Business process method for saving electric power and electric quantity

Abstract

The invention discloses a business process method for saving electric power and electric quantity. Comprising the following steps: S1, obtaining business parameter data of a to-be-measured project; s2, calling a calculation algorithm template and a validity threshold rule of the standardized calculation model library; s3, generating a comprehensive evaluation signal by using adaptive measurement and calculation and a quality sentinel engine; s4, analyzing a data state identifier in the comprehensive evaluation signal, and generating a control instruction packet; s5, in response to the index updating instruction, creating a to-be-processed abnormal work order and locking the business process of the current project; according to the method, physical high fidelity of a data source is determined through dynamic correction and self-adaptive verification; the multi-dimensional risk entropy and the service weight are fused, so that the collaborative gain of differential risk control is realized; a full-link quality closed loop is constructed in combination with aging attenuation and confidence coefficient weighted aggregation; a data trust footstone is reconstructed, and digital qualitative change of electric power and electric quantity business from microscopic accurate measurement and calculation to macroscopic credible decision making is realized.

Description

Technical Field

[0001] This invention belongs to the field of power energy data processing and management technology, and in particular relates to a method for saving power consumption in business processes. Background Technology

[0002] As a core component of energy conservation and emission reduction, electricity saving services not only involve energy efficiency monitoring of massive amounts of equipment, but also encompass the entire lifecycle management from project application and energy saving calculation to indicator control. Currently, digital technology has gradually penetrated all levels of the energy system, aiming to optimize energy allocation through the fusion and analysis of multi-source data. However, facing increasingly complex energy-saving retrofit scenarios (such as high-voltage frequency conversion and waste heat power generation) and ever-increasing requirements for data granularity and accuracy, how to build a business processing system that can adapt to the characteristics of differentiated projects while ensuring the authenticity and reliability of data throughout the entire process has become a key focus of the industry.

[0003] Although existing power saving calculation services have partially incorporated electronic tools, they still face several significant technical challenges and limitations in practical applications:

[0004] The standardization and reliability of data sources are difficult to quantify: Existing technologies mostly rely on general spreadsheet tools (such as Excel) for manual offline statistics and data entry. This method lacks a mechanism for distinguishing and verifying the quality of heterogeneous data sources (such as manual reporting and automatic instrument data collection). Due to the lack of dynamic constraints on input fields and real-time evaluation of source reliability, the completeness and accuracy of the entered data vary greatly, making it easy to generate data pollution problems of "garbage in, garbage out".

[0005] The static rigidity and insufficient environmental adaptability of the calculation model: Traditional calculation methods usually use fixed physical formulas, often ignoring the actual impact of fluctuations in the equipment's operating environment (such as temperature and load rate) on energy efficiency, and lacking a dynamic threshold verification mechanism based on historical statistical patterns. This static calculation model is difficult to accurately reflect the real energy-saving effect under non-standard operating conditions, resulting in deviations between the calculation results and the actual situation.

[0006] Lack of open-loop control and synergy in business processes: Existing audit processes mostly adopt a serial mode of "data entry first, then manual auditing," lacking an automatic risk control mechanism that is triggered in real time during the calculation process. The discovery of data anomalies is often delayed, and it is impossible to achieve automated closed-loop control from "anomaly identification" to "business lock-in," making it difficult to form an effective driving force for rectification and resulting in limited overall management efficiency.

[0007] To address the aforementioned challenges, this invention establishes a robust quality access defense at the data collection source by introducing dynamic rendering of structured forms and a comprehensive interface reliability scoring mechanism, ensuring the standardized acquisition of parameters for different project categories. It employs a layered and decoupled computational architecture, superimposing environmentally sensitive dynamic correction factors and statistically based elastic threshold rules on top of fundamental physical logic, thus achieving a shift from static calculation to adaptive dynamic measurement. A multi-parameter fusion quality sentinel engine is embedded in the measurement process, calculating a comprehensive risk entropy value to assess data credibility in real time, and using state machine logic to transform the assessment results into specific control commands, performing streaming aggregation updates based on confidence weights. For abnormal data, a transactional rectification lock and time-lapse mechanism is triggered. Summary of the Invention

[0008] To address the problems existing in the prior art, the purpose of this invention is to provide a method for a business process that saves electricity. This method solves problems such as weak data quality control, poor environmental adaptability, and lagging process control in the prior art. It achieves a synergistic benefit effect of calculation-verification and anomaly-blocking. Through environmental correction and statistical deviation analysis, it significantly improves the physical authenticity and logical rigor of electricity saving calculation. By deeply coupling data quality scoring with business flow control, it achieves a closed-loop management of the entire chain, from micro-level project rectification to macro-level weighted aggregation of indicators, which greatly improves the automation level and decision-making credibility of electricity saving business.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a method for saving electricity in a business process, comprising the following steps:

[0010] S1. Obtain the business parameter data of the project to be measured. The business parameter data includes a type identifier used to uniquely represent the project category and a set of operating characteristic parameters used to quantify the project's operating status.

[0011] S2. Call the pre-set standardized calculation model library. The model library stores multiple calculation algorithm templates corresponding to different types of identifiers, as well as validity threshold rules corresponding to the running feature parameter set.

[0012] S3. Utilizing adaptive calculation and quality sentinel engine, business parameter data is routed and matched to the target calculation algorithm template based on type identifier. While performing calculation, the validity threshold rule is used to perform real-time verification of the running feature parameter set to generate a comprehensive evaluation signal containing power saving value and data status identifier.

[0013] S4. Analyze the data status identifier in the comprehensive evaluation signal and generate a control instruction package. If the identifier is compliant, the control instruction package contains an indicator update instruction; if the identifier is abnormal, the control instruction package contains a rectification and blocking instruction.

[0014] S5. In response to the indicator update command, automatically update the hierarchical annual energy saving indicator data; or in response to the rectification and blocking command, create pending abnormal work orders and lock the business process of the current project.

