New energy asset business insight evaluation method and system for operation and maintenance investment decision

Abstract

The application discloses a new energy asset business insight evaluation method and system for operation and maintenance investment decision, and particularly relates to the technical field of data analysis and decision support, and the method comprises the following steps: acquiring multi-source heterogeneous data, and performing standardized storage and association according to a three-dimensional structure of asset identification, time dimension and data type; based on the standardized data, a quantitative evaluation result is calculated and generated through a set of preset evaluation models with a synergistic feedback mechanism; based on the quantitative evaluation result, a decision suggestion for operation and maintenance and investment is generated through a preset decision algorithm; the evaluation result and the decision suggestion are output through an integrated visual display module, and a closed loop of evaluation-decision-execution-feedback-optimization is formed. The system comprises corresponding modules. The application realizes data-driven quantitative evaluation and intelligent decision for the whole life cycle of assets, and solves the problems of information dispersion, static model and decision lag in traditional asset management.

Description

Technical Field

[0001] This invention relates to the field of data analysis and decision support technology, and more specifically, to a method and system for evaluating the business insights of new energy assets for operation and maintenance investment decisions. Background Technology

[0002] Currently, in the field of asset management for new energy equipment such as charging piles for new energy vehicles, the publicly available charging pile asset management solutions mainly use two-dimensional relational databases to store equipment operation data. The evaluation indicators are limited to failure rate and utilization rate. The models are static and cannot integrate external environmental variables, and the decision-making relies on human experience.

[0003] However, in actual use, it still has some shortcomings, such as: on the one hand, multi-source heterogeneous data (operational data, environmental data, business data) are stored in a scattered manner and lack a unified three-dimensional data structure that supports spatiotemporal correlation queries, resulting in insufficient data value mining.

[0004] On the other hand, the evaluation models are mostly static, single-task models, such as simply predicting failures or calculating revenues. The models operate in isolation and cannot collaboratively output comprehensive business insights (such as scores that integrate health, value and risk). Furthermore, the model parameters cannot be dynamically optimized based on decision feedback. The evaluation results are separated from the decision recommendations, and there is a lack of a closed-loop link for automated decision generation and execution tracking, resulting in delayed management response and the inability to continuously optimize the value of assets throughout their entire lifecycle.

[0005] To address the above issues, we propose a business insight assessment method and system for new energy assets aimed at operation and maintenance investment decisions. Summary of the Invention

[0006] To overcome the aforementioned deficiencies in the prior art, embodiments of the present invention provide a method and system for evaluating the business insights of new energy assets for operation and maintenance investment decisions. The following solutions address the problems of low data integration efficiency and isolated and static evaluation models mentioned in the background art.

[0007] on the one hand

[0008] This invention provides the following technical solution: a business insight assessment method for new energy assets oriented towards operation and maintenance investment decisions, comprising:

[0009] S1. Data Integration: Acquire and process multi-source heterogeneous data of new energy equipment assets, associate the multi-source heterogeneous data, and store it in a standardized manner according to the three-dimensional structure of asset identification, time dimension and data type;

[0010] S2. Quantitative Business Insights: Based on the standardized stored data, a set of preset evaluation models are used to calculate and generate quantitative evaluation results; wherein, the evaluation models include: business value evaluation model, equipment health evaluation model, investment return prediction model, and risk warning model;

[0011] Specifically, the investment return prediction model calls the output results of the business value assessment model; the risk warning model calls the output results of the equipment health assessment model; and the parameters of the set of assessment models can be collaboratively optimized based on feedback data.

[0012] S3. Decision Support Generation: Based on the quantitative evaluation results, a pre-set decision algorithm is used to generate decision suggestions for asset operation and maintenance and for asset investment.

[0013] S4. Decision Result Output: Output the quantitative evaluation results and decision recommendations to a display module for integrated and visual display.

[0014] By adopting the above technical solutions, the problems of "data silos" caused by the dispersion of multi-source heterogeneous data and inconsistent standards in the existing new energy equipment asset management, the single and isolated nature of traditional assessment methods, the lack of linkage between business and operation and maintenance perspectives, the reliance on experience for decision-making, and the lack of data-driven and intuitive presentation are overcome.

[0015] Furthermore, in S1, standardized storage is performed according to a three-dimensional structure of asset identification, time dimension, and data type, specifically as follows:

[0016] Establish a data warehouse structure with asset identification as the first index dimension, time point as the second index dimension, and data category as the third index dimension;

[0017] The three-dimensional structure is used to support multi-dimensional correlation queries and traceability based on the entire life cycle of assets, providing a structured data foundation for the set of preset evaluation models.

[0018] By adopting the above technical solutions, a three-dimensional integrated data warehouse structure of "asset-time-data type" was established, realizing the efficient integration of multi-source heterogeneous data and multi-dimensional correlation query and traceability based on the entire asset life cycle. This provides a high-quality, structured, and unified data foundation for a series of subsequent evaluation models, and solves the problem of "data silos" where information is scattered and difficult to correlate and analyze.

