mechanism to process the evaluation information and initial evaluation results obtained by the information acquisition module through the obtained initial evaluation results and evaluation information to obtain the target evaluation results of each model to be evaluated.

[0049] Specifically, the target evaluation module can use the Pareto mechanism to evaluate the evaluation information of each model to be evaluated and determine at least one optimal model. For example, the evaluation scores of each dimension of the model A to be evaluated are: accuracy evaluation 70, stability evaluation 80, adaptability evaluation 90, and efficiency evaluation 90. The evaluation scores of each dimension of the model B to be evaluated are: accuracy evaluation 60, stability evaluation 80, adaptability evaluation 70, and efficiency evaluation 70. The evaluation scores of each dimension of the model C to be evaluated are: accuracy evaluation 80, stability evaluation 80, adaptability evaluation 90, and efficiency evaluation 70. Since the scores of each dimension of the model B to be evaluated are lower than those of the model A to be evaluated, and neither model A nor model C is better than the other in all indicators, the models A and C to be evaluated are the optimal models.

[0050] At this point, the initial evaluation results of the optimal model can be adjusted to further highlight the advantages of the optimal model over other models, so that the final target evaluation results can more accurately reflect the comprehensive performance of the model to be evaluated.

[0051] The present invention does not specifically limit the method for obtaining the adjustment coefficient, which can be a fixed value or obtained by calculation.

[0052] Exemplarily, it can be calculated based on the distance from the prediction results obtained by each optimal model to the positive ideal solution. Among them, the adjustment coefficient is used to regulate the evaluation results of each optimal model, and the adjustment coefficient is greater than 1. The adjustment coefficient and the distance are negatively correlated; the positive ideal solution can be understood as the optimal value of each data. The smaller the distance between the two, the closer the prediction result of the optimal model is to the positive ideal solution, and its performance is better. The larger the adjustment coefficient, the higher the target evaluation result obtained. In this way, for those models whose prediction results are close to the positive ideal solution, a larger adjustment coefficient can further improve their evaluation results and strengthen their advantages over other models. This helps to clarify the objects of focus and priority use among multiple models, and provide more targeted guidance for practical applications.

[0053] Specifically, the target evaluation module performs normalization processing based on the distance between each optimal model and its positive ideal solution to obtain the processing distance of each optimal model; the target evaluation module uses the processing distance and the distance between each optimal model and the positive ideal solution to obtain the adjustment coefficient of each optimal model. For example, the distance between model A to be evaluated and its positive ideal solution is 24, the distance between model B to be evaluated and its positive ideal solution is 13, and the distance between model C to be evaluated and its positive ideal solution is 9. After normalization, the distance between model A to be evaluated and its positive ideal solution is 0.52, model B to be evaluated and its positive ideal solution is 0.28, and model C to be evaluated is 0.20. The adjustment coefficient is the coefficient plus 1, that is, the adjustment coefficient of model A is 1.52, the adjustment coefficient of model B is 1.28, and the adjustment coefficient of model C is 1.20.

[0054] It should be noted that the present invention does not specifically limit the method for obtaining the adjustment coefficient, and the user can decide based on actual conditions.

[0055] The information acquisition module calculates the positive ideal solution and the negative ideal solution of each model to be evaluated; and the information acquisition module obtains evaluation information of each model to be evaluated based on the positive ideal solution and the negative ideal solution.

[0056] Optionally, the system further includes a report generation module, which generates an evaluation report for each model to be evaluated based on the target evaluation result obtained by the target evaluation module; wherein the evaluation report includes at least one of the following: a radar chart, a bar chart; wherein the bar chart includes evaluation information of each model to be evaluated. It can display the performance of each model to be evaluated in four dimensions: accuracy, efficiency, stability and applicability. The present invention can use a radar chart to display the comprehensive evaluation results of each model, a bar chart to compare the performance of different models on the same indicator, and a comparative analysis table to list the score and ranking of each model in detail. In this way, the intuitiveness and readability of the evaluation results can be significantly improved through a variety of visualization methods, providing users with powerful decision-making support.

[0057] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0058] In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "connect," "fixed," etc. should be understood broadly. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0059] In the present invention, unless otherwise expressly specified or limited, when a first feature is "above" or "below" a second feature, it may mean that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Furthermore, when a first feature is "above," "above," or "above" a second feature, it may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is at a higher level than the second feature. When a first feature is "below," "below," or "below" a second feature, it may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is at a lower level than the second feature.