[0015] Furthermore, the set of operating characteristic parameters in S1 is collected through a structured form interface. The structured form interface dynamically renders input fields based on different project categories. The input fields include energy consumption baseline values, equipment operating power, and operating time parameters.

[0016] Furthermore, the quality assessment steps for business parameter data are as follows:

[0017] The system captures user business parameter data through the front-end interactive interface, obtains project category feature identification codes, and retrieves the corresponding dynamic form metadata structure from the configuration set of the non-relational database. The dynamic form metadata structure contains a list of all fields required for the project category, as well as the pre-defined field importance weight for each field.

[0018] The rendering engine parses the dynamic form metadata structure and draws the corresponding input controls on the user interface. It also binds a data change listener to each input control to capture the data input method in real time.

[0019] When data is populated into any field, the listener identifies the data source type and sets the field integrity status value to 1 based on the data source confidence coefficient.

[0020] When each data change event is triggered, the comprehensive reliability score of the calculation interface is updated in real time by using the field importance weight, data source confidence coefficient and field integrity status value of all current fields.

[0021] If the overall reliability score of the interface is less than the admission reliability threshold, the current data collection quality is deemed insufficient. The submit button is disabled on the front end, and fields with low confidence coefficients of data sources are highlighted to prompt users to supplement data using a more reliable method.

[0022] If the overall reliability score of the interface is greater than or equal to the admission reliability threshold: the data quality is deemed compliant, the submit button is unlocked, and the data is allowed to proceed to the next stage.

[0023] Furthermore, the standardized computational model library in S2 is constructed using a layered and decoupled architecture, including a basic algorithm layer and a parameter configuration layer. The basic algorithm layer stores general physical computation logic, while the parameter configuration layer stores specific coefficients and validity threshold rules associated with type identifiers.

[0024] Furthermore, the adaptive measurement deviation index is calculated as follows:

[0025] User-entered set of operational characteristic parameters; real-time collected environmental status parameters; preset standard operating condition basic energy efficiency coefficient, which includes the historical data distribution mean and the historical data distribution standard deviation; statistical characteristic quantities provided by the historical database;

[0026] The original energy saving is calculated by calling the physical formula in the basic algorithm layer; the environmentally sensitive dynamic correction factor is calculated by using the empirical formula for operating condition correction in thermodynamics and electrical engineering, and the original energy saving is multiplied by the environmentally sensitive dynamic correction factor to obtain the corrected energy saving.

[0027] The elastic threshold boundary scaling coefficient in the configuration layer is read in parallel. Combining the mean and standard deviation of historical data distribution, the upper and lower bound thresholds of dynamic effectiveness are calculated in real time using the normal distribution interval estimation theory in statistics.

[0028] Using Z-Score standardization and nonlinear mapping, an adaptive measurement deviation index is calculated based on the corrected post-segment electricity consumption, the mean of historical data distribution, and the standard deviation of historical data distribution.

[0029] If the adaptive calculation deviation index > deviation blocking threshold or the corrected power consumption > dynamic validity upper limit threshold: the calculation result is determined to be seriously deviating from historical patterns or physical boundaries; strong verification blocking is triggered, an abnormal work order is generated, and the adaptive calculation deviation index value is written into the work order as an abnormality severity indicator.

[0030] If the adaptive measurement deviation index is less than or equal to the deviation blocking threshold, the calculation result is determined to be within a reasonable fluctuation range. The corrected energy saving is adopted as the final energy saving and stored in the database. At the same time, the historical statistical sample database is updated for the iteration of the historical data distribution mean in the next cycle.

[0031] The final output is: corrected power consumption and adaptive measurement deviation index.

[0032] Furthermore, the calculation logic executed by the adaptive measurement and quality sentinel engine in S3 is as follows: Through the algorithm routing mechanism, business parameter data is injected into the matching target measurement algorithm template for trial calculation; the deviation of the running feature parameter set from the validity threshold rule is calculated synchronously. If the deviation of all parameters is within the preset compliance range, a comprehensive evaluation signal containing the effective power saving value and compliance status identifier is output; if the deviation of any parameter exceeds the compliance range, or if there is a logical conflict in the trial calculation result, a comprehensive evaluation signal containing the error code and abnormal status identifier is output.

[0033] Furthermore, the calculation steps for the comprehensive evaluation signal are as follows:

[0034] Input business parameter data, target calculation algorithm template, and validity threshold rules;

[0035] Based on the project type, the corresponding target calculation algorithm template is called, the business parameters are substituted into the calculation to obtain the temporary calculation result;

[0036] Three subtasks are executed in parallel: the physical boundary deviation index of the input parameters relative to the physical boundary is calculated based on the linear normalization method; the historical statistical deviation index of the temporary measurement results relative to the historical average is calculated; and the logical relationship between the business parameter data is verified and the logical association consistency coefficient is calculated.

[0037] The comprehensive risk entropy value of the current task is calculated based on the physical boundary deviation index, historical statistical deviation index, and logical correlation consistency coefficient.

[0038] Compare the comprehensive risk entropy value with the dynamic blocking warning threshold;

[0039] If the comprehensive risk entropy value is less than or equal to the dynamic blocking warning threshold: the risk is determined to be controllable, and a signal containing temporary calculation results and compliance indicators is output.

[0040] If the comprehensive risk entropy value is greater than the dynamic blocking warning threshold: a high-risk anomaly is determined, the circuit breaker is triggered, and an error code containing the specific violation dimension and an anomaly identifier are generated;

[0041] The final output is a comprehensive evaluation signal.

[0042] Furthermore, the steps for parsing the comprehensive evaluation signal in S4 are as follows: extract the data status identifier field from the comprehensive evaluation signal, use state machine logic to determine the current business flow direction; when the data status identifier is abnormal, extract the error code and map it to specific rectification prompt information, and encapsulate it into the rectification blocking instruction.