[0019] Furthermore, the business value assessment model is a multi-dimensional aggregation assessment model, which generates the business value score by aggregating the quantitative results of the operational efficiency dimension, growth potential dimension, and risk dimension.

[0020] The quantitative results of the operational efficiency dimension are calculated based on multiple operational performance indicators of the new energy equipment assets within a preset time period;

[0021] The quantitative results of the growth potential dimension are derived from market demand data, development planning data, and competitive environment data of the target region through trend prediction analysis.

[0022] The quantitative results of the aforementioned risk dimensions are derived from an analysis of the impact of multiple risk factors, including policy, market, environment, and technology.

[0023] By adopting the above technical solutions, a multi-dimensional aggregated evaluation model is constructed, which integrates the originally isolated operations, growth and risk analysis into a unified quantitative score of business value. This overcomes the drawbacks of the traditional evaluation perspective being singular and static, and provides comprehensive, dynamic and horizontally comparable data for asset investment decisions.

[0024] Furthermore, the device health assessment model is a multi-dimensional collaborative scoring model, which generates the device health score by scoring the hardware status dimension, the operational stability dimension, and the maintenance level dimension.

[0025] The hardware status dimension score is derived from the equipment's years of operation, core component wear data, and historical fault data.

[0026] The operational stability score is derived from the equipment's fault interval data, operating parameter fluctuation data, and downtime data.

[0027] The maintenance level score is based on an evaluation of maintenance timeliness data, spare parts replacement quality data, and preventive maintenance coverage data.

[0028] By adopting the above technical solutions, a multi-dimensional collaborative scoring model integrating "hardware status, operational stability, and maintenance level" is constructed, enabling a comprehensive and refined assessment of equipment health. This method overcomes the limitations of traditional single-indicator (such as failure rate) assessments, comprehensively quantifying inherent equipment attributes, dynamic operational performance, and human maintenance factors. This allows for earlier and more accurate identification of potential equipment risks and performance degradation trends, providing reliable data for predictive maintenance and precise resource allocation.

[0029] Furthermore, the investment return prediction model is a dynamic multi-scenario prediction model, and the process of generating the investment return prediction value includes:

[0030] Based on the standardized historical data, determine the core financial parameters required for forecasting;

[0031] Based on the aforementioned core financial parameters, a baseline scenario prediction value is calculated and generated.

[0032] By combining real-time acquired external variable data, the core financial parameters are dynamically adjusted, and multiple scenario prediction values ​​corresponding to different external conditions are generated simultaneously.

[0033] By adopting the above technical solutions, the drawbacks of traditional models, such as fixed parameters and inability to adapt to real-time fluctuations in external conditions such as market and policy, are overcome. By generating benchmark values ​​and predicted values ​​under various scenarios, more flexible and forward-looking quantitative basis is provided for investment decisions, significantly improving the robustness of decisions and risk response capabilities.

[0034] Furthermore, the risk warning model is a multi-level response warning model, which monitors risk indicators of multiple categories such as operation, investment, policy and maintenance in parallel, and triggers different levels of warnings based on the level at which the monitored indicator values ​​exceed preset thresholds; and associates corresponding response handling suggestions and decision execution processes.

[0035] By adopting the above technical solutions, comprehensive, hierarchical, and automated monitoring of multiple types of risks has been achieved, and differentiated response measures can be directly triggered, forming a complete risk warning and disposal closed loop, which significantly improves the accuracy and response efficiency of risk management.

[0036] Furthermore, S3 generates decision recommendations for asset operation and maintenance, specifically including:

[0037] Based on the equipment health score and historical fault data, a predictive maintenance plan is generated for the specific equipment.

[0038] The predictive maintenance plan generates maintenance instructions that include maintenance items, execution time, and suggested resources by associating them with a preset maintenance strategy rule base based on the segmented intervals of the equipment health score.

[0039] By adopting the above technical solutions, a closed loop from "assessing and identifying problems" to "automatically generating precise maintenance plans" has been achieved, thereby improving the automation level of operation and maintenance decisions and the accuracy of resource allocation, effectively avoiding "over-maintenance" or "under-maintenance", ensuring equipment reliability and optimizing operation and maintenance costs.

[0040] Furthermore, the S3 process generates decision recommendations for asset investment, specifically including:

[0041] Based on the target area's commercial value score, projected return on investment, and competitive environment data, site selection assessment recommendations for new asset locations are generated.

[0042] The site selection assessment recommendations are generated through a site selection scoring model. The site selection scoring model takes business value score and investment return prediction as core input parameters, and combines geographic information system spatial analysis technology to quantify and calibrate competitive environment data, and outputs quantified site selection scores and corresponding equipment configuration recommendations.

[0043] By adopting the above technical solutions, integrating business value scoring, investment return prediction and quantitative analysis of the competitive environment, and combining spatial calibration with geographic information systems, data-driven and intelligent decision support for investment site selection of new energy equipment has been achieved.