[0060] In the description of this specification, the terms "one embodiment", "some embodiments", "embodiments", "examples", "specific examples" or "some examples" refer to the specific features, structures, materials or characteristics described in conjunction with the embodiment or example and included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and features of different embodiments or examples, unless they are mutually inconsistent.

[0061] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A general evaluation system for model operation results, characterized in that: Applied to the field of energy management, the system includes: A result acquisition module is used to run multiple models to be evaluated, and use each model to be evaluated to process the information to be processed, and obtain the operation results of each model to be evaluated; An information acquisition module is used to evaluate the operation results obtained by the result acquisition module using the TOPSIS mechanism to obtain evaluation information and initial evaluation results of each model to be evaluated; wherein the evaluation information includes at least: accuracy evaluation, stability evaluation, adaptability evaluation and efficiency evaluation; the initial evaluation result is the degree of adaptation of the model to be evaluated to the application scenario; The target evaluation module is used to process the evaluation information and the initial evaluation results obtained by the information acquisition module using the Pareto mechanism to obtain the target evaluation results of each model to be evaluated.

2. The universal evaluation system for model operation results according to claim 1, characterized in that: The information acquisition module also includes: The initial evaluation module is used to obtain the initial evaluation results of each model to be evaluated based on the evaluation information and its evaluation weight.

3. The universal evaluation system for model operation results according to claim 2, characterized in that: The system further includes a sensor network, wherein the sensor network is used to obtain environmental information; wherein the environmental information includes at least one of the following: water conservancy information, energy information, and meteorological information; The information acquisition module is further configured to input the environmental information acquired by the sensor network into the trained weight evaluation model, and output an evaluation weight corresponding to the environmental information; The weight evaluation model includes at least one of the following: a hierarchical analysis model, a principal hierarchical analysis model, a fuzzy comprehensive evaluation model, and a grey correlation evaluation model.

4. The universal evaluation system for model operation results according to claim 1, characterized in that: The target evaluation module is further configured to evaluate the evaluation information of each model to be evaluated using the Pareto mechanism to determine at least one optimal model; The target evaluation module is specifically configured to calculate an adjustment coefficient of each optimal model based on the distance between each optimal model and the positive ideal solution; wherein the adjustment coefficient is negatively correlated with the distance; the adjustment coefficient is used to regulate the evaluation result of each optimal model, and the adjustment coefficient is greater than 1; and The target evaluation results of the respective optimal models are obtained by using the initial evaluation results and the adjustment coefficients of the optimal models.

5. The universal evaluation system for model operation results according to claim 4, characterized in that: The target evaluation module is further configured to perform normalization processing based on the distance between each optimal model and the positive ideal solution to obtain the processing distance of each optimal model; Furthermore, the adjustment coefficient of each optimal model is obtained by utilizing the processing distance and the distance between each optimal model and its corresponding positive ideal solution.

6. The universal evaluation system for model operation results according to claim 1, characterized in that: The information acquisition module is used to calculate the positive ideal solution and the negative ideal solution of each model to be evaluated; The information acquisition module is further configured to obtain evaluation information of each to-be-evaluated model based on the positive ideal solution and the negative ideal solution.

7. The universal evaluation system for model operation results according to claim 1, characterized in that: The information acquisition module is further used to obtain the running time and resource usage information of each model to be evaluated; Furthermore, the running time and the resource usage information are used to perform evaluation to obtain efficiency evaluation of each model to be evaluated.

8. The universal evaluation system for model operation results according to claim 1, characterized in that: The system further includes a report generation module, the report generation module being configured to generate an evaluation report for each model to be evaluated based on the target evaluation result obtained by the target evaluation module; wherein the evaluation report includes at least one of the following: a radar chart, a bar chart; The histogram includes evaluation information of each model to be evaluated.

9. The universal evaluation system for model operation results according to claim 1, characterized in that: The model to be evaluated includes at least one of the following: a cascade reservoir model, a distributed hydrological model, an energy system planning model, and a numerical weather forecast model.

10. The universal evaluation system for model operation results according to claim 1, characterized in that: The result acquisition module is used to obtain the operation results of each model to be evaluated based on the processing time and / or resource consumption of the information to be processed of each model to be evaluated.

Images