[0043] Furthermore, the steps for generating the control instruction packet are as follows:

[0044] Input the comprehensive evaluation signal; analyze the comprehensive evaluation signal and extract the data status flag bits;

[0045] If the data status flag is 0, the state machine transitions to the index update state and directly generates an index update instruction;

[0046] If the data status flag bit = 1, the state machine transitions to the abnormal resolution state and proceeds to the next step;

[0047] In the abnormal parsing state, the error code is extracted from the comprehensive evaluation signal, the fault dictionary table is queried, and the corresponding natural language rectification prompt and error code severity coefficient are retrieved.

[0048] The project business weight factor of the current project is calculated in parallel; the rectification urgency index is calculated by using a linear weighted model based on multi-attribute utility theory, combined with the error code severity coefficient and the project business weight factor.

[0049] Construct a rectification blocking instruction object; encapsulate the following information into the object: natural language rectification prompt; calculated rectification urgency index; locked business document ID;

[0050] Before generating control instructions, determine the relationship between the urgency index of rectification and the threshold of graded blocking response;

[0051] If the urgency index of rectification is greater than or equal to the graded blocking response threshold: it is determined to be an emergency blocking measure; the SLA level is marked as P1 - the highest priority in the control instructions, requiring a response within 4 hours;

[0052] If the urgency index for rectification is less than the tiered blocking response threshold: it is judged as a general blocking; the SLA level is marked as P3-standard priority in the control instructions, and a response is allowed within 24 hours;

[0053] The final output is a fully encapsulated control instruction package.

[0054] Furthermore, the steps in S5 in response to the rectification blocking command are as follows: instantiate a new abnormal work order object, write the rectification prompt information into the work order object, and lock the status of the project to be measured as pending rectification until a closed-loop feedback signal for the abnormal work order is received; the steps in response to the indicator update command are as follows: use the aggregation algorithm to accumulate the energy saving value into the corresponding municipal and provincial annual energy saving indicator completion values, and refresh the visual monitoring dashboard in real time.

[0055] The technical effects and advantages of this invention are as follows:

[0056] This invention breaks away from the arbitrariness of traditional general text box input by using dynamic rendering technology and an interface-based reliability scoring mechanism, forcibly establishing data collection standards and ensuring the integrity and compliance of parameters specific to different types of projects. This invention abandons rigid static formulas, introducing environmentally sensitive dynamic correction factors and adaptive calculation deviation indices based on historical statistics. It can sense fluctuations in physical operating conditions such as operating temperature and load rate, and automatically adjust the verification boundaries according to historical data distribution, thereby eliminating calculation deviations caused by environmental differences. This ensures that every unit of electricity saving data is not only standardized in format but also possesses extremely high realism and fidelity in physical meaning.

[0057] This invention constructs an adaptive measurement and quality sentinel engine, which calculates a comprehensive risk entropy value by integrating three dimensions: physical boundaries, historical statistics, and logical correlations. This achieves a three-dimensional situational awareness of data risks, effectively avoiding false alarms and missed alarms caused by single-dimensional verification. It introduces a business context-aware decision-making mechanism, deeply integrating the severity of error codes with project business weights using a rectification urgency index. This results in a significant synergistic gain effect: instead of mechanically blocking all anomalies, it can intelligently distinguish between minor deviations in core business and major errors in peripheral business, dynamically generating tiered SLA blocking instructions. This ensures the efficiency of core business processes and guarantees that limited operation and rectification resources can be accurately allocated to the highest-risk or highest-value links, improving the overall operational efficiency of the system.

[0058] This invention elevates microscopic calculation results to macroscopic management tools through a two-way response mechanism of state locking and streaming aggregation. On the rectification side, by introducing a rectification timeliness decay coefficient and dynamic project credit scoring, a time-cost-driven mechanism is constructed, compelling applicants to efficiently complete data governance within the locked period, achieving rigid control of the problem loop. On the indicator side, the traditional simple accumulation mode is abandoned, and a confidence-weighted streaming aggregation algorithm is adopted to calculate the effective indicator contribution value based on the quality weight of the data. This ensures that what is finally presented on the dashboard is not an inflated stack of numbers, but effective assets that have been filtered and weighted by quality. This greatly enhances the robustness and anti-interference ability of macroscopic decision-making data, and provides truly objective, reliable, and quantifiable digital support for enterprises' energy conservation and emission reduction strategies.

[0059] Other features and advantages of the invention will become clear from the following detailed description of exemplary embodiments of the invention with reference to the accompanying drawings. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the steps provided by the present invention;

[0061] Figure 2 This is a schematic diagram of the quality determination steps for business parameter data provided by the present invention;

[0062] Figure 3 This is a schematic diagram of the calculation process for the adaptive measurement deviation index provided by the present invention;

[0063] Figure 4 This is a schematic diagram of the calculation steps for the comprehensive evaluation signal provided by the present invention;

[0064] Figure 5 This is a schematic diagram of the steps for generating the control instruction package provided by the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0066] like Figures 1 to 5 As shown in the figure, an embodiment of the present invention provides a method for saving electricity in a business process, which includes the following steps:

[0067] S1. Obtain the business parameter data of the project to be measured. The business parameter data includes a type identifier used to uniquely represent the project category and a set of operating characteristic parameters used to quantify the project's operating status.

[0068] The set of operating characteristic parameters in S1 is collected through a structured form interface. The structured form interface dynamically renders input fields based on different project categories. The input fields include energy consumption baseline values, equipment operating power, and operating time parameters.

[0069] Based on the project category selected by the user (motor modification in this example), the corresponding JSON configuration is obtained from the server, and a project category-specific input box is dynamically rendered (motor rated power, power consumption before modification in this example). The data collection is not a general text box, but a strongly typed field with units and data types. This solves the standardization problem of multi-source heterogeneous data collection. Through dynamic rendering, it ensures that the unique parameters of different project categories are not missed, providing a standardized data foundation for subsequent accurate calculations.