[0044] on the other hand

[0045] A business insight assessment system for new energy assets aimed at operation and maintenance investment decisions, including:

[0046] Data integration module: used to acquire and process multi-source heterogeneous data of new energy equipment assets. The data integration module includes a data fusion engine, used to clean and associate the multi-source heterogeneous data, and to standardize and store it according to a three-dimensional structure of asset identification, time dimension and data type.

[0047] Business Insight Assessment Module: Connected to the data integration module, it is used to execute a set of preset assessment models to generate quantitative assessment results; the business insight assessment module includes: a business value assessment unit, an equipment health scoring unit, an investment return prediction unit, and a risk warning unit;

[0048] Decision support module: Connected to the business insight assessment module, it is used to generate decision suggestions for asset operation and maintenance and decision suggestions for asset investment based on the quantitative assessment results and through a preset decision algorithm;

[0049] Display module: Connected to the business insight assessment module and decision support module, it is used to integrate and visualize the quantitative assessment results and decision recommendations;

[0050] Model Iteration and Optimization Module: Connected to the Business Insight Assessment Module, Decision Support Module, and Data Integration Module respectively, this module is used to periodically collect the deviations between assessment results, decision execution feedback, and actual operational data. It updates the parameters of the set of preset assessment models and decision algorithms through parameter optimization algorithms, forming a closed-loop learning mechanism.

[0051] By adopting the above technical solutions, the system achieves self-learning and continuous evolution, overcoming the problems of fixed model parameters and inability to adapt to dynamic business environments in traditional asset management. This module collects actual operational data and decision execution feedback, and automatically adjusts the parameters of the evaluation model and decision-making algorithm through algorithms, forming an intelligent closed loop of "evaluation-decision-feedback-optimization," thereby continuously improving the system's evaluation accuracy, decision adaptability, and long-term business value.

[0052] The technical effects and advantages of this invention are as follows:

[0053] 1. This invention constructs a three-dimensional integrated data warehouse of "asset-time-data type", which realizes efficient fusion and spatiotemporal correlation query of multi-source heterogeneous data, provides a high-quality and structured data foundation for the upper-level model, and solves the problem of information dispersion and fragmentation;

[0054] 2. This invention designs and coordinates the operation of a set of evaluation models with feedback mechanisms (business value, health, return on investment, risk warning), transforming isolated data into multidimensional, dynamic, and quantifiable business insight indicators, overcoming the drawbacks of single-model evaluation perspective limitations and inability to adapt to changes in the external environment.

[0055] 3. This invention constructs a complete chain from quantitative assessment to automatic generation of decision recommendations, and then to visualization and execution feedback. It also introduces a model iteration and optimization module, forming an intelligent closed-loop management system of "assessment-decision-execution-feedback-optimization". This achieves proactive, precise and continuous optimization of asset management, and significantly enhances the commercial value of assets throughout their entire life cycle. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0057] Figure 2 This is a schematic diagram of the system flow of the present invention;

[0058] Figure 3 This is a schematic diagram of the business insight quantification process of the present invention;

[0059] Figure 4 This is a schematic diagram of the decision support and closed-loop feedback process of the present invention;

[0060] Figure 5 This is a schematic diagram of the overall system modules of the present invention.

[0061] In the diagram: 1. Data integration module; 2. Business insight assessment module; 3. Decision support module; 4. Display module; 5. Model iteration and optimization module. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] This embodiment uses the asset management of urban electric vehicle charging pile networks as an application scenario to illustrate the new energy asset business insight assessment method and system for operation and maintenance investment decision-making provided by the present invention.

[0064] Example 1

[0065] refer to Figure 1-4 A business insight assessment methodology for new energy assets aimed at operation and maintenance investment decisions includes:

[0066] S1. Data Integration: Acquire and process multi-source heterogeneous data of new energy equipment assets, associate the multi-source heterogeneous data, and store it in a standardized manner according to the three-dimensional structure of asset identification, time dimension and data type;

[0067] It should be specifically noted that the multi-source heterogeneous data includes asset performance data, business environment data, cost-benefit data, and external impact data.

[0068] It should be further explained that the asset performance data is collected from the historical operating performance data (average daily revenue per pile, efficiency per square meter, load rate, transaction success rate, energy efficiency ratio) of 1,200 charging piles in the city over the past 3 years, as well as equipment operation data (fault records, running time, maintenance frequency, and wear data of core components). The data comes from real-time uploads and historical archives of the charging pile operation management system.

[0069] Business environment data is obtained through geographic information platforms, including latitude and longitude of each location, population density within 1 kilometer, and traffic flow data; data on regional new energy vehicle ownership and charging demand growth rate are obtained from urban traffic management departments and new energy vehicle industry associations; data on the distribution of charging piles of surrounding competitors and user spending power are collected through market research institutions; and local subsidy policies, charging price control policies, land use policy texts and update records are retrieved from government information disclosure platforms.

[0070] Cost and revenue data are obtained from the enterprise's purchasing department for equipment procurement costs and spare parts procurement costs; from construction partners for detailed installation and construction costs; from the finance department for data on land lease costs, electricity expenses, maintenance personnel salaries, network communication costs, etc.; and from the operations department for revenue data such as charging service fees, advertising cooperation revenue, and value-added service revenue.

[0071] External impact data is obtained through a macroeconomic data platform to track electricity price fluctuations and the growth rate of the new energy vehicle industry; data on the frequency of extreme weather events, rainfall, and number of high-temperature days over the past three years are obtained from meteorological departments; and information on emergencies such as regional traffic control and large-scale events is collected through an urban event early warning platform.

[0072] It should be specifically noted that S1 uses a standardized storage method based on a three-dimensional structure of asset identification, time dimension, and data type, specifically as follows:

[0073] Establish a data warehouse structure with asset identification as the first index dimension, time point as the second index dimension, and data category as the third index dimension;

[0074] The three-dimensional structure is used to support multi-dimensional correlation queries and traceability based on the entire life cycle of assets, providing a structured data foundation for the set of preset evaluation models.

[0075] It should be further explained that the data cleaning process employs outlier detection algorithms (such as the 3σ principle) to remove extreme outliers in indicators such as average daily revenue per pile and load rate; it fills in short-term data gaps at some locations using a missing value completion algorithm (based on the average data of similar equipment in the same region during the same period); and it segments and extracts keywords from unstructured data such as policy texts and event information, transforming them into quantifiable tag data, such as subsidy policy strength tags: high / medium / low, corresponding to quantitative values ​​of 1 / 0.5 / 0 respectively.

[0076] Data association is achieved through a data fusion engine, using "charging pile asset number" as the primary key to link various performance data, operation data, and cost-benefit data of the same asset; "time stamp" is used as an auxiliary linking item to achieve alignment of data from different sources in the time dimension, such as linking traffic flow data for a certain period with charging pile usage frequency data for the same period.

[0077] Standardized storage establishes a data warehouse structure with asset identification as the first index dimension, time point as the second index dimension, and data category as the third index dimension. For example, the index "Asset No. A001-May 10, 2024-Operational Performance Data" corresponds to storing specific indicator values ​​such as the average daily revenue per charging pile and load rate on a specified date. This three-dimensional structure supports multi-dimensional correlation queries and traceability based on the entire life cycle of the charging pile, ensuring a consistent, complete, and timely structured data foundation for subsequent evaluation models.

[0078] S2. Quantitative Business Insights: Based on the standardized stored data, a set of preset evaluation models are used to calculate and generate quantitative evaluation results; wherein, the evaluation models include: business value evaluation model, equipment health evaluation model, investment return prediction model, and risk warning model;

[0079] Specifically, the investment return prediction model calls the output results of the business value assessment model; the risk warning model calls the output results of the equipment health assessment model; and the parameters of the set of assessment models can be collaboratively optimized based on feedback data.

[0080] It should be specifically noted that the business value assessment model is a multi-dimensional aggregate assessment model, which generates the business value score by aggregating the quantitative results of the operational efficiency dimension, growth potential dimension, and risk dimension.

[0081] The quantitative results of the operational efficiency dimension are calculated based on multiple operational performance indicators of the new energy equipment assets within a preset time period;

[0082] The quantitative results of the growth potential dimension are derived from market demand data, development planning data, and competitive environment data of the target region through trend prediction analysis.

[0083] The quantitative results of the aforementioned risk dimensions are derived from an analysis of the impact of multiple risk factors, including policy, market, environment, and technology.

[0084] It should be further explained that the business valuation model adopts a three-dimensional evaluation framework of "operational efficiency + growth potential + risk coefficient", and its technical implementation includes the following steps:

[0085] Input feature engineering extracts the operational indicator series for the past 12 months from the 3D data warehouse and calculates the operational efficiency score (OE); extracts regional market, policy, and competition data, predicts trends through a linear regression model, and calculates the growth potential score (GP); extracts policy texts, market fluctuations, and environmental event data, and uses the analytic hierarchy process (AHP) to quantify the risk coefficient (R).

[0086] Aggregate calculation, business value score ,in, , , For the corresponding weighting coefficients, , , The default values ​​are 0.4, 0.3, and 0.3, respectively, determined by domain experts using the Delphi method, and can be adjusted through the system interface; by uniformly weighting and integrating indicators of different dimensions and sources, a standardized business value score of 0-100 is output, providing a unified scale for horizontal comparison of assets.

[0087] It should be specifically noted that the device health assessment model is a multi-dimensional collaborative scoring model, which generates the device health score by scoring the hardware status dimension, the operational stability dimension, and the operation and maintenance level dimension.

[0088] The hardware status dimension score is derived from the equipment's years of operation, core component wear data, and historical fault data.

[0089] The operational stability score is derived from the equipment's fault interval data, operating parameter fluctuation data, and downtime data.

[0090] The maintenance level score is based on an evaluation of maintenance timeliness data, spare parts replacement quality data, and preventive maintenance coverage data.