[0070] In this embodiment, the key parameters are defined and calculated as follows:

[0071] The parameter symbol for the project category feature identifier code is: ; Used to uniquely identify the specific category of energy-saving projects. In this embodiment, it includes high-voltage frequency conversion retrofitting, waste heat power generation, etc., and is the index key for retrieving configuration information; its physical meaning is the pre-set enumeration type integer value in the database; it is obtained by capturing the user's selection instruction in the project application drop-down menu through the front-end interactive interface, or by parsing the fields in the XML message transmitted by the upstream business through the API interface;

[0072] The parameter symbol for field importance weight is: ; represents the core importance of a specific input field (device operating power in this embodiment) in calculating the power saving of the current type of project; is a dimensionless normalization coefficient with a value range of (0,1];

[0073] The symbol for the confidence coefficient of the data source is: ; represents the reliability level of the source when the current field data is populated; is a probability value derived based on the statistical false alarm rate, with a value range of (0,1];

[0074] The parameter symbol for the field integrity status value is ; indicates whether the required fields for dynamic rendering have been effectively filled; is a binary logical variable, 1 indicates that it has been filled and the format validation has passed, and 0 indicates that it has not been filled or the format is incorrect;

[0075] The parameter symbol for the overall interface reliability score is: ; represents the overall credibility of the data collection process; is a comprehensive probability index calculated based on the weighted average algorithm, with the value range limited to the interval [0,1]; the calculation logic based on the weighted arithmetic average algorithm is as follows: for each dynamically rendered input field, calculate the weighted contribution value of the field, which is the product of the field integrity status value, the data source confidence coefficient, and the field importance weight; sum up the weighted contribution values ​​of all fields; divide the above sum by the sum of the field importance weights of all fields. In this embodiment, due to weight normalization, the denominator is usually 1. If there are non-mandatory fields that are dynamically adjusted, they need to be normalized again to obtain the final score;

[0076] The specific calculation process for determining the quality of business parameter data is as follows:

[0077] The input is the item category feature identifier code triggered by the user or system. ;

[0078] Based on the received project category feature identifier code The system retrieves the corresponding dynamic form metadata structure from the configuration set of a non-relational database (in this embodiment, the non-relational database used is MongoDB). The dynamic form metadata structure contains a list of all fields required for the project category, as well as a pre-defined field importance weight for each field. ;

[0079] The rendering engine parses the dynamic form metadata structure and draws the corresponding input controls on the user interface. In this embodiment, the input controls include text boxes and drop-down boxes. A data change listener is bound to each input control to capture the data input method in real time.

[0080] When data is populated into any field, the listener identifies the data source type and determines the data source confidence level based on the data source confidence coefficient. At the same time, the field integrity status value of the field will be set. Set to 1;

[0081] Each time a data change event is triggered, the importance weights of all current fields are utilized. Confidence coefficient of data source and field integrity status value Real-time updates of the overall reliability score of the computing interface ;

[0082] Preset access reliability threshold In this embodiment, the admission reliability threshold Set to 0.70;

[0083] Branch execution:

[0084] If the overall reliability score of the interface <Admission reliability threshold The current data collection quality is deemed insufficient. In this embodiment, this may be due to excessive reliance on manual input or missing key fields. The submit button is disabled on the front end, and the confidence coefficient of the data source is highlighted. For fields with lower values, the user is prompted to supplement the data using a more credible method (in this example, uploading supporting materials for OCR).

[0085] If the overall reliability score of the interface ≥ Admission Reliability Threshold Once the data quality is deemed compliant, the submit button is unlocked, allowing the data to proceed to the next stage.

[0086] The final output is an encapsulated structured business data packet. This structured business data packet not only contains user-inputted entity data (in this embodiment, entity data includes energy consumption baseline values, power, etc.), but also a metadata header, which records the calculated comprehensive interface reliability score. Interface overall reliability score Accurate to two decimal places, within the range [0,1].

[0087] S2. Call the pre-set standardized calculation model library. The model library stores multiple calculation algorithm templates corresponding to different types of identifiers, as well as validity threshold rules corresponding to the running feature parameter set.

[0088] The standardized computational model library in S2 is built using a layered and decoupled architecture, including a basic algorithm layer and a parameter configuration layer. The basic algorithm layer stores general physical computation logic, while the parameter configuration layer stores specific coefficients and validity threshold rules associated with type identifiers.

[0089] The calculation formula for the basic algorithm layer is as follows: And stored in the database, where, This is the basic logic, which in this embodiment includes multiplication and accumulation. These are coefficients for different project types. In this embodiment, the energy efficiency ratios of different devices are used. These are the input parameters; the validity threshold rules are defined. and Within the legal range, in this embodiment, the daily running time is ≤24 hours; the decoupling of calculation logic and code is achieved; when a new project type is added or the calculation standard is adjusted, there is no need to rewrite the code, only to update the parameter configuration layer, which greatly improves scalability and maintenance efficiency;

[0090] In this embodiment, the following key parameters are defined and calculated as follows:

[0091] The parameter symbol for the basic energy efficiency coefficient under standard operating conditions is: ; represents the theoretical energy efficiency ratio or power saving rate of a specific type of energy-saving equipment under a standard test environment, which in this embodiment is a full load at 25°C; is a dimensionless ratio, stored in the configuration table of the database; directly reads the nominal value from the equipment's factory nameplate data or the energy efficiency test report issued by a third-party authoritative testing agency, and enters it into the parameter configuration layer;

[0092] The parameter symbol of the environmentally sensitive dynamic correction factor is: ; used to compensate for the impact of differences between actual operating environment and standard operating conditions on energy efficiency; is a dimensionless correction coefficient, typically ranging from [0.8, 1.2]; based on empirical formulas for operating condition correction in thermodynamics and electrical engineering, the calculation logic is as follows: Let the actual operating environment temperature be... The standard test temperature is In this embodiment, the temperature is set to 25°C. The calculation logic of the correction factor is as follows: calculate the difference between the actual temperature and the standard temperature; then divide the difference by the standard temperature; multiply the resulting ratio by the preset temperature sensitivity coefficient, and perform an algebraic sum operation with the result and the constant 1; if it is necessary to normalize to the [0,1] interval for subsequent fusion, the Sigmoid function can be used for mapping.