[0091] It should be further explained that the equipment health assessment model constructs a three-dimensional assessment framework of "hardware status + operational stability + maintenance level". The scores for each dimension are normalized to generate a health score of 0-100. A score ≥85 is "Excellent", 60-84 is "Good", 40-59 is "Average", and a score <40 is "Poor". The specific scoring logic is as follows:

[0092] By calculating the hardware status dimension score The aging coefficient A is a weighted function of three sub-indicators: years of operation, wear rate of core components, and frequency of historical failures; the weighting is based on historical failure root cause analysis data.

[0093] Calculate the runtime stability dimension score Among them, the stability coefficient It is calculated by combining the mean time between failures (MTBF), the variance of operating parameters (voltage, current), and the proportion of unplanned downtime;

[0094] Calculate the operation and maintenance level score Among them, the operation and maintenance support coefficient O is calculated based on data such as the timely response rate of maintenance work orders, the completion rate of preventive maintenance plans, and the recurrence rate of faults after spare parts replacement;

[0095] The final health score is obtained. The model automatically outputs labels such as "excellent", "good", "average" and "poor" based on the scoring threshold, and these labels directly drive subsequent predictive maintenance decisions.

[0096] It should be specifically noted that the investment return prediction model is a dynamic multi-scenario prediction model, and the process of generating the investment return prediction value includes:

[0097] Based on the standardized historical data, determine the core financial parameters required for forecasting;

[0098] Based on the aforementioned core financial parameters, a baseline scenario prediction value is calculated and generated.

[0099] By combining real-time acquired external variable data, the core financial parameters are dynamically adjusted, and multiple scenario prediction values ​​corresponding to different external conditions are generated simultaneously.

[0100] It should be further explained that the specific implementation steps of the investment return prediction model are as follows:

[0101] First, extract the initial total investment I, historical average annual operating cost Cc, and historical average annual revenue Rc for a specific asset or target area from the data warehouse;

[0102] Then, the growth potential dimension score GP output by the business value assessment model is called, and converted into a growth potential coefficient g through a regression equation fitted with historical data: Where a and b are regression coefficients fitted based on historical data, used to calculate the annual expected return of the benchmark scenario. Substitute into the formula The baseline ROI value is obtained, where T is the number of years of operation.

[0103] It should be specifically noted that the risk warning model is a multi-level response warning model, which monitors risk indicators in multiple categories such as operation, investment, policy and maintenance in parallel; triggers different levels of warnings based on the level at which the monitored indicator values ​​exceed preset thresholds; and associates corresponding response handling suggestions and decision execution processes.

[0104] It should be further explained that the risk indicators and thresholds are set as follows:

[0105] Operational risks: Equipment health score below 40, load rate below 30% for 3 consecutive months, and revenue decline of ≥15% for 3 consecutive months. The thresholds are set based on the average failure loss data and profit bottom line analysis of the industry over the past 3 years. Equipment health score of 40 corresponds to the point of sharp increase in failure rate, and load rate of 30% is the break-even critical load rate.

[0106] Investment risks: The ROI of the new site is predicted to be lower than the industry benchmark by 8%, and the growth rate of regional competitors exceeds 50%. The industry benchmark of 8% is based on the average return on investment of the charging pile industry in the past 3 years. The growth rate of competitors by 50% corresponds to the critical value of market share being rapidly eroded.

[0107] Policy risks: The introduction of subsidy reduction policies and adjustments to charging price controls by ≥10% are based on sensitivity analysis of the impact of policy changes on revenue.

[0108] Operational risks: Core spare parts inventory is less than 30 days' worth of usage, and the maintenance team's response efficiency decreases by ≥20%. The threshold is set based on the spare parts procurement cycle and fault response standards. The industry average response time is 2 hours, and a 20% decrease corresponds to a significant increase in the risk of response timeout.

[0109] It should be further explained that the warnings are divided into the following levels:

[0110] Level 1 Warning (High Risk): Triggers emergency handling procedures, such as emergency equipment replacement and suspension of investment in new sites. It is applicable to scenarios where indicators seriously exceed the threshold and the impact is irreversible (such as equipment health score of 35 points and new site ROI prediction of 5%).

[0111] Level 2 Warning (Medium Risk): Triggers optimization measures, such as strengthening maintenance and adjusting operational strategies. Applicable to scenarios where indicators are close to the threshold or the impact can be mitigated through adjustments (e.g., load rate of 32% or revenue decline of 14%).

[0112] Level 3 Warning (Low Risk): Only provides a suggestive display and is applicable to scenarios where indicators slightly deviate from the threshold and have a minor impact (such as 35 days' worth of core spare parts inventory or a 10% decrease in maintenance response efficiency).

[0113] The system monitors relevant indicators in real time, updates data hourly, and automatically generates early warning information and associates corresponding response and handling suggestions when indicators reach preset thresholds, pointing to specific decision-making and execution processes.

[0114] S3. Decision Support Generation: Based on the quantitative evaluation results, a pre-set decision algorithm is used to generate decision suggestions for asset operation and maintenance and for asset investment.

[0115] It should be specifically noted that the decision recommendations generated in S3 for asset operation and maintenance include:

[0116] Based on the equipment health score and historical fault data, a predictive maintenance plan is generated for the specific equipment.

[0117] The predictive maintenance plan generates maintenance instructions that include maintenance items, execution time, and suggested resources by associating them with a preset maintenance strategy rule base based on the segmented intervals of the equipment health score.

[0118] It should be further explained that, based on equipment health scores and historical fault data, predictive maintenance plans are generated. The specific process is as follows:

[0119] Maintenance strategy rule base establishment: Maintenance strategies are preset according to equipment health score segments. The rule base is built based on equipment maintenance effectiveness data and cost-benefit analysis over the past 3 years.

[0120] Excellent (≥85 points): Maintain the regular maintenance frequency (monthly inspection), and configure spare parts reserves at 80% of the regular usage, based on the statistical data that the equipment failure rate in this scoring segment is less than 5%;

[0121] Good (60-84 points): Inspection once every half month, focusing on monitoring the operating status of core components, and spare parts are reserved at 100% of the normal usage, based on the characteristic of equipment failure rate of 10%-15% in this scoring range;

[0122] General (40-59 points): Weekly inspection, increase the inspection items of core components, replace components with obvious signs of aging in advance, and configure spare parts reserves at 150% of the regular usage. The equipment failure rate in this scoring range is 20%-30%, and preventive maintenance needs to be strengthened.

[0123] Poor (<40 points): Suspend operations for comprehensive overhaul, replace core faulty components or assess the feasibility of replacement, and allocate spare parts reserves according to emergency needs. The equipment failure rate in this scoring segment exceeds 50%, and the losses from continued operation outweigh the maintenance costs.

[0124] The system polls the device health score H in real time. When H falls into the "normal" range (40-59 points), it automatically queries the device's historical fault database, matches frequently failed components, and then calls the rule engine to generate maintenance instructions containing specific components, recommended replacement time, and required spare parts based on the rule database (e.g., "if the health score is normal and the charging module failure frequency is >3 times / year, it is recommended to replace it within 1 month"). These instructions are then automatically dispatched to the operation and maintenance management system via an interface.

[0125] Spare parts demand forecasting: This is an independent forecasting sub-model. The inputs are the health distribution of all equipment, historical spare parts consumption data, and generated maintenance plans. The model uses time series forecasting methods (such as seasonal ARIMA) and outputs a forecast list of the demand for various spare parts in the next 3 months, providing data input for the procurement system.

[0126] It should be specifically noted that the decision-making recommendations for asset investment generated in S3 include:

[0127] Based on the target area's commercial value score, projected return on investment, and competitive environment data, site selection assessment recommendations for new asset locations are generated.

[0128] The site selection assessment recommendations are generated through a site selection scoring model. The site selection scoring model takes business value score and investment return prediction as core input parameters, and combines geographic information system spatial analysis technology to quantify and calibrate competitive environment data, and outputs quantified site selection scores and corresponding equipment configuration recommendations.

[0129] It should be further explained that the site selection scoring model technically integrates GIS spatial analysis capabilities. Its execution process is as follows: obtain the target area's commercial value score (SV) and return on investment score (S). ROI (Normalized from ROI predicted value to 0-100 points), competitive pressure score S compInput; acquire competitor location geographic data through a GIS engine, perform kernel density analysis, and generate a competitive pressure quantification layer; perform weighted calculations: Areas with a score of 70 or higher will be recommended, along with device configuration suggestions;

[0130] It should be further explained that the weighting coefficients of the site selection scoring model are based on a retrospective analysis of 500 charging piles built in the past 3 years. The model uses a multiple linear regression method to analyze the impact of business value score, investment return prediction, and competitive environment on actual operating revenue, and obtains the standardized regression coefficients of each factor. The system automatically collects actual operating data of new sites every quarter and recalculates the impact coefficients of each factor on actual revenue. If the change exceeds 10%, it is recommended to adjust the weights. Users can also manually adjust the weights in the system interface according to the characteristics of specific areas, with an adjustment range of ±0.1.

[0131] Priority analysis of aging equipment replacement: based on equipment health score H and industry benchmark ROI ( ), Current equipment ROI ( ), ROI improvement rate of new equipment ( ),calculate The system periodically calculates and sorts the scores of all devices to provide data support for the annual capital expenditure plan.

[0132] It should be further explained that the default weight values ​​were obtained through regression analysis using historical data from 200 equipment replacement decisions over the past 5 years, with the overall benefits (failure reduction rate + benefit increase rate) after actual replacement as the target.

[0133] S4. Decision Result Output: Output the quantitative evaluation results and decision recommendations to a display module for integrated and visual display.

[0134] It should be further explained that the display module includes a panoramic view of asset performance, core business indicators, a list of decision-making recommendations, and a full lifecycle view. Figure 4 The core module supports custom configuration and interactive functions:

[0135] Asset Performance Panorama: Visualizes the commercial value score, health score, and real-time load rate of all charging pile locations on a map. It supports filtering by region (such as city division) and rating level (such as excellent / good commercial value). Clicking on a specific location allows you to drill down and view detailed evaluation data for that location (such as operational efficiency score and fault records).