[0093] The parameter sign of the historical data distribution mean is: It represents the arithmetic average level of key operating parameters (in this embodiment, the key operating parameter is the daily electricity saving) of a type of project over a past statistical period; it is a statistical quantity with specific physical units.

[0094] The sign of the parameter for the standard deviation of historical data distribution is: It characterizes the dispersion of historical data for a given type of item; it is a statistical quantity with specific physical units.

[0095] The parameter sign of the elastic threshold boundary scaling factor is: Used to adjust the width of the anomaly detection boundary to adapt to different confidence level requirements; is a dimensionless positive real number; set by the system administrator according to the business risk control strategy; set to 3.0 in the lenient mode and 2.0 in the strict mode;

[0096] The parameter sign of the dynamic validity upper bound threshold is: The upper limit of the dynamic criterion used to verify whether the entered data is abnormal; a numerical value with specific physical units; based on the normal distribution interval estimation theory in statistics; the calculation logic is as follows: based on historical statistical data; extract data samples of the most recent N periods of the project type from the historical database; calculate the historical data distribution mean and historical data distribution standard deviation of the samples; and scale the mean and standard deviation with the elastic threshold boundary coefficient. Adding the products together yields the dynamic upper bound;

[0097] The parameter sign of the adaptive measurement deviation index is: ; represents the degree of deviation of the current calculation result from the expected model; is a normalized dimensionless index, with a value range limited to the [0,1] interval; the calculation logic is as follows: based on Z-Score standardization and nonlinear mapping; obtain the corrected energy saving obtained from the current calculation; calculate the absolute value of the difference between the energy saving and the mean of the historical data distribution; divide the absolute value by the standard deviation of the historical data distribution to obtain the standard score; use the hyperbolic tangent function to map the standard score to the [0,1] interval;

[0098] The adaptive measurement deviation index is calculated as follows:

[0099] The user-entered set of operational characteristic parameters, in this embodiment, includes running time and load rate; real-time collected environmental status parameters, in this embodiment, include the actual operating environment temperature. and standard test temperature Pre-set standard operating condition basic energy efficiency coefficient Statistical features provided by historical databases, including the mean of historical data distributions. and the standard deviation of historical data distribution ;

[0100] The original energy saving is calculated by calling the physical formulas in the basic algorithm layer; the environmentally sensitive dynamic correction factor is calculated using empirical formulas for operating condition correction from thermodynamics and electrical engineering. The original energy savings are then multiplied by an environmentally sensitive dynamic correction factor. After obtaining the corrected power saving ;

[0101] Parallel reading of elastic threshold boundary scaling coefficients in the configuration layer Combined with the historical data distribution mean and the standard deviation of historical data distribution The upper bound threshold of dynamic effectiveness is calculated in real time using the normal distribution interval estimation theory in statistics. and lower bound threshold;

[0102] Using Z-Score normalization and nonlinear mapping, based on the corrected energy saving... Historical data distribution mean and the standard deviation of historical data distribution Calculate the adaptive measurement deviation index ;

[0103] Set deviation blocking threshold In this embodiment, it is set to 0.95, which corresponds to approximately 2 times the standard deviation;

[0104] If adaptive measurement deviation index Deviation blocking threshold Or adjust the power saving after the adjustment >Dynamic validity upper bound threshold The calculation result is determined to deviate significantly from historical patterns or physical boundaries; a strong verification block is triggered, an abnormal work order is generated, and the adaptive deviation index is calculated. The value is written into the work order as an indicator of the severity of the anomaly;

[0105] If adaptive measurement deviation index ≤ Deviation blocking threshold The calculation results are deemed to be within a reasonable fluctuation range; the corrected power saving is accepted. The final energy savings are recorded and stored in the database, while the historical statistical sample database is updated for use as the historical data distribution mean in the next cycle. Iteration;

[0106] The final output is: corrected battery level. Used for report display and indicator accumulation; adaptive calculation of deviation index. Used for data quality labeling and risk control analysis.

[0107] S3. Utilizing adaptive calculation and quality sentinel engine, business parameter data is routed and matched to the target calculation algorithm template based on type identifier. While performing calculation, the validity threshold rule is used to perform real-time verification of the running feature parameter set to generate a comprehensive evaluation signal containing power saving value and data status identifier.

[0108] The calculation logic executed by the adaptive measurement and quality sentinel engine in S3 is as follows: Through the algorithm routing mechanism, business parameter data is injected into the matching target measurement algorithm template for trial calculation; the deviation of the running feature parameter set from the validity threshold rule is calculated synchronously. If the deviation of all parameters is within the preset compliance range, a comprehensive evaluation signal containing the effective power saving value and compliance status identifier is output; if the deviation of any parameter exceeds the compliance range, or if there is a logical conflict in the trial calculation result, a comprehensive evaluation signal containing the error code and abnormal status identifier is output.

[0109] In this embodiment, the following key parameters are defined and calculated as follows:

[0110] The parameter sign of the physical boundary deviation index is: ; represents the degree of deviation of the input parameters from the physical limits of the device; is a dimensionless normalized value, with a value range of [0,1];

[0111] Based on the linear normalization method, the calculation logic is as follows: determine whether the input parameter value is within the physical lower limit and physical upper limit interval; if it is within the interval, calculate the distance between the input parameter value and the midpoint of the interval, and divide it by the half width of the interval to obtain the original deviation; if it exceeds the interval, directly take the value as 1; to facilitate fusion, the deviation within the interval is mapped to [0,0.5], and the deviation exceeding the interval is mapped to (0.5,1].

[0112] The parameter sign of the historical statistical deviation index is ; represents the degree of deviation of the calculation result from the historical average of similar projects; is a dimensionless normalized value, with a value range of [0,1];

[0113] The parameter sign of the logical association consistency coefficient is: ; represents the logical consistency between multiple input parameters; is a dimensionless probability value, ranging from [0,1];

[0114] The parameter symbol for the comprehensive risk entropy value is: ; represents the overall data unreliability of the current measurement task; is a dimensionless value calculated based on information entropy theory, with the value range limited to the interval [0,1];

[0115] The parameter symbol for the dynamic blocking warning threshold is: It is the critical point that triggers the system's circuit breaker mechanism; it is a dimensionless value that is dynamically adjusted based on the global false alarm rate.

[0116] The calculation steps for the comprehensive evaluation signal are as follows:

[0117] Input business parameter data, target calculation algorithm template, and validity threshold rules;

[0118] Based on the project type, the corresponding target calculation algorithm template is called, the business parameters are substituted into the calculation to obtain the temporary calculation result;

[0119] Three subtasks are executed in parallel: calculating the physical boundary deviation index of the input parameters relative to the physical boundary based on the linear normalization method. ; Calculate the historical statistical deviation index of the provisional calculation results relative to the historical average. Verify the logical relationships between business parameter data and calculate the logical association consistency coefficient. ;

[0120] Based on physical boundary deviation index Historical statistical deviation index Consistency coefficient of logical association Calculate the overall risk entropy value of the current task. ;

[0121] Comparison of comprehensive risk entropy values With dynamic blocking warning threshold In this embodiment, it is set to Example 0.6;

[0122] If the comprehensive risk entropy value ≤Dynamic blocking warning threshold If the risk is deemed manageable, output a signal including provisional calculation results and compliance indicators.

[0123] If the comprehensive risk entropy value Dynamic blocking warning threshold If a high-risk anomaly is detected, a circuit breaker is triggered, and an error code containing the specific violation dimension (in this embodiment, it is set to excessive historical deviation) and an anomaly identifier is generated.

[0124] The final output is a comprehensive evaluation signal.

[0125] S4. Analyze the data status identifier in the comprehensive evaluation signal and generate a control instruction package. If the identifier is compliant, the control instruction package contains an indicator update instruction; if the identifier is abnormal, the control instruction package contains a rectification and blocking instruction.

[0126] The steps for parsing the comprehensive evaluation signal in S4 are as follows: extract the data status identifier field from the comprehensive evaluation signal, use state machine logic to determine the current business flow direction; when the data status identifier is abnormal, extract the error code and map it to specific rectification prompt information, and encapsulate it into the rectification blocking instruction;

[0127] In this embodiment, the following key parameters are defined and calculated:

[0128] The parameter symbol for the data status flag is: ; represents the original state conclusion of the upstream output; is a binary logic enumeration quantity, in this embodiment, 0 represents compliance and 1 represents abnormality;

[0129] The parameter symbol for the error code severity coefficient is: It represents the degree of damage to data quality caused by a specific error code; it is a dimensionless normalized value with a range of (0,1]; it is based on failure mode and effects analysis theory; it uses a pre-set fault dictionary table; each error code in the table corresponds to a severity value calibrated by experts;

[0130] Error code E001, logical conflict. In this embodiment, power saving > power consumption: set the severity coefficient of the error code. =1.0, absolutely unreliable;

[0131] Error code E002, statistical deviation (in this embodiment, exceeding twice the historical average): set the error code severity coefficient. =0.7, highly suspicious;

[0132] Error code E003, incomplete information (in this embodiment, non-mandatory fields are empty): Set the severity coefficient of the error code. =0.3, minor defect;

[0133] The parameter symbol for the project business weight factor is: ; represents the importance of the current project to be measured in the overall energy saving target. In this embodiment, the project with the larger the annual electricity saving is, the higher the weight is; is a dimensionless normalized value with a value range of (0,1]; normalization processing based on Pareto law; the calculation logic is as follows: the estimated annual electricity saving of the current project is used to set a high-value project baseline; the ratio of the estimated annual electricity saving to the high-value project baseline is used as input, and nonlinear mapping is performed through the hyperbolic tangent function to make its output approximate the interval [0,1];

[0134] The parameter symbol for the urgency index of rectification is: Used to determine the execution level of blocking instructions and the priority of work orders; a dimensionless value calculated based on a weighted algorithm, with a value range limited to the [0,1] interval; based on a linear weighted model of multi-attribute utility theory; the calculation logic is as follows, integrating the severity of the error with the importance of the project; the error code severity coefficient Multiply by the business impact adjustment coefficient (set to 0.6 in this embodiment); then apply the project business weight factor. Multiply by the project impact adjustment factor (set to 0.4 in this example); add the two products together; to ensure normalization, the sum of the adjustment factors must equal 1;

[0135] The parameter symbol for the graded blocking response threshold is: ; used to distinguish between general rectification and emergency rectification; is a dimensionless value;

[0136] The steps for generating the control instruction packet are as follows:

[0137] Input the comprehensive evaluation signal; analyze the comprehensive evaluation signal and extract the data status flag bits. ;

[0138] If the data status flag is =0, the state machine transitions to the indicator update state and directly generates the indicator update instruction;

[0139] If the data status flag is =1, the state machine transitions to the abnormal resolution state and proceeds to the next step;

[0140] In the abnormal parsing state, error codes are extracted from the comprehensive evaluation signal, the fault dictionary table is queried, and the corresponding natural language rectification prompts and error code severity coefficients are retrieved. ;

[0141] Parallel computation of project business weight factors for the current project Using a linear weighted model based on multi-attribute utility theory, combined with error code severity coefficients... and project business weighting factors Calculate the urgency index of rectification ;

[0142] Construct a rectification blocking instruction object; encapsulate the following information into the object: natural language rectification prompt; and calculate the rectification urgency index. The locked business document ID;

[0143] Before generating control instructions, assess the urgency index of rectification. Regarding the relationship with the graded blocking response threshold, in this embodiment, the graded blocking response threshold is set to 0.8;

[0144] If the urgency index of rectification ≥ Tiered blocking response threshold: Determined as emergency blocking; SLA level marked as P1 - highest priority in control instructions, requiring a response within 4 hours;

[0145] If the urgency index of rectification <Graded blocking response threshold: Determined as general blocking; SLA level marked as P3-standard priority in control instructions, allowing response within 24 hours;

[0146] The final output is a fully encapsulated control instruction package.