[0136] Core business metrics: Dynamically displays core KPIs such as overall asset ROI, average health score, investment return forecast, and number of risk warnings. It supports year-on-year (compared to the same period last year) and month-on-month (compared to the previous month) comparisons. The data update frequency is real-time (core metrics) and daily (aggregate metrics such as ROI).

[0137] Decision Recommendation List: Displays operation and maintenance optimization suggestions, investment decision suggestions, and risk management suggestions by priority (operation and maintenance suggestions prioritize high-risk equipment maintenance suggestions, and investment suggestions prioritize high-scoring site selection suggestions). Supports viewing suggestion details (such as maintenance plan details and site selection scoring basis) and execution status (not executed / in execution / completed).

[0138] Full lifecycle view: Displays the core indicators and decision-making points of each stage in three phases: site selection and deployment, operation and maintenance, and replacement and recycling. It supports drill-down to view the evaluation data and decision-making trajectory of specific locations / equipment at each stage.

[0139] Interaction and push functions: Supports exporting dashboard data (Excel / PDF format). Core decision suggestions and risk warnings can be automatically pushed to the management's mobile / PC terminals via email and system messages. The push frequency is set according to the warning level (level 1 warning is pushed in real time, level 2 warning is pushed daily, and level 3 warning is pushed weekly).

[0140] It should be further explained that, in summary, this invention constitutes a dynamically evolving intelligent closed-loop system; its core process is embodied in three interlocking technical cycles: a data-driven evaluation cycle: based on a three-dimensional data warehouse, four major evaluation models work together to generate quantitative insights; an automatic decision-making execution cycle: the insight results trigger pre-set algorithms to generate and push executable operation and maintenance and investment instructions; and a feedback parameter optimization cycle: the model iteration optimization module collects real business data after instruction execution, compares it with the model's predicted values, and dynamically adjusts the adjustable parameters of each model (such as business value weights α, β, γ, and the investment return growth mapping function) through optimization algorithms; these three cycles operate continuously, enabling the system's evaluation accuracy and decision-making effectiveness to continuously improve over time.

[0141] Example 2

[0142] refer to Figure 5 A business insight and evaluation system for new energy assets aimed at operation and maintenance investment decisions, specifically including:

[0143] Data integration module: used to acquire and process multi-source heterogeneous data of new energy equipment assets. The data integration module includes a data fusion engine, used to clean and associate the multi-source heterogeneous data, and to standardize and store it according to a three-dimensional structure of asset identification, time dimension and data type.

[0144] Business Insight Assessment Module: Connected to the data integration module, it is used to execute a set of preset assessment models to generate quantitative assessment results; the business insight assessment module includes: a business value assessment unit, an equipment health scoring unit, an investment return prediction unit, and a risk warning unit;

[0145] Decision support module: Connected to the business insight assessment module, it is used to generate decision suggestions for asset operation and maintenance and decision suggestions for asset investment based on the quantitative assessment results and through a preset decision algorithm;

[0146] Display module: Connected to the business insight assessment module and decision support module, it is used to integrate and visualize the quantitative assessment results and decision recommendations;

[0147] Model Iteration and Optimization Module: Connected to the Business Insight Assessment Module, Decision Support Module, and Data Integration Module respectively, this module is used to periodically collect the deviations between assessment results, decision execution feedback, and actual operational data. It updates the parameters of the set of preset assessment models and decision algorithms through parameter optimization algorithms, forming a closed-loop learning mechanism.

[0148] It should be further explained that the model iteration and optimization module regularly (e.g., quarterly) collects real data from the business system after the decision is implemented, such as "the change in failure rate after replacing core components" and "the actual ROI of new investment locations". The module compares the real data with the model's previous predictions / evaluations, calculates the deviation, and uses this deviation data to fine-tune adjustable parameters such as the weights (α, β, γ) in the business value assessment model and the growth coefficient mapping function in the investment return prediction model, through optimization algorithms such as gradient descent. This makes the model predictions more in line with actual business development, thereby achieving the system's self-learning and continuous optimization.

[0149] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0150] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A business insight assessment method for new energy assets oriented towards operation and maintenance investment decisions, characterized in that, Includes the following steps: S1. Data Integration: Acquire and process multi-source heterogeneous data of new energy equipment assets, associate the multi-source heterogeneous data, and store it in a standardized manner according to the three-dimensional structure of asset identification, time dimension and data type; S2. Quantitative Business Insights: Based on the standardized stored data, a set of preset evaluation models are used to calculate and generate quantitative evaluation results; wherein, the evaluation models include: business value evaluation model, equipment health evaluation model, investment return prediction model, and risk warning model; Specifically, the investment return prediction model calls the output results of the business value assessment model; the risk warning model calls the output results of the equipment health assessment model; and the parameters of the set of assessment models can be collaboratively optimized based on feedback data. S3. Decision Support Generation: Based on the quantitative evaluation results, a pre-set decision algorithm is used to generate decision suggestions for asset operation and maintenance and for asset investment. S4. Decision Result Output: Output the quantitative evaluation results and decision recommendations to a display module for integrated and visual display.