[0147] S5. In response to the indicator update command, automatically update the hierarchical annual energy saving indicator data; or in response to the rectification and blocking command, create pending abnormal work orders and lock the business process of the current project.

[0148] The steps in S5 to respond to the rectification blocking command are as follows: instantiate a new abnormal work order object, write the rectification prompt information into the work order object, and lock the status of the project to be measured to be under rectification until a closed-loop feedback signal for the abnormal work order is received; the steps to respond to the indicator update command are as follows: use the aggregation algorithm to accumulate the energy saving value into the corresponding municipal and provincial annual energy saving indicator completion values, and refresh the visual monitoring dashboard in real time.

[0149] In this embodiment, the following key parameters are defined and calculated as follows:

[0150] The parameter symbol for the rectification time reduction coefficient is: This represents the rate at which a project's credit rating decreases over time when it is in a state of pending rectification and locked-in status; it is the reciprocal of the time dimension.

[0151] The parameter symbols for the project's dynamic credit score are: ; used to measure the cooperation and efficiency of the project applicant during the rectification process; is a dimensionless normalized value, with a range of [0,1];

[0152] The parameter symbol for the original value of power saving is: ; represents the calculated corrected energy saving; is a real number with a specific physical unit;

[0153] The parameter symbol for the overall confidence weight of the data is: ; represents the reliability of the current power saving data; is a dimensionless normalized value, with a range of (0,1];

[0154] The parameter symbol for the effective indicator contribution value is: This refers to a virtual unit of electrical quantity after quality weighting.

[0155] The sign of the aggregate volatility variance parameter is: Used to display the range of uncertainty for current statistical indicators on the dashboard;

[0156] The steps for generating a business process are as follows:

[0157] Based on control command packets; responding to rectification and blocking commands:

[0158] Parse the control instruction packet and create an exception work order instance in memory; write the error code, rectification prompt, and SLA priority to the instance properties;

[0159] Update the status field in the project master table from "under calculation" to "pending rectification"; persist the work order instance to the work order table; initialize the rectification time-out counter and record the lock start timestamp; if any operation fails, perform a full rollback to ensure data consistency; start a background daemon process to periodically (in this embodiment, hourly) calculate the project's dynamic credit score. If the score is lower than the preset threshold, in this embodiment, the preset threshold is set to 0.2, and a serious warning notification is automatically sent to the project manager; only when the work order completion signal is received and passes the secondary verification (in this embodiment, steps S1-S3 are re-executed) is the unlocking transaction triggered to restore the project status.

[0160] Response indicator update instruction: Parse the control instruction packet and extract the raw power saving value. Combined confidence weight of data ; Calculate the effective indicator contribution value of this update And individual fluctuation variance; using atomic increment commands in in-memory databases such as Redis, the contribution value of effective indicators is... The data is accumulated into the cache keys of the provincial and municipal annual energy-saving targets, and the fluctuation variance is also accumulated; incremental update messages are pushed to the front-end visualization screen through the WebSocket channel.

[0161] Final output: The project status in the database changes to pending rectification, and the work order system generates a new record; the numbers on the monitoring dashboard update in real time and display the latest confidence interval.

[0162] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. 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 method for saving electricity in business processes, characterized in that, Includes the following steps: S1. Obtain the business parameter data of the project to be measured. The business parameter data includes a type identifier used to uniquely represent the project category and a set of operating characteristic parameters used to quantify the project's operating status. S2. Call the pre-set standardized calculation model library. The model library stores multiple calculation algorithm templates corresponding to different types of identifiers, as well as validity threshold rules corresponding to the running feature parameter set. S3. Utilizing adaptive calculation and quality sentinel engine, business parameter data is routed and matched to the target calculation algorithm template based on type identifier. While performing calculation, the validity threshold rule is used to perform real-time verification of the running feature parameter set to generate a comprehensive evaluation signal containing power saving value and data status identifier. S4. Analyze the data status identifier in the comprehensive evaluation signal and generate a control instruction package. If the identifier is compliant, the control instruction package contains an indicator update instruction; if the identifier is abnormal, the control instruction package contains a rectification and blocking instruction. S5. In response to the indicator update command, automatically update the hierarchical annual energy saving indicator data; or in response to the rectification and blocking command, create pending abnormal work orders and lock the business process of the current project.

2. The method for a power-saving business process according to claim 1, characterized in that: The set of operating characteristic parameters in S1 is collected through a structured form interface. The structured form interface dynamically renders input fields based on different project categories. The input fields include energy consumption baseline values, equipment operating power, and operating time parameters.

3. The method for a power-saving business process according to claim 2, characterized in that: The steps for determining the quality of business parameter data are as follows: The system captures user business parameter data through the front-end interactive interface, obtains project category feature identification codes, and retrieves the corresponding dynamic form metadata structure from the configuration set of the non-relational database. The dynamic form metadata structure contains a list of all fields required for the project category, as well as the pre-defined field importance weight for each field. The rendering engine parses the dynamic form metadata structure and draws the corresponding input controls on the user interface. It also binds a data change listener to each input control to capture the data input method in real time. When data is populated into any field, the listener identifies the data source type and sets the field integrity status value to 1 based on the data source confidence coefficient. When each data change event is triggered, the comprehensive reliability score of the calculation interface is updated in real time by using the field importance weight, data source confidence coefficient and field integrity status value of all current fields. If the overall reliability score of the interface is less than the admission reliability threshold, the current data collection quality is deemed insufficient. The submit button is disabled on the front end, and fields with low confidence coefficients of data sources are highlighted to prompt users to supplement data using a more reliable method. If the overall reliability score of the interface is greater than or equal to the admission reliability threshold: the data quality is deemed compliant, the submit button is unlocked, and the data is allowed to proceed to the next stage.

4. The method for saving electricity in a business process according to claim 3, characterized in that: The standardized computational model library in S2 is built using a layered and decoupled architecture, including a basic algorithm layer and a parameter configuration layer. The basic algorithm layer stores general physical computation logic, while the parameter configuration layer stores specific coefficients and validity threshold rules associated with type identifiers.