2. The new energy asset business insight assessment method for operation and maintenance investment decision-making as described in claim 1, characterized in that: S1 is standardized and stored according to a three-dimensional structure of asset identification, time dimension, and data type, specifically as follows: Establish a data warehouse structure with asset identification as the first index dimension, time point as the second index dimension, and data category as the third index dimension; The three-dimensional structure is used to support multi-dimensional correlation queries and traceability based on the entire life cycle of assets, providing a structured data foundation for the set of preset evaluation models.

3. The new energy asset business insight assessment method for operation and maintenance investment decision-making as described in claim 1, characterized in that: The business value assessment model is a multi-dimensional aggregate assessment model, which generates the business value score by aggregating the quantitative results of the operational efficiency dimension, growth potential dimension, and risk dimension. The quantitative results of the operational efficiency dimension are calculated based on multiple operational performance indicators of the new energy equipment assets within a preset time period; The quantitative results of the growth potential dimension are derived from market demand data, development planning data, and competitive environment data of the target region through trend prediction analysis. The quantitative results of the aforementioned risk dimensions are derived from an analysis of the impact of multiple risk factors, including policy, market, environment, and technology.

4. The new energy asset business insight assessment method for operation and maintenance investment decision-making as described in claim 1, characterized in that: The device health assessment model is a multi-dimensional collaborative scoring model, which generates the device health score by scoring the hardware status dimension, the operational stability dimension, and the maintenance level dimension. The hardware status dimension score is derived from the equipment's years of operation, core component wear data, and historical fault data. The operational stability score is derived from the equipment's fault interval data, operating parameter fluctuation data, and downtime data. The maintenance level score is based on an evaluation of maintenance timeliness data, spare parts replacement quality data, and preventive maintenance coverage data.

5. The new energy asset business insight assessment method for operation and maintenance investment decision-making as described in claim 1, characterized in that: The investment return prediction model is a dynamic multi-scenario prediction model, and the process of generating the investment return prediction value includes: Based on the standardized historical data, determine the core financial parameters required for forecasting; Based on the aforementioned core financial parameters, a baseline scenario prediction value is calculated and generated. By combining real-time acquired external variable data, the core financial parameters are dynamically adjusted, and multiple scenario prediction values ​​corresponding to different external conditions are generated simultaneously.

6. The new energy asset business insight assessment method for operation and maintenance investment decision-making as described in claim 1, characterized in that: The risk warning model is a multi-level response warning model, which monitors risk indicators in multiple categories such as operation, investment, policy and maintenance in parallel. It triggers different levels of warnings based on the level at which the monitored indicator values ​​exceed preset thresholds, and associates them with corresponding response handling suggestions and decision execution processes.

7. The new energy asset business insight assessment method for operation and maintenance investment decision-making as described in claim 1, characterized in that: The S3 process generates decision recommendations for asset operation and maintenance, specifically including: Based on the equipment health score and historical fault data, a predictive maintenance plan is generated for the specific equipment. The predictive maintenance plan generates maintenance instructions that include maintenance items, execution time, and suggested resources by associating them with a preset maintenance strategy rule base based on the segmented intervals of the equipment health score.

8. The new energy asset business insight assessment method for operation and maintenance investment decision-making as described in claim 1, characterized in that: The S3 process generates decision recommendations for asset investment, specifically including: Based on the target area's commercial value score, projected return on investment, and competitive environment data, site selection assessment recommendations for new asset locations are generated. The site selection assessment recommendations are generated through a site selection scoring model. The site selection scoring model takes business value score and investment return prediction as core input parameters, and combines geographic information system spatial analysis technology to quantify and calibrate competitive environment data, and outputs quantified site selection scores and corresponding equipment configuration recommendations.

9. A new energy asset business insight assessment system for operation and maintenance investment decision-making, used to implement the new energy asset business insight assessment method for operation and maintenance investment decision-making as described in any one of claims 1-8, characterized in that, include: Data integration module: used to acquire and process multi-source heterogeneous data of new energy equipment assets. The data integration module includes a data fusion engine, used to clean and associate the multi-source heterogeneous data, and to standardize and store it according to a three-dimensional structure of asset identification, time dimension and data type. Business Insight Assessment Module: Connected to the data integration module, it is used to execute a set of preset assessment models to generate quantitative assessment results; The business insight assessment module includes: a business value assessment unit, an equipment health rating unit, an investment return prediction unit, and a risk warning unit; Decision support module: Connected to the business insight assessment module, it is used to generate decision suggestions for asset operation and maintenance and decision suggestions for asset investment based on the quantitative assessment results and through a preset decision algorithm; Display module: Connected to the business insight assessment module and decision support module, it is used to integrate and visualize the quantitative assessment results and decision recommendations; Model Iteration and Optimization Module: Connected to the Business Insight Assessment Module, Decision Support Module, and Data Integration Module respectively, this module is used to periodically collect the deviations between assessment results, decision execution feedback, and actual operational data. It updates the parameters of the set of preset assessment models and decision algorithms through parameter optimization algorithms, forming a closed-loop learning mechanism.

Images