5. The method for a power-saving business process according to claim 4, characterized in that: The adaptive measurement deviation index is calculated as follows: User-entered set of operational characteristic parameters; real-time collected environmental status parameters; preset standard operating condition basic energy efficiency coefficient, which includes the historical data distribution mean and the historical data distribution standard deviation; statistical characteristic quantities provided by the historical database; The original energy saving is calculated by calling the physical formula in the basic algorithm layer; the environmentally sensitive dynamic correction factor is calculated by using the empirical formula for operating condition correction in thermodynamics and electrical engineering, and the original energy saving is multiplied by the environmentally sensitive dynamic correction factor to obtain the corrected energy saving. The elastic threshold boundary scaling coefficient in the configuration layer is read in parallel. Combining the mean and standard deviation of historical data distribution, the upper and lower bound thresholds of dynamic effectiveness are calculated in real time using the normal distribution interval estimation theory in statistics. Using Z-Score standardization and nonlinear mapping, an adaptive measurement deviation index is calculated based on the corrected post-segment electricity consumption, the mean of historical data distribution, and the standard deviation of historical data distribution. If the adaptive calculation deviation index > deviation blocking threshold or the corrected power consumption > dynamic validity upper limit threshold: the calculation result is determined to be seriously deviating from historical patterns or physical boundaries; strong verification blocking is triggered, an abnormal work order is generated, and the adaptive calculation deviation index value is written into the work order as an abnormality severity indicator. If the adaptive measurement deviation index is less than or equal to the deviation blocking threshold, the calculation result is determined to be within a reasonable fluctuation range. The corrected energy saving is adopted as the final energy saving and stored in the database. At the same time, the historical statistical sample database is updated for the iteration of the historical data distribution mean in the next cycle. The final output is: corrected power consumption and adaptive measurement deviation index.

6. The method for a power-saving business process according to claim 5, characterized in that: The calculation logic executed by the adaptive measurement and quality sentinel engine in S3 is as follows: Through the algorithm routing mechanism, business parameter data is injected into the matching target measurement algorithm template for trial calculation; the deviation of the running feature parameter set from the validity threshold rule is calculated simultaneously. If the deviation of all parameters is within the preset compliance range, a comprehensive evaluation signal containing the effective power saving value and compliance status identifier is output; if the deviation of any parameter exceeds the compliance range, or if there is a logical conflict in the trial calculation result, a comprehensive evaluation signal containing the error code and abnormal status identifier is output.

7. The method for a power-saving business process according to claim 6, characterized in that: The calculation steps for the comprehensive evaluation signal are as follows: Input business parameter data, target calculation algorithm template, and validity threshold rules; Based on the project type, the corresponding target calculation algorithm template is called, the business parameters are substituted into the calculation to obtain the temporary calculation result; Three subtasks are executed in parallel: the physical boundary deviation index of the input parameters relative to the physical boundary is calculated based on the linear normalization method; Calculate the historical statistical deviation index of the provisional calculation results relative to the historical mean; Verify the logical relationships between business parameter data and calculate the logical association consistency coefficient; The comprehensive risk entropy value of the current task is calculated based on the physical boundary deviation index, historical statistical deviation index, and logical correlation consistency coefficient. Compare the comprehensive risk entropy value with the dynamic blocking warning threshold; If the comprehensive risk entropy value is less than or equal to the dynamic blocking warning threshold: the risk is determined to be controllable, and a signal containing the temporary calculation results and compliance indicators is output. If the comprehensive risk entropy value is greater than the dynamic blocking warning threshold: a high-risk anomaly is determined, the circuit breaker is triggered, and an error code containing the specific violation dimension and an anomaly identifier are generated; The final output is a comprehensive evaluation signal.

8. The method for a power-saving business process according to claim 7, characterized in that: The steps for parsing the comprehensive evaluation signal in S4 are as follows: extract the data status identifier field from the comprehensive evaluation signal, use state machine logic to determine the current business flow direction; when the data status identifier is abnormal, extract the error code and map it to specific rectification prompt information, and encapsulate it into the rectification blocking instruction.

9. The method for a power-saving business process according to claim 8, characterized in that: The steps for generating the control instruction packet are as follows: Input the comprehensive evaluation signal; analyze the comprehensive evaluation signal and extract the data status flag bits; If the data status flag is 0, the state machine transitions to the index update state and directly generates an index update instruction; If the data status flag bit = 1, the state machine transitions to the abnormal resolution state and proceeds to the next step; In the abnormal parsing state, the error code is extracted from the comprehensive evaluation signal, the fault dictionary table is queried, and the corresponding natural language rectification prompt and error code severity coefficient are retrieved. The project business weight factor of the current project is calculated in parallel; the rectification urgency index is calculated by using a linear weighted model based on multi-attribute utility theory, combined with the error code severity coefficient and the project business weight factor. Construct a rectification and blocking command object; Encapsulate the following information into an object: natural language rectification prompts; calculated rectification urgency index; and the ID of the locked business document. Before generating control instructions, determine the relationship between the urgency index of rectification and the threshold of graded blocking response; If the urgency index of rectification is greater than or equal to the graded blocking response threshold: it is determined to be an emergency blocking measure; the SLA level is marked as P1 - the highest priority in the control instructions, requiring a response within 4 hours; If the urgency index for rectification is less than the tiered blocking response threshold: it is judged as a general blocking; the SLA level is marked as P3-standard priority in the control instructions, and a response is allowed within 24 hours; The final output is a fully encapsulated control instruction package.

10. A method for saving electricity in a business process according to claim 9, characterized in that: The steps in S5 to respond to the rectification blocking command are as follows: instantiate a new abnormal work order object, write the rectification prompt information into the work order object, and lock the status of the project to be measured as pending rectification until a closed-loop feedback signal for the abnormal work order is received; the steps to respond to the indicator update command are as follows: use the aggregation algorithm to accumulate the energy saving value into the corresponding municipal and provincial annual energy saving indicator completion values, and refresh the visual monitoring dashboard in real time.

